diff --git a/cts/cts-fencing.in b/cts/cts-fencing.in index 034304a974..30fb39f298 100644 --- a/cts/cts-fencing.in +++ b/cts/cts-fencing.in @@ -1,954 +1,925 @@ #!@PYTHON@ """Regression tests for Pacemaker's fencer.""" # pylint doesn't like the module name "cts-fencing" which is an invalid complaint for this file # but probably something we want to continue warning about elsewhere # pylint: disable=invalid-name # pacemaker imports need to come after we modify sys.path, which pylint will complain about. # pylint: disable=wrong-import-position __copyright__ = "Copyright 2012-2025 the Pacemaker project contributors" __license__ = "GNU General Public License version 2 or later (GPLv2+) WITHOUT ANY WARRANTY" import argparse import os import sys import subprocess import tempfile # These imports allow running from a source checkout after running `make`. # Note that while this doesn't necessarily mean it will successfully run tests, # but being able to see --help output can be useful. if os.path.exists("@abs_top_srcdir@/python"): sys.path.insert(0, "@abs_top_srcdir@/python") # pylint: disable=comparison-of-constants,comparison-with-itself,condition-evals-to-constant if os.path.exists("@abs_top_builddir@/python") and "@abs_top_builddir@" != "@abs_top_srcdir@": sys.path.insert(0, "@abs_top_builddir@/python") from pacemaker.buildoptions import BuildOptions from pacemaker.exitstatus import ExitStatus from pacemaker._cts.corosync import Corosync, localname from pacemaker._cts.process import killall, exit_if_proc_running from pacemaker._cts.test import Test, Tests TEST_DIR = sys.path[0] def update_path(): """Set the PATH environment variable appropriately for the tests.""" new_path = os.environ['PATH'] if os.path.exists(f"{TEST_DIR}/cts-fencing.in"): # pylint: disable=protected-access print(f"Running tests from the source tree: {BuildOptions._BUILD_DIR} ({TEST_DIR})") # For pacemaker-fenced and cts-fence-helper new_path = f"{BuildOptions._BUILD_DIR}/daemons/fenced:{new_path}" new_path = f"{BuildOptions._BUILD_DIR}/tools:{new_path}" # For stonith_admin new_path = f"{BuildOptions._BUILD_DIR}/cts/support:{new_path}" # For cts-support else: print(f"Running tests from the install tree: {BuildOptions.DAEMON_DIR} (not {TEST_DIR})") # For pacemaker-fenced, cts-fence-helper, and cts-support new_path = f"{BuildOptions.DAEMON_DIR}:{new_path}" print(f'Using PATH="{new_path}"') os.environ['PATH'] = new_path class FenceTest(Test): """Executor for a single test.""" def __init__(self, name, description, **kwargs): """ Create a new FenceTest instance. Arguments: name -- A unique name for this test. This can be used on the command line to specify that only a specific test should be executed. description -- A meaningful description for the test. """ Test.__init__(self, name, description, **kwargs) self._daemon_location = "pacemaker-fenced" def _kill_daemons(self): killall(["pacemakerd", "pacemaker-fenced"]) def _start_daemons(self): cmd = ["pacemaker-fenced", "--stand-alone", "--logfile", self.logpath] if self.verbose: cmd += ["-V"] s = " ".join(cmd) print(f"Starting {s}") # pylint: disable=consider-using-with self._daemon_process = subprocess.Popen(cmd) class FenceTests(Tests): """Collection of all fencing regression tests.""" def __init__(self, **kwargs): """Create a new FenceTests instance.""" Tests.__init__(self, **kwargs) self._corosync = Corosync(self.verbose, self.logdir, "cts-fencing") def new_test(self, name, description): """Create a named test.""" test = FenceTest(name, description, verbose=self.verbose, timeout=self.timeout, force_wait=self.force_wait, logdir=self.logdir) self._tests.append(test) return test def build_api_sanity_tests(self): """Register tests to verify basic API usage.""" verbose_arg = "" if self.verbose: verbose_arg = "-V" test = self.new_test("low_level_api_test", "Sanity-test client API") test.add_cmd("cts-fence-helper", args=f"-t {verbose_arg}", validate=False) test = self.new_test("low_level_api_mainloop_test", "Sanity-test client API using mainloop") test.add_cmd("cts-fence-helper", args=f"-m {verbose_arg}", validate=False) def build_custom_timeout_tests(self): """Register tests to verify custom timeout usage.""" # custom timeout without topology test = self.new_test("custom_timeout_1", "Verify per device timeouts work as expected without using topology") test.add_cmd('stonith_admin', args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd('stonith_admin', args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3 -o pcmk_off_timeout=1') test.add_cmd('stonith_admin', args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3 -o pcmk_off_timeout=4') test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") # timeout is 5+1+4 = 10 test.add_log_pattern("Total timeout set to 12s") # custom timeout _WITH_ topology test = self.new_test("custom_timeout_2", "Verify per device timeouts work as expected _WITH_ topology") test.add_cmd('stonith_admin', args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd('stonith_admin', args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3 -o pcmk_off_timeout=1000ms') test.add_cmd('stonith_admin', args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3 -o pcmk_off_timeout=4000s') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") # timeout is 5+1+4000 = 4006 test.add_log_pattern("Total timeout set to 4807s") def build_fence_merge_tests(self): """Register tests to verify when fence operations should be merged.""" # Simple test that overlapping fencing operations get merged test = self.new_test("custom_merge_single", "Verify overlapping identical fencing operations are merged, no fencing levels used") test.add_cmd("stonith_admin", args="--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10") # one merger will happen test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") # the pattern below signifies that both the original and duplicate operation completed test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") # Test that multiple mergers occur test = self.new_test("custom_merge_multiple", "Verify multiple overlapping identical fencing operations are merged") test.add_cmd("stonith_admin", args="--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o \"mode=pass\" -o delay=2 -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10") # 4 mergers should occur test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") # the pattern below signifies that both the original and duplicate operation completed test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") # Test that multiple mergers occur with topologies used test = self.new_test("custom_merge_with_topology", "Verify multiple overlapping identical fencing operations are merged with fencing levels") test.add_cmd("stonith_admin", args="--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10") # 4 mergers should occur test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client") # the pattern below signifies that both the original and duplicate operation completed test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") test.add_log_pattern("Operation 'off' targeting node3 by ") def build_fence_no_merge_tests(self): """Register tests to verify when fence operations should not be merged.""" test = self.new_test("custom_no_merge", "Verify differing fencing operations are not merged") test.add_cmd("stonith_admin", args="--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node3 node2") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3 node2") test.add_cmd("stonith_admin", args="--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node3 node2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -F node2 -t 10", no_wait=True) test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 10") test.add_log_pattern("Merging fencing action 'off' targeting node3 originating from client", negative=True) def build_standalone_tests(self): """Register a grab bag of tests.""" # test what happens when all devices timeout test = self.new_test("fence_multi_device_failure", "Verify that all devices timeout, a fencing failure is returned") test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false3 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 2", expected_exitcode=ExitStatus.TIMEOUT) test.add_log_pattern("Total timeout set to 7s") test.add_log_pattern("targeting node3 using false1 returned ") test.add_log_pattern("targeting node3 using false2 returned ") test.add_log_pattern("targeting node3 using false3 returned ") # test what happens when multiple devices can fence a node, but the first device fails test = self.new_test("fence_device_failure_rollover", "Verify that when one fence device fails for a node, the others are tried") test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") test.add_log_pattern("Total timeout set to 18s") # test what happens when we try to use a missing fence-agent test = self.new_test("fence_missing_agent", "Verify proper error-handling when using a non-existent fence-agent") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_missing -o mode=pass -o pcmk_host_list=node3") test.add_cmd("stonith_admin", args="--output-as=xml -R true2 -a fence_dummy -o mode=pass -o pcmk_host_list=node2") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5", expected_exitcode=ExitStatus.NOSUCH) test.add_cmd("stonith_admin", args="--output-as=xml -F node2 -t 5") # simple topology test for one device test = self.new_test("topology_simple", "Verify all fencing devices at a level are used") test.add_cmd("stonith_admin", args='--output-as=xml -R true -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") test.add_log_pattern("Total timeout set to 6s") test.add_log_pattern("targeting node3 using true returned 0") # add topology, delete topology, verify fencing still works test = self.new_test("topology_add_remove", "Verify fencing occurrs after all topology levels are removed") test.add_cmd("stonith_admin", args='--output-as=xml -R true -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true") test.add_cmd("stonith_admin", args="--output-as=xml -d node3 -i 1") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") test.add_log_pattern("Total timeout set to 6s") test.add_log_pattern("targeting node3 using true returned 0") # test what happens when the first fencing level has multiple devices test = self.new_test("topology_device_fails", "Verify if one device in a level fails, the other is tried") test.add_cmd("stonith_admin", args='--output-as=xml -R false -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 20") test.add_log_pattern("Total timeout set to 48s") test.add_log_pattern("targeting node3 using false returned 1") test.add_log_pattern("targeting node3 using true returned 0") # test what happens when the first fencing level fails test = self.new_test("topology_multi_level_fails", "Verify if one level fails, the next leve is tried") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true4 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true3") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true4") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 3") test.add_log_pattern("Total timeout set to 21s") test.add_log_pattern("targeting node3 using false1 returned 1") test.add_log_pattern("targeting node3 using false2 returned 1") test.add_log_pattern("targeting node3 using true3 returned 0") test.add_log_pattern("targeting node3 using true4 returned 0") # test what happens when the first fencing level had devices that no one has registered test = self.new_test("topology_missing_devices", "Verify topology can continue with missing devices") test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true4 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true3") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true4") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") # Test what happens if multiple fencing levels are defined, and then the first one is removed test = self.new_test("topology_level_removal", "Verify level removal works") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true4 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v false2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true3") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 3 -v true4") # Now remove level 2, verify none of the devices in level two are hit test.add_cmd("stonith_admin", args="--output-as=xml -d node3 -i 2") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 20") test.add_log_pattern("Total timeout set to 96s") test.add_log_pattern("targeting node3 using false1 returned 1") test.add_log_pattern("targeting node3 using false2 returned ", negative=True) test.add_log_pattern("targeting node3 using true3 returned 0") test.add_log_pattern("targeting node3 using true4 returned 0") # Test targeting a topology level by node name pattern test = self.new_test("topology_level_pattern", "Verify targeting topology by node name pattern works") test.add_cmd("stonith_admin", args='--output-as=xml -R true -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r '@node.*' -i 1 -v true") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5") test.add_log_pattern("targeting node3 using true returned 0") # test allowing commas and semicolons as delimiters in pcmk_host_list test = self.new_test("host_list_delimiters", "Verify commas and semicolons can be used as pcmk_host_list delimiters") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1,node2,node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=pcmk1;pcmk2;pcmk3"') test.add_cmd("stonith_admin", args="stonith_admin --output-as=xml -F node2 -t 5") test.add_cmd("stonith_admin", args="stonith_admin --output-as=xml -F pcmk3 -t 5") test.add_log_pattern("targeting node2 using true1 returned 0") test.add_log_pattern("targeting pcmk3 using true2 returned 0") # test the stonith builds the correct list of devices that can fence a node test = self.new_test("list_devices", "Verify list of devices that can fence a node is correct") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -l node1 -V", stdout_match="true2", stdout_no_match="true1") test.add_cmd("stonith_admin", args="--output-as=xml -l node1 -V", stdout_match="true3", stdout_no_match="true1") # simple test of device monitor test = self.new_test("monitor", "Verify device is reachable") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node3"') test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node3"') test.add_cmd("stonith_admin", args="--output-as=xml -Q true1") test.add_cmd("stonith_admin", args="--output-as=xml -Q false1") test.add_cmd("stonith_admin", args="--output-as=xml -Q true2", expected_exitcode=ExitStatus.NOSUCH) # Verify monitor occurs for duration of timeout period on failure test = self.new_test("monitor_timeout", "Verify monitor uses duration of timeout period given") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=fail -o monitor_mode=fail -o pcmk_host_list=node3') test.add_cmd("stonith_admin", args="--output-as=xml -Q true1 -t 5", expected_exitcode=ExitStatus.ERROR) test.add_log_pattern("Attempt 2 to execute") # Verify monitor occurs for duration of timeout period on failure, but stops at max retries test = self.new_test("monitor_timeout_max_retries", "Verify monitor retries until max retry value or timeout is hit") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=fail -o monitor_mode=fail -o pcmk_host_list=node3') test.add_cmd("stonith_admin", args="--output-as=xml -Q true1 -t 15", expected_exitcode=ExitStatus.ERROR) test.add_log_pattern("Attempted to execute agent fence_dummy (list) the maximum number of times") # simple register test test = self.new_test("register", "Verify devices can be registered and un-registered") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3') test.add_cmd("stonith_admin", args="--output-as=xml -Q true1") test.add_cmd("stonith_admin", args="--output-as=xml -D true1") test.add_cmd("stonith_admin", args="--output-as=xml -Q true1", expected_exitcode=ExitStatus.NOSUCH) # simple reboot test test = self.new_test("reboot", "Verify devices can be rebooted") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3') test.add_cmd("stonith_admin", args="--output-as=xml -B node3 -t 5") test.add_cmd("stonith_admin", args="--output-as=xml -D true1") test.add_cmd("stonith_admin", args="--output-as=xml -Q true1", expected_exitcode=ExitStatus.NOSUCH) # test fencing history test = self.new_test("fence_history", "Verify last fencing operation is returned") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node3') test.add_cmd("stonith_admin", args="--output-as=xml -F node3 -t 5 -V") test.add_cmd("stonith_admin", args="--output-as=xml -H node3", stdout_match='action="off" target="node3" .* status="success"') # simple test of dynamic list query test = self.new_test("dynamic_list_query", "Verify dynamic list of fencing devices can be retrieved") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -R true2 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -R true3 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -l fake_port_1", stdout_match='count="3"') # fence using dynamic list query test = self.new_test("fence_dynamic_list_query", "Verify dynamic list of fencing devices can be retrieved") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -R true2 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -R true3 -a fence_dummy -o mode=pass -o mock_dynamic_hosts=fake_port_1") test.add_cmd("stonith_admin", args="--output-as=xml -F fake_port_1 -t 5 -V") # simple test of query using status action test = self.new_test("status_query", "Verify dynamic list of fencing devices can be retrieved") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_check=status') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o pcmk_host_check=status') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o pcmk_host_check=status') test.add_cmd("stonith_admin", args="--output-as=xml -l fake_port_1", stdout_match='count="3"') # test what happens when no reboot action is advertised test = self.new_test("no_reboot_support", "Verify reboot action defaults to off when no reboot action is advertised by agent") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy_no_reboot -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -B node1 -t 5 -V") test.add_log_pattern("does not support reboot") test.add_log_pattern("using true1 returned 0") # make sure reboot is used when reboot action is advertised test = self.new_test("with_reboot_support", "Verify reboot action can be used when metadata advertises it") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -B node1 -t 5 -V") test.add_log_pattern("does not advertise support for 'reboot', performing 'off'", negative=True) test.add_log_pattern("using true1 returned 0") # make sure all fencing delays are applied correctly and taken into account by fencing timeouts with topology test = self.new_test("topology_delays", "Verify all fencing delays are applied correctly and taken into account by fencing timeouts with topology") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3" -o pcmk_delay_base=1') test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list=node1 node2 node3" -o pcmk_delay_base=1') # Resulting "random" delay will always be 1 since (rand() % (delay_max - delay_base)) is always 0 here test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3" -o pcmk_delay_base=1 -o pcmk_delay_max=2') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o "pcmk_host_list=node1 node2 node3"') test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -r node3 -i 2 -v true3") test.add_cmd("stonith_admin", args="--output-as=xml -F node3 --delay 1") # Total fencing timeout takes all fencing delays into account test.add_log_pattern("Total timeout set to 582s") # Fencing timeout for the first device takes the requested fencing delay # and pcmk_delay_base into account test.add_log_pattern(r"Requesting that .* perform 'off' action targeting node3 using true1 .*146s.*", regex=True) # Requested fencing delay is applied only for the first device in the # first level, with the static delay from pcmk_delay_base added test.add_log_pattern("Delaying 'off' action targeting node3 using true1 for 2s | timeout=120s requested_delay=1s base=1s max=1s") # Fencing timeout no longer takes the requested fencing delay into account for further devices test.add_log_pattern(r"Requesting that .* perform 'off' action targeting node3 using false1 .*145s.*", regex=True) # Requested fencing delay is no longer applied for further devices test.add_log_pattern("Delaying 'off' action targeting node3 using false1 for 1s | timeout=120s requested_delay=0s base=1s max=1s") # Fencing timeout takes pcmk_delay_max into account test.add_log_pattern(r"Requesting that .* perform 'off' action targeting node3 using true2 .*146s.*", regex=True) test.add_log_pattern("Delaying 'off' action targeting node3 using true2 for 1s | timeout=120s requested_delay=0s base=1s max=2s") test.add_log_pattern("Delaying 'off' action targeting node3 using true3", negative=True) - def build_nodeid_tests(self): - """Register tests that use a corosync node id.""" - our_uname = localname() - - # verify nodeid is supplied when nodeid is in the metadata parameters - test = self.new_test("supply_nodeid", - "Verify nodeid is given when fence agent has nodeid as parameter") - - test.add_cmd("stonith_admin", - args=f'--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname}"') - test.add_cmd("stonith_admin", args=f"--output-as=xml -F {our_uname} -t 3") - test.add_log_pattern(f"as nodeid with fence action 'off' targeting {our_uname}") - - # verify nodeid is _NOT_ supplied when nodeid is not in the metadata parameters - test = self.new_test("do_not_supply_nodeid", - "Verify nodeid is _NOT_ given when fence agent does not have nodeid as parameter") - - # use a host name that won't be in corosync.conf - test.add_cmd("stonith_admin", - args='--output-as=xml -R true1 -a fence_dummy_no_nodeid ' - f'-o mode=pass -o pcmk_host_list="regr-test {our_uname}"') - test.add_cmd("stonith_admin", args="--output-as=xml -F regr-test -t 3") - test.add_log_pattern("as nodeid with fence action 'off' targeting regr-test", - negative=True) - test.add_cmd("stonith_admin", args=f"--output-as=xml -F {our_uname} -t 3") - test.add_log_pattern("as nodeid with fence action 'off' targeting {our_uname}", - negative=True) - def build_unfence_tests(self): """Register tests that verify unfencing.""" our_uname = localname() # verify unfencing using automatic unfencing test = self.new_test("unfence_required_1", "Verify require unfencing on all devices when automatic=true in agent's metadata") test.add_cmd('stonith_admin', args=f'--output-as=xml -R true1 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname}"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true2 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname}"') test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 3") # both devices should be executed test.add_log_pattern("using true1 returned 0") test.add_log_pattern("using true2 returned 0") # verify unfencing using automatic unfencing fails if any of the required agents fail test = self.new_test("unfence_required_2", "Verify require unfencing on all devices when automatic=true in agent's metadata") test.add_cmd('stonith_admin', args=f'--output-as=xml -R true1 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname}"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true2 -a fence_dummy_auto_unfence -o mode=fail -o "pcmk_host_list={our_uname}"') test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 6", expected_exitcode=ExitStatus.ERROR) # verify unfencing using automatic devices with topology test = self.new_test("unfence_required_3", "Verify require unfencing on all devices even when at different topology levels") test.add_cmd('stonith_admin', args=f'--output-as=xml -R true1 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true2 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 1 -v true1") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v true2") test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 3") test.add_log_pattern("using true1 returned 0") test.add_log_pattern("using true2 returned 0") # verify unfencing using automatic devices with topology test = self.new_test("unfence_required_4", "Verify all required devices are executed even with topology levels fail") test.add_cmd('stonith_admin', args=f'--output-as=xml -R true1 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true2 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true3 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R true4 -a fence_dummy_auto_unfence -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R false1 -a fence_dummy -o mode=fail -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R false2 -a fence_dummy -o mode=fail -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R false3 -a fence_dummy -o mode=fail -o "pcmk_host_list={our_uname} node3"') test.add_cmd('stonith_admin', args=f'--output-as=xml -R false4 -a fence_dummy -o mode=fail -o "pcmk_host_list={our_uname} node3"') test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 1 -v true1") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 1 -v false1") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v false2") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v true2") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v false3") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v true3") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 3 -v false4") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 4 -v true4") test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 3") test.add_log_pattern("using true1 returned 0") test.add_log_pattern("using true2 returned 0") test.add_log_pattern("using true3 returned 0") test.add_log_pattern("using true4 returned 0") def build_unfence_on_target_tests(self): """Register tests that verify unfencing that runs on the target.""" our_uname = localname() # verify unfencing using on_target device test = self.new_test("unfence_on_target_1", "Verify unfencing with on_target = true") test.add_cmd("stonith_admin", args=f'--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname}"') test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 3") test.add_log_pattern("(on) to be executed on target") # verify failure of unfencing using on_target device test = self.new_test("unfence_on_target_2", "Verify failure unfencing with on_target = true") test.add_cmd("stonith_admin", args=f'--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname} node_fake_1234"') test.add_cmd("stonith_admin", args="--output-as=xml -U node_fake_1234 -t 3", expected_exitcode=ExitStatus.NOSUCH) test.add_log_pattern("(on) to be executed on target") # verify unfencing using on_target device with topology test = self.new_test("unfence_on_target_3", "Verify unfencing with on_target = true using topology") test.add_cmd("stonith_admin", args=f'--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd("stonith_admin", args=f'--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname} node3"') test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 1 -v true1") test.add_cmd("stonith_admin", args=f"--output-as=xml -r {our_uname} -i 2 -v true2") test.add_cmd("stonith_admin", args=f"--output-as=xml -U {our_uname} -t 3") test.add_log_pattern("(on) to be executed on target") # verify unfencing using on_target device with topology fails when target node doesn't exist test = self.new_test("unfence_on_target_4", "Verify unfencing failure with on_target = true using topology") test.add_cmd("stonith_admin", args=f'--output-as=xml -R true1 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname} node_fake"') test.add_cmd("stonith_admin", args=f'--output-as=xml -R true2 -a fence_dummy -o mode=pass -o "pcmk_host_list={our_uname} node_fake"') test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v true1") test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 2 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -U node_fake -t 3", expected_exitcode=ExitStatus.NOSUCH) test.add_log_pattern("(on) to be executed on target") def build_remap_tests(self): """Register tests that verify remapping of reboots to off-on.""" test = self.new_test("remap_simple", "Verify sequential topology reboot is remapped to all-off-then-all-on") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake ' '-o pcmk_off_timeout=1 -o pcmk_reboot_timeout=10') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake ' '-o pcmk_off_timeout=2 -o pcmk_reboot_timeout=20') test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v true1 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -B node_fake -t 5") test.add_log_pattern("Remapping multiple-device reboot targeting node_fake") # timeout should be sum of off timeouts (1+2=3), not reboot timeouts (10+20=30) test.add_log_pattern("Total timeout set to 3s for peer's fencing targeting node_fake") test.add_log_pattern("perform 'off' action targeting node_fake using true1") test.add_log_pattern("perform 'off' action targeting node_fake using true2") test.add_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'") # fence_dummy sets "on" as an on_target action test.add_log_pattern("Ignoring true1 'on' failure (no capable peers) targeting node_fake") test.add_log_pattern("Ignoring true2 'on' failure (no capable peers) targeting node_fake") test.add_log_pattern("Undoing remap of reboot targeting node_fake") test = self.new_test("remap_simple_off", "Verify sequential topology reboot skips 'on' if " "pcmk_reboot_action=off or agent doesn't support " "'on'") test.add_cmd("stonith_admin", args="--output-as=xml -R true1 -a fence_dummy -o mode=pass " "-o pcmk_host_list=node_fake -o pcmk_off_timeout=1 " "-o pcmk_reboot_timeout=10 -o pcmk_reboot_action=off") test.add_cmd("stonith_admin", args="--output-as=xml -R true2 -a fence_dummy_no_on " "-o mode=pass -o pcmk_host_list=node_fake " "-o pcmk_off_timeout=2 -o pcmk_reboot_timeout=20") test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v true1 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -B node_fake -t 5") test.add_log_pattern("Remapping multiple-device reboot targeting node_fake") # timeout should be sum of off timeouts (1+2=3), not reboot timeouts (10+20=30) test.add_log_pattern("Total timeout set to 3s for peer's fencing targeting node_fake") test.add_log_pattern("perform 'off' action targeting node_fake using true1") test.add_log_pattern("perform 'off' action targeting node_fake using true2") test.add_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'") # "on" should be skipped test.add_log_pattern("Not turning node_fake back on using " "true1 because the device is configured " "to stay off") test.add_log_pattern("Not turning node_fake back on using true2" " because the agent doesn't support 'on'") test.add_log_pattern("Undoing remap of reboot targeting node_fake") test = self.new_test("remap_automatic", "Verify remapped topology reboot skips automatic 'on'") test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy_auto_unfence ' '-o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy_auto_unfence ' '-o "mode=pass" -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v true1 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -B node_fake -t 5") test.add_log_pattern("Remapping multiple-device reboot targeting node_fake") test.add_log_pattern("perform 'off' action targeting node_fake using true1") test.add_log_pattern("perform 'off' action targeting node_fake using true2") test.add_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'") test.add_log_pattern("Undoing remap of reboot targeting node_fake") test.add_log_pattern("perform 'on' action targeting node_fake using", negative=True) test.add_log_pattern("'on' failure", negative=True) test = self.new_test("remap_complex_1", "Verify remapped topology reboot in second level works if non-remapped first level fails") test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 2 -v true1 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -B node_fake -t 5") test.add_log_pattern("perform 'reboot' action targeting node_fake using false1") test.add_log_pattern("Remapping multiple-device reboot targeting node_fake") test.add_log_pattern("perform 'off' action targeting node_fake using true1") test.add_log_pattern("perform 'off' action targeting node_fake using true2") test.add_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'") test.add_log_pattern("Ignoring true1 'on' failure (no capable peers) targeting node_fake") test.add_log_pattern("Ignoring true2 'on' failure (no capable peers) targeting node_fake") test.add_log_pattern("Undoing remap of reboot targeting node_fake") test = self.new_test("remap_complex_2", "Verify remapped topology reboot failure in second level proceeds to third level") test.add_cmd("stonith_admin", args='--output-as=xml -R false1 -a fence_dummy -o mode=fail -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R false2 -a fence_dummy -o mode=fail -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true1 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true2 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args='--output-as=xml -R true3 -a fence_dummy -o mode=pass -o pcmk_host_list=node_fake') test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 1 -v false1") test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 2 -v true1 -v false2 -v true3") test.add_cmd("stonith_admin", args="--output-as=xml -r node_fake -i 3 -v true2") test.add_cmd("stonith_admin", args="--output-as=xml -B node_fake -t 5") test.add_log_pattern("perform 'reboot' action targeting node_fake using false1") test.add_log_pattern("Remapping multiple-device reboot targeting node_fake") test.add_log_pattern("perform 'off' action targeting node_fake using true1") test.add_log_pattern("perform 'off' action targeting node_fake using false2") test.add_log_pattern("Attempted to execute agent fence_dummy (off) the maximum number of times") test.add_log_pattern("Undoing remap of reboot targeting node_fake") test.add_log_pattern("perform 'reboot' action targeting node_fake using true2") test.add_log_pattern("node_fake with true3", negative=True) def build_query_tests(self): """Run stonith_admin --metadata for the fence_dummy agent and check command output.""" test = self.new_test("get_metadata", "Run stonith_admin --metadata for the fence_dummy agent") test.add_cmd("stonith_admin", args="--output-as=xml -a fence_dummy --metadata", stdout_match=' %s %s """ % (agent(), SHORT_DESC, AGENT_VERSION, OCF_VERSION, LONG_DESC)) for (option, _) in sorted_options(avail_opt): if "shortdesc" not in ALL_OPT[option]: continue print(' ' % (option, ALL_OPT[option]["required"])) default = "" default_name_arg = "-%s" % ALL_OPT[option]["getopt"][:-1] default_name_no_arg = "-%s" % ALL_OPT[option]["getopt"] if "default" in ALL_OPT[option]: default = 'default="%s"' % ALL_OPT[option]["default"] elif options.get(default_name_arg) is not None: try: default = 'default="%s"' % options[default_name_arg] except TypeError: # @todo/@note: Currently there is no clean way how to handle lists # we can create a string from it but we can't set it on command line default = 'default="%s"' % str(options[default_name_arg]) elif default_name_no_arg in options: default = 'default="true"' mixed = ALL_OPT[option]["help"] # split it between option and help text res = re.compile(r"^(.*--\S+)\s+", re.IGNORECASE | re.S).search(mixed) if res is not None: mixed = res.group(1) mixed = mixed.replace("<", "<").replace(">", ">") print(' ' % mixed) if ALL_OPT[option]["getopt"].count(":") > 0: print(' ' % default) else: print(' ' % default) print(' %s' % ALL_OPT[option]["shortdesc"]) print(' ') print(' \n ') if not no_on: if auto_unfence: attr_name = 'automatic' else: attr_name = 'on_target' print(' ' % attr_name) print(' ') if not no_reboot: print(' ') print(' ') print(' ') print(' ') print(' ') print(' ') print('') def option_longopt(option): """Return the getopt-compatible long-option name of the given option.""" if ALL_OPT[option]["getopt"].endswith(":"): return ALL_OPT[option]["longopt"] + "=" return ALL_OPT[option]["longopt"] def opts_from_command_line(argv, avail_opt): """Read options from command-line arguments.""" # Prepare list of options for getopt getopt_string = "" longopt_list = [] for k in avail_opt: if k in ALL_OPT: getopt_string += ALL_OPT[k]["getopt"] else: fail_usage("Parse error: unknown option '%s'" % k) if k in ALL_OPT and "longopt" in ALL_OPT[k]: longopt_list.append(option_longopt(k)) try: (opt, _) = getopt.gnu_getopt(argv, getopt_string, longopt_list) except getopt.GetoptError as error: fail_usage("Parse error: %s" % error.msg) # Transform longopt to short one which are used in fencing agents old_opt = opt opt = {} for old_option in dict(old_opt): if old_option.startswith("--"): for rec in ALL_OPT.values(): if rec.get("longopt") is None: continue long = "--%s" % rec["longopt"] if long == old_option: short = "-%s" % rec["getopt"][0] opt[short] = dict(old_opt)[old_option] else: opt[old_option] = dict(old_opt)[old_option] # Compatibility Layer (with what? probably not needed for fence_dummy) new_opt = dict(opt) if "-T" in new_opt: new_opt["-o"] = "status" if "-n" in new_opt: new_opt["-m"] = new_opt["-n"] opt = new_opt return opt def opts_from_stdin(avail_opt): """Read options from standard input.""" opt = {} name = "" for line in sys.stdin.readlines(): line = line.strip() if line.startswith("#") or (len(line) == 0): continue (name, value) = (line + "=").split("=", 1) value = value[:-1] # Compatibility Layer (with what? probably not needed for fence_dummy) if name == "option": name = "action" if name not in avail_opt: print("Parse error: Ignoring unknown option '%s'" % line, file=sys.stderr) continue if ALL_OPT[name]["getopt"].endswith(":"): short = "-%s" % ALL_OPT[name]["getopt"][0] opt[short] = value elif value.lower() in ["1", "yes", "on", "true"]: short = "-%s" % ALL_OPT[name]["getopt"] opt[short] = "1" return opt def process_input(avail_opt): """Set standard environment variables, and parse all options.""" # Set standard environment os.putenv("LANG", "C") os.putenv("LC_ALL", "C") # Read options from command line or standard input if len(sys.argv) > 1: return opts_from_command_line(sys.argv[1:], avail_opt) return opts_from_stdin(avail_opt) def atexit_handler(): """Close stdout on exit.""" try: sys.stdout.close() os.close(1) except IOError: sys.exit("%s failed to close standard output" % agent()) def success_mode(options, option, default_value): """Return exit code specified by option.""" if option in options: test_value = options[option] else: test_value = default_value if test_value == "pass": exitcode = 0 elif test_value == "fail": exitcode = 1 else: exitcode = random.randint(0, 1) return exitcode def write_options(options): """Write out all options to debug file.""" with contextlib.suppress(IOError): with io.open(options["-D"], "at", encoding="utf-8") as debugfile: debugfile.write("### %s ###\n" % time.strftime("%Y-%m-%d %H:%M:%S")) for option in sorted(options): debugfile.write("%s=%s\n" % (option, options[option])) debugfile.write("###\n") def main(): """Run the dummy fencing agent.""" auto_unfence = False no_reboot = False no_on = False # Meta-data can't take parameters, so we simulate different meta-data # behavior based on the executable name (which can be a symbolic link). if sys.argv[0].endswith("_auto_unfence"): auto_unfence = True elif sys.argv[0].endswith("_no_reboot"): no_reboot = True elif sys.argv[0].endswith("_no_on"): no_on = True - elif sys.argv[0].endswith("_no_nodeid"): - del ALL_OPT["nodeid"] device_opt = ALL_OPT.keys() # Defaults for fence agent atexit.register(atexit_handler) options = process_input(device_opt) options["device_opt"] = device_opt show_docs(options, auto_unfence, no_reboot, no_on) action = options.get("-o", "reboot") # dump input to file if "-D" in options and action != "validate-all": write_options(options) if "-f" in options and action != "validate-all": val = int(options["-f"]) print("delay sleep for %d seconds" % val, file=sys.stderr) time.sleep(val) # random sleep for testing if "-R" in options and action != "validate-all": val = int(options["-R"]) ran = random.randint(1, val) print("random sleep for %d seconds" % ran, file=sys.stderr) time.sleep(ran) if action == "monitor": if "-d" in options: time.sleep(int(options["-d"])) exitcode = success_mode(options, "-m", "pass") elif action == "list": print("fence_dummy action (list) called", file=sys.stderr) if "-H" in options: print(options["-H"]) exitcode = 0 else: print("dynamic hostlist requires mock_dynamic_hosts to be set", file=sys.stderr) exitcode = 1 elif action == "validate-all": if "-f" in options: val = int(options["-f"]) if val > 10: exitcode = 1 else: exitcode = 0 else: exitcode = 1 elif action == "off": if "-F" in options: time.sleep(int(options["-F"])) exitcode = success_mode(options, "-M", "random") else: exitcode = success_mode(options, "-M", "random") # Ensure we generate some error output on failure exit. if exitcode == 1: print("simulated %s failure" % action, file=sys.stderr) sys.exit(exitcode) if __name__ == "__main__": main() diff --git a/daemons/fenced/fenced_cib.c b/daemons/fenced/fenced_cib.c index 887486d04e..90c225569e 100644 --- a/daemons/fenced/fenced_cib.c +++ b/daemons/fenced/fenced_cib.c @@ -1,621 +1,658 @@ /* * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include // xmlNode #include // xmlXPathObject, etc. #include #include #include #include #include #include static xmlNode *local_cib = NULL; static cib_t *cib_api = NULL; static bool have_cib_devices = FALSE; /*! * \internal * \brief Check whether a node has a specific attribute name/value * * \param[in] node Name of node to check * \param[in] name Name of an attribute to look for * \param[in] value The value the named attribute needs to be set to in order to be considered a match * * \return TRUE if the locally cached CIB has the specified node attribute */ gboolean node_has_attr(const char *node, const char *name, const char *value) { GString *xpath = NULL; xmlNode *match; CRM_CHECK((local_cib != NULL) && (node != NULL) && (name != NULL) && (value != NULL), return FALSE); /* Search for the node's attributes in the CIB. While the schema allows * multiple sets of instance attributes, and allows instance attributes to * use id-ref to reference values elsewhere, that is intended for resources, * so we ignore that here. */ xpath = g_string_sized_new(256); pcmk__g_strcat(xpath, "//" PCMK_XE_NODES "/" PCMK_XE_NODE "[@" PCMK_XA_UNAME "='", node, "']" "/" PCMK_XE_INSTANCE_ATTRIBUTES "/" PCMK_XE_NVPAIR "[@" PCMK_XA_NAME "='", name, "' " "and @" PCMK_XA_VALUE "='", value, "']", NULL); match = pcmk__xpath_find_one(local_cib->doc, xpath->str, LOG_NEVER); g_string_free(xpath, TRUE); return (match != NULL); } static void add_topology_level(xmlNode *match) { char *desc = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(match != NULL, return); fenced_register_level(match, &desc, &result); fenced_send_config_notification(STONITH_OP_LEVEL_ADD, &result, desc); pcmk__reset_result(&result); free(desc); } static void topology_remove_helper(const char *node, int level) { char *desc = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; xmlNode *data = pcmk__xe_create(NULL, PCMK_XE_FENCING_LEVEL); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add_int(data, PCMK_XA_INDEX, level); crm_xml_add(data, PCMK_XA_TARGET, node); fenced_unregister_level(data, &desc, &result); fenced_send_config_notification(STONITH_OP_LEVEL_DEL, &result, desc); pcmk__reset_result(&result); pcmk__xml_free(data); free(desc); } static void remove_topology_level(xmlNode *match) { int index = 0; char *key = NULL; CRM_CHECK(match != NULL, return); key = stonith_level_key(match, fenced_target_by_unknown); crm_element_value_int(match, PCMK_XA_INDEX, &index); topology_remove_helper(key, index); free(key); } static void register_fencing_topology(xmlXPathObjectPtr xpathObj) { int max = pcmk__xpath_num_results(xpathObj); for (int lpc = 0; lpc < max; lpc++) { xmlNode *match = pcmk__xpath_result(xpathObj, lpc); if (match == NULL) { continue; } remove_topology_level(match); add_topology_level(match); } } /* Fencing */ void fencing_topology_init(void) { xmlXPathObject *xpathObj = NULL; const char *xpath = "//" PCMK_XE_FENCING_LEVEL; crm_trace("Full topology refresh"); free_topology_list(); init_topology_list(); /* Grab everything */ xpathObj = pcmk__xpath_search(local_cib->doc, xpath); register_fencing_topology(xpathObj); xmlXPathFreeObject(xpathObj); } #define XPATH_WATCHDOG_TIMEOUT "//" PCMK_XE_NVPAIR \ "[@" PCMK_XA_NAME "='" \ PCMK_OPT_STONITH_WATCHDOG_TIMEOUT "']" static void update_stonith_watchdog_timeout_ms(xmlNode *cib) { long long timeout_ms = 0; xmlNode *stonith_watchdog_xml = NULL; const char *value = NULL; // @TODO An XPath search can't handle multiple instances or rules stonith_watchdog_xml = pcmk__xpath_find_one(cib->doc, XPATH_WATCHDOG_TIMEOUT, LOG_NEVER); if (stonith_watchdog_xml) { value = crm_element_value(stonith_watchdog_xml, PCMK_XA_VALUE); } if (value) { timeout_ms = crm_get_msec(value); } if (timeout_ms < 0) { timeout_ms = pcmk__auto_stonith_watchdog_timeout(); } stonith_watchdog_timeout_ms = timeout_ms; } +/*! + * \internal + * \brief Mark a fence device dirty if its \c cib_registered flag is \c TRUE + * + * \param[in] key Ignored + * \param[in,out] value Fence device (fenced_device_t *) + * \param[in] user_data Ignored + * + * \note This function is suitable for use with \c g_hash_table_foreach(). + */ +static void +mark_dirty_if_cib_registered(gpointer key, gpointer value, gpointer user_data) +{ + fenced_device_t *device = value; + + if (device->cib_registered) { + device->dirty = TRUE; + } +} + +/*! + * \internal + * \brief Return the value of a fence device's \c dirty flag + * + * \param[in] key Ignored + * \param[in] value Fence device (fenced_device_t *) + * \param[in] user_data Ignored + * + * \return \c dirty flag of \p value + * + * \note This function is suitable for use with + * \c g_hash_table_foreach_remove(). + */ +static gboolean +device_is_dirty(gpointer key, gpointer value, gpointer user_data) +{ + fenced_device_t *device = value; + + return device->dirty; +} + /*! * \internal * \brief Update all STONITH device definitions based on current CIB */ static void cib_devices_update(void) { - GHashTableIter iter; - stonith_device_t *device = NULL; - crm_info("Updating devices to version %s.%s.%s", crm_element_value(local_cib, PCMK_XA_ADMIN_EPOCH), crm_element_value(local_cib, PCMK_XA_EPOCH), crm_element_value(local_cib, PCMK_XA_NUM_UPDATES)); - g_hash_table_iter_init(&iter, device_list); - while (g_hash_table_iter_next(&iter, NULL, (void **)&device)) { - if (device->cib_registered) { - device->dirty = TRUE; - } - } + fenced_foreach_device(mark_dirty_if_cib_registered, NULL); /* have list repopulated if cib has a watchdog-fencing-resource TODO: keep a cached list for queries happening while we are refreshing */ g_list_free_full(stonith_watchdog_targets, free); stonith_watchdog_targets = NULL; fenced_scheduler_run(local_cib); - g_hash_table_iter_init(&iter, device_list); - while (g_hash_table_iter_next(&iter, NULL, (void **)&device)) { - if (device->dirty) { - g_hash_table_iter_remove(&iter); - } - } + fenced_foreach_device_remove(device_is_dirty); } +#define PRIMITIVE_ID_XP_FRAGMENT "/" PCMK_XE_PRIMITIVE "[@" PCMK_XA_ID "='" + static void -update_cib_stonith_devices(const char *event, xmlNode * msg) +update_cib_stonith_devices(const xmlNode *patchset) { - int format = 1; - xmlNode *wrapper = pcmk__xe_first_child(msg, PCMK__XE_CIB_UPDATE_RESULT, - NULL, NULL); - xmlNode *patchset = pcmk__xe_first_child(wrapper, NULL, NULL, NULL); char *reason = NULL; - CRM_CHECK(patchset != NULL, return); - crm_element_value_int(patchset, PCMK_XA_FORMAT, &format); - - if (format != 2) { - crm_warn("Unknown patch format: %d", format); - return; - } - - for (xmlNode *change = pcmk__xe_first_child(patchset, NULL, NULL, NULL); + for (const xmlNode *change = pcmk__xe_first_child(patchset, NULL, NULL, + NULL); change != NULL; change = pcmk__xe_next(change, NULL)) { const char *op = crm_element_value(change, PCMK_XA_OPERATION); const char *xpath = crm_element_value(change, PCMK_XA_PATH); - const char *shortpath = NULL; + const char *primitive_xpath = NULL; if (pcmk__str_eq(op, PCMK_VALUE_MOVE, pcmk__str_null_matches) || (strstr(xpath, "/" PCMK_XE_STATUS) != NULL)) { continue; } - if (pcmk__str_eq(op, PCMK_VALUE_DELETE, pcmk__str_none) - && (strstr(xpath, "/" PCMK_XE_PRIMITIVE) != NULL)) { + primitive_xpath = strstr(xpath, PRIMITIVE_ID_XP_FRAGMENT); + if ((primitive_xpath != NULL) + && pcmk__str_eq(op, PCMK_VALUE_DELETE, pcmk__str_none)) { + const char *rsc_id = NULL; - char *search = NULL; - char *mutable = NULL; + const char *end_quote = NULL; - if ((strstr(xpath, PCMK_XE_INSTANCE_ATTRIBUTES) != NULL) - || (strstr(xpath, PCMK_XE_META_ATTRIBUTES) != NULL)) { + if ((strstr(primitive_xpath, PCMK_XE_INSTANCE_ATTRIBUTES) != NULL) + || (strstr(primitive_xpath, PCMK_XE_META_ATTRIBUTES) != NULL)) { reason = pcmk__str_copy("(meta) attribute deleted from " "resource"); break; } - mutable = pcmk__str_copy(xpath); - rsc_id = strstr(mutable, PCMK_XE_PRIMITIVE "[@" PCMK_XA_ID "=\'"); - if (rsc_id != NULL) { - rsc_id += strlen(PCMK_XE_PRIMITIVE "[@" PCMK_XA_ID "=\'"); - search = strchr(rsc_id, '\''); + + rsc_id = primitive_xpath + sizeof(PRIMITIVE_ID_XP_FRAGMENT) - 1; + end_quote = strchr(rsc_id, '\''); + + CRM_LOG_ASSERT(end_quote != NULL); + if (end_quote == NULL) { + crm_err("Bug: Malformed item in Pacemaker-generated patchset"); + continue; } - if (search != NULL) { - *search = 0; - stonith_device_remove(rsc_id, true); + + if (strchr(end_quote, '/') == NULL) { + /* The primitive element itself was deleted. If this was a + * fencing resource, it's faster to remove it directly than to + * run the scheduler and update all device registrations. + */ + char *copy = strndup(rsc_id, end_quote - rsc_id); + + pcmk__assert(copy != NULL); + stonith_device_remove(copy, true); + /* watchdog_device_update called afterwards to fall back to implicit definition if needed */ - } else { - crm_warn("Ignoring malformed CIB update (resource deletion)"); + + free(copy); + continue; } - free(mutable); - - } else if (strstr(xpath, "/" PCMK_XE_RESOURCES) - || strstr(xpath, "/" PCMK_XE_CONSTRAINTS) - || strstr(xpath, "/" PCMK_XE_RSC_DEFAULTS)) { - shortpath = strrchr(xpath, '/'); - pcmk__assert(shortpath != NULL); - reason = crm_strdup_printf("%s %s", op, shortpath+1); + } + + if (strstr(xpath, "/" PCMK_XE_RESOURCES) + || strstr(xpath, "/" PCMK_XE_CONSTRAINTS) + || strstr(xpath, "/" PCMK_XE_RSC_DEFAULTS)) { + + const char *shortpath = strrchr(xpath, '/'); + + reason = crm_strdup_printf("%s %s", op, shortpath + 1); break; } } if (reason != NULL) { crm_info("Updating device list from CIB: %s", reason); cib_devices_update(); free(reason); } else { crm_trace("No updates for device list found in CIB"); } } static void watchdog_device_update(void) { if (stonith_watchdog_timeout_ms > 0) { - if (!g_hash_table_lookup(device_list, STONITH_WATCHDOG_ID) && - !stonith_watchdog_targets) { + if (!fenced_has_watchdog_device() + && (stonith_watchdog_targets == NULL)) { /* getting here watchdog-fencing enabled, no device there yet and reason isn't stonith_watchdog_targets preventing that */ int rc; xmlNode *xml; xml = create_device_registration_xml( STONITH_WATCHDOG_ID, st_namespace_internal, STONITH_WATCHDOG_AGENT, - NULL, /* stonith_device_register will add our + NULL, /* fenced_device_register() will add our own name as PCMK_STONITH_HOST_LIST param so we can skip that here */ NULL); - rc = stonith_device_register(xml, TRUE); + rc = fenced_device_register(xml, true); pcmk__xml_free(xml); - if (rc != pcmk_ok) { - rc = pcmk_legacy2rc(rc); + if (rc != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; crm_crit("Cannot register watchdog pseudo fence agent: %s", pcmk_rc_str(rc)); stonith_shutdown(0); } } - } else if (g_hash_table_lookup(device_list, STONITH_WATCHDOG_ID) != NULL) { + } else if (fenced_has_watchdog_device()) { /* be silent if no device - todo parameter to stonith_device_remove */ stonith_device_remove(STONITH_WATCHDOG_ID, true); } } /*! * \internal * \brief Query the full CIB * * \return Standard Pacemaker return code */ static int fenced_query_cib(void) { int rc = pcmk_ok; crm_trace("Re-requesting full CIB"); rc = cib_api->cmds->query(cib_api, NULL, &local_cib, cib_sync_call); rc = pcmk_legacy2rc(rc); if (rc == pcmk_rc_ok) { pcmk__assert(local_cib != NULL); } else { crm_err("Couldn't retrieve the CIB: %s " QB_XS " rc=%d", pcmk_rc_str(rc), rc); } return rc; } static void update_fencing_topology(const char *event, xmlNode *msg) { xmlNode *wrapper = pcmk__xe_first_child(msg, PCMK__XE_CIB_UPDATE_RESULT, NULL, NULL); xmlNode *patchset = pcmk__xe_first_child(wrapper, NULL, NULL, NULL); int format = 1; int add[] = { 0, 0, 0 }; int del[] = { 0, 0, 0 }; CRM_CHECK(patchset != NULL, return); crm_element_value_int(patchset, PCMK_XA_FORMAT, &format); if (format != 2) { crm_warn("Unknown patch format: %d", format); return; } xml_patch_versions(patchset, add, del); for (xmlNode *change = pcmk__xe_first_child(patchset, NULL, NULL, NULL); change != NULL; change = pcmk__xe_next(change, NULL)) { const char *op = crm_element_value(change, PCMK_XA_OPERATION); const char *xpath = crm_element_value(change, PCMK_XA_PATH); if (op == NULL) { continue; } if (strstr(xpath, "/" PCMK_XE_FENCING_LEVEL) != NULL) { // Change to a specific entry crm_trace("Handling %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); if (strcmp(op, PCMK_VALUE_DELETE) == 0) { /* We have only path and ID, which is not enough info to remove * a specific entry. Re-initialize the whole topology. */ crm_info("Re-initializing fencing topology after %s operation " "%d.%d.%d for %s", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } if (strcmp(op, PCMK_VALUE_CREATE) == 0) { add_topology_level(change->children); } else if (strcmp(op, PCMK_VALUE_MODIFY) == 0) { xmlNode *match = pcmk__xe_first_child(change, PCMK_XE_CHANGE_RESULT, NULL, NULL); if (match != NULL) { remove_topology_level(match->children); add_topology_level(match->children); } } continue; } if (strstr(xpath, "/" PCMK_XE_FENCING_TOPOLOGY) != NULL) { // Change to the topology in general crm_info("Re-initializing fencing topology after top-level " "%s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } if ((strstr(xpath, "/" PCMK_XE_CONFIGURATION) != NULL) && (pcmk__xe_first_child(change, PCMK_XE_FENCING_TOPOLOGY, NULL, NULL) != NULL) && pcmk__str_any_of(op, PCMK_VALUE_CREATE, PCMK_VALUE_DELETE, NULL)) { // Topology was created or entire configuration section was deleted crm_info("Re-initializing fencing topology after top-level " "%s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } crm_trace("Nothing for us in %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); } } static void update_cib_cache_cb(const char *event, xmlNode * msg) { + xmlNode *patchset = NULL; long long timeout_ms_saved = stonith_watchdog_timeout_ms; bool need_full_refresh = false; if(!have_cib_devices) { crm_trace("Skipping updates until we get a full dump"); return; } else if(msg == NULL) { crm_trace("Missing %s update", event); return; } /* Maintain a local copy of the CIB so that we have full access * to device definitions, location constraints, and node attributes */ if (local_cib != NULL) { int rc = pcmk_ok; xmlNode *wrapper = NULL; - xmlNode *patchset = NULL; crm_element_value_int(msg, PCMK__XA_CIB_RC, &rc); if (rc != pcmk_ok) { return; } wrapper = pcmk__xe_first_child(msg, PCMK__XE_CIB_UPDATE_RESULT, NULL, NULL); patchset = pcmk__xe_first_child(wrapper, NULL, NULL, NULL); rc = xml_apply_patchset(local_cib, patchset, TRUE); switch (rc) { case pcmk_ok: case -pcmk_err_old_data: + /* @TODO Full refresh (with or without query) in case of + * -pcmk_err_old_data? It seems wrong to call + * stonith_device_remove() based on primitive deletion in an + * old diff. + */ break; case -pcmk_err_diff_resync: case -pcmk_err_diff_failed: crm_notice("[%s] Patch aborted: %s (%d)", event, pcmk_strerror(rc), rc); pcmk__xml_free(local_cib); local_cib = NULL; break; default: crm_warn("[%s] ABORTED: %s (%d)", event, pcmk_strerror(rc), rc); pcmk__xml_free(local_cib); local_cib = NULL; } } if (local_cib == NULL) { if (fenced_query_cib() != pcmk_rc_ok) { return; } need_full_refresh = true; } pcmk__refresh_node_caches_from_cib(local_cib); update_stonith_watchdog_timeout_ms(local_cib); if (timeout_ms_saved != stonith_watchdog_timeout_ms) { need_full_refresh = true; } if (need_full_refresh) { fencing_topology_init(); cib_devices_update(); } else { // Partial refresh update_fencing_topology(event, msg); - update_cib_stonith_devices(event, msg); + update_cib_stonith_devices(patchset); } watchdog_device_update(); } static void init_cib_cache_cb(xmlNode * msg, int call_id, int rc, xmlNode * output, void *user_data) { crm_info("Updating device list from CIB"); have_cib_devices = TRUE; local_cib = pcmk__xml_copy(NULL, output); pcmk__refresh_node_caches_from_cib(local_cib); update_stonith_watchdog_timeout_ms(local_cib); fencing_topology_init(); cib_devices_update(); watchdog_device_update(); } static void cib_connection_destroy(gpointer user_data) { if (stonith_shutdown_flag) { crm_info("Connection to the CIB manager closed"); return; } else { crm_crit("Lost connection to the CIB manager, shutting down"); } if (cib_api) { cib_api->cmds->signoff(cib_api); } stonith_shutdown(0); } /*! * \internal * \brief Disconnect from CIB manager */ void fenced_cib_cleanup(void) { if (cib_api != NULL) { cib_api->cmds->del_notify_callback(cib_api, PCMK__VALUE_CIB_DIFF_NOTIFY, update_cib_cache_cb); cib__clean_up_connection(&cib_api); } pcmk__xml_free(local_cib); local_cib = NULL; } void setup_cib(void) { int rc, retries = 0; cib_api = cib_new(); if (cib_api == NULL) { crm_err("No connection to the CIB manager"); return; } do { sleep(retries); rc = cib_api->cmds->signon(cib_api, crm_system_name, cib_command); } while (rc == -ENOTCONN && ++retries < 5); if (rc != pcmk_ok) { crm_err("Could not connect to the CIB manager: %s (%d)", pcmk_strerror(rc), rc); return; } rc = cib_api->cmds->add_notify_callback(cib_api, PCMK__VALUE_CIB_DIFF_NOTIFY, update_cib_cache_cb); if (rc != pcmk_ok) { crm_err("Could not set CIB notification callback"); return; } rc = cib_api->cmds->query(cib_api, NULL, NULL, cib_none); cib_api->cmds->register_callback(cib_api, rc, 120, FALSE, NULL, "init_cib_cache_cb", init_cib_cache_cb); cib_api->cmds->set_connection_dnotify(cib_api, cib_connection_destroy); crm_info("Watching for fencing topology changes"); } diff --git a/daemons/fenced/fenced_commands.c b/daemons/fenced/fenced_commands.c index 51666992bf..b31c52bf94 100644 --- a/daemons/fenced/fenced_commands.c +++ b/daemons/fenced/fenced_commands.c @@ -1,3669 +1,3677 @@ /* * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include +#include // bool #include #include #include #include #include #include #include #include #include #include #include // xmlNode #include // xmlXPathObject, etc. #include #include #include #include #include #include #include #include #include -GHashTable *device_list = NULL; +static GHashTable *device_table = NULL; + GHashTable *topology = NULL; static GList *cmd_list = NULL; static GHashTable *fenced_handlers = NULL; struct device_search_s { /* target of fence action */ char *host; /* requested fence action */ char *action; /* timeout to use if a device is queried dynamically for possible targets */ // @TODO This name is misleading now, it's the value of stonith-timeout int per_device_timeout; /* number of registered fencing devices at time of request */ int replies_needed; /* number of device replies received so far */ int replies_received; /* whether the target is eligible to perform requested action (or off) */ bool allow_self; /* private data to pass to search callback function */ void *user_data; /* function to call when all replies have been received */ void (*callback) (GList * devices, void *user_data); /* devices capable of performing requested action (or off if remapping) */ GList *capable; /* Whether to perform searches that support the action */ uint32_t support_action_only; }; static gboolean stonith_device_dispatch(gpointer user_data); static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data); static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence); static int get_agent_metadata(const char *agent, xmlNode **metadata); -static void read_action_metadata(stonith_device_t *device); +static void read_action_metadata(fenced_device_t *device); static enum fenced_target_by unpack_level_kind(const xmlNode *level); -typedef struct async_command_s { - +typedef struct { int id; - int pid; - int fd_stdout; uint32_t options; int default_timeout; /* seconds */ int timeout; /* seconds */ int start_delay; // seconds (-1 means disable static/random fencing delays) int delay_id; char *op; char *origin; char *client; char *client_name; char *remote_op_id; char *target; - uint32_t target_nodeid; char *action; char *device; + //! Head of device list (used only for freeing list with command object) GList *device_list; - GList *next_device_iter; // device_list entry for next device to execute + + //! Next item to process in \c device_list + GList *next_device_iter; void *internal_user_data; void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data); - guint timer_sigterm; - guint timer_sigkill; - /*! If the operation timed out, this is the last signal - * we sent to the process to get it to terminate */ - int last_timeout_signo; - - stonith_device_t *active_on; - stonith_device_t *activating_on; + + fenced_device_t *active_on; + fenced_device_t *activating_on; } async_command_t; static xmlNode *construct_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result); +/*! + * \internal + * \brief Check whether the fencer's device table contains a watchdog device + * + * \retval \c true If the device table contains a watchdog device + * \retval \c false Otherwise + */ +bool +fenced_has_watchdog_device(void) +{ + return (device_table != NULL) + && (g_hash_table_lookup(device_table, STONITH_WATCHDOG_ID) != NULL); +} + +/*! + * \internal + * \brief Call a function for each known fence device + * + * \param[in] fn Function to call for each device + * \param[in,out] user_data User data + */ +void +fenced_foreach_device(GHFunc fn, gpointer user_data) +{ + if (device_table != NULL) { + g_hash_table_foreach(device_table, fn, user_data); + } +} + +/*! + * \internal + * \brief Remove each known fence device matching a given predicate + * + * \param[in] fn Function that returns \c TRUE to remove a fence device or + * \c FALSE to keep it + */ +void +fenced_foreach_device_remove(GHRFunc fn) +{ + if (device_table != NULL) { + g_hash_table_foreach_remove(device_table, fn, NULL); + } +} + static gboolean -is_action_required(const char *action, const stonith_device_t *device) +is_action_required(const char *action, const fenced_device_t *device) { return (device != NULL) && device->automatic_unfencing && pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none); } static int -get_action_delay_max(const stonith_device_t *device, const char *action) +get_action_delay_max(const fenced_device_t *device, const char *action) { const char *value = NULL; guint delay_max = 0U; if (!pcmk__is_fencing_action(action)) { return 0; } value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_MAX); if (value) { pcmk_parse_interval_spec(value, &delay_max); delay_max /= 1000; } return (int) delay_max; } static int -get_action_delay_base(const stonith_device_t *device, const char *action, +get_action_delay_base(const fenced_device_t *device, const char *action, const char *target) { char *hash_value = NULL; guint delay_base = 0U; if (!pcmk__is_fencing_action(action)) { return 0; } hash_value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_BASE); if (hash_value) { char *value = pcmk__str_copy(hash_value); char *valptr = value; if (target != NULL) { for (char *val = strtok(value, "; \t"); val != NULL; val = strtok(NULL, "; \t")) { char *mapval = strchr(val, ':'); if (mapval == NULL || mapval[1] == 0) { crm_err("pcmk_delay_base: empty value in mapping", val); continue; } if (mapval != val && strncasecmp(target, val, (size_t)(mapval - val)) == 0) { value = mapval + 1; crm_debug("pcmk_delay_base mapped to %s for %s", value, target); break; } } } if (strchr(value, ':') == 0) { pcmk_parse_interval_spec(value, &delay_base); delay_base /= 1000; } free(valptr); } return (int) delay_base; } /*! * \internal * \brief Override STONITH timeout with pcmk_*_timeout if available * * \param[in] device STONITH device to use * \param[in] action STONITH action name * \param[in] default_timeout Timeout to use if device does not have * a pcmk_*_timeout parameter for action * * \return Value of pcmk_(action)_timeout if available, otherwise default_timeout * \note For consistency, it would be nice if reboot/off/on timeouts could be * set the same way as start/stop/monitor timeouts, i.e. with an * entry in the fencing resource configuration. However that * is insufficient because fencing devices may be registered directly via * the fencer's register_device() API instead of going through the CIB * (e.g. stonith_admin uses it for its -R option, and the executor uses it * to ensure a device is registered when a command is issued). As device * properties, pcmk_*_timeout parameters can be grabbed by the fencer when * the device is registered, whether by CIB change or API call. */ static int -get_action_timeout(const stonith_device_t *device, const char *action, +get_action_timeout(const fenced_device_t *device, const char *action, int default_timeout) { if (action && device && device->params) { char buffer[64] = { 0, }; const char *value = NULL; /* If "reboot" was requested but the device does not support it, * we will remap to "off", so check timeout for "off" instead */ if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_trace("%s doesn't support reboot, using timeout for off instead", device->id); action = PCMK_ACTION_OFF; } /* If the device config specified an action-specific timeout, use it */ snprintf(buffer, sizeof(buffer), "pcmk_%s_timeout", action); value = g_hash_table_lookup(device->params, buffer); if (value) { long long timeout_ms = crm_get_msec(value); return (int) QB_MIN(pcmk__timeout_ms2s(timeout_ms), INT_MAX); } } return default_timeout; } /*! * \internal * \brief Get the currently executing device for a fencing operation * * \param[in] cmd Fencing operation to check * * \return Currently executing device for \p cmd if any, otherwise NULL */ -static stonith_device_t * +static fenced_device_t * cmd_device(const async_command_t *cmd) { - if ((cmd == NULL) || (cmd->device == NULL) || (device_list == NULL)) { + if ((cmd == NULL) || (cmd->device == NULL) || (device_table == NULL)) { return NULL; } - return g_hash_table_lookup(device_list, cmd->device); + return g_hash_table_lookup(device_table, cmd->device); } /*! * \internal * \brief Return the configured reboot action for a given device * * \param[in] device_id Device ID * * \return Configured reboot action for \p device_id */ const char * fenced_device_reboot_action(const char *device_id) { const char *action = NULL; - if ((device_list != NULL) && (device_id != NULL)) { - stonith_device_t *device = g_hash_table_lookup(device_list, device_id); + if ((device_table != NULL) && (device_id != NULL)) { + fenced_device_t *device = g_hash_table_lookup(device_table, device_id); if ((device != NULL) && (device->params != NULL)) { action = g_hash_table_lookup(device->params, "pcmk_reboot_action"); } } return pcmk__s(action, PCMK_ACTION_REBOOT); } /*! * \internal * \brief Check whether a given device supports the "on" action * * \param[in] device_id Device ID * * \return true if \p device_id supports "on", otherwise false */ bool fenced_device_supports_on(const char *device_id) { - if ((device_list != NULL) && (device_id != NULL)) { - stonith_device_t *device = g_hash_table_lookup(device_list, device_id); + if ((device_table != NULL) && (device_id != NULL)) { + fenced_device_t *device = g_hash_table_lookup(device_table, device_id); if (device != NULL) { return pcmk_is_set(device->flags, st_device_supports_on); } } return false; } static void free_async_command(async_command_t * cmd) { if (!cmd) { return; } if (cmd->delay_id) { g_source_remove(cmd->delay_id); } cmd_list = g_list_remove(cmd_list, cmd); g_list_free_full(cmd->device_list, free); free(cmd->device); free(cmd->action); free(cmd->target); free(cmd->remote_op_id); free(cmd->client); free(cmd->client_name); free(cmd->origin); free(cmd->op); free(cmd); } /*! * \internal * \brief Create a new asynchronous fencing operation from request XML * * \param[in] msg Fencing request XML (from IPC or CPG) * * \return Newly allocated fencing operation on success, otherwise NULL * * \note This asserts on memory errors, so a NULL return indicates an * unparseable message. */ static async_command_t * create_async_command(xmlNode *msg) { xmlNode *op = NULL; async_command_t *cmd = NULL; int rc = pcmk_rc_ok; if (msg == NULL) { return NULL; } op = pcmk__xpath_find_one(msg->doc, "//*[@" PCMK__XA_ST_DEVICE_ACTION "]", LOG_ERR); if (op == NULL) { return NULL; } cmd = pcmk__assert_alloc(1, sizeof(async_command_t)); // All messages must include these cmd->action = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ACTION); cmd->op = crm_element_value_copy(msg, PCMK__XA_ST_OP); cmd->client = crm_element_value_copy(msg, PCMK__XA_ST_CLIENTID); if ((cmd->action == NULL) || (cmd->op == NULL) || (cmd->client == NULL)) { free_async_command(cmd); return NULL; } crm_element_value_int(msg, PCMK__XA_ST_CALLID, &(cmd->id)); crm_element_value_int(msg, PCMK__XA_ST_DELAY, &(cmd->start_delay)); crm_element_value_int(msg, PCMK__XA_ST_TIMEOUT, &(cmd->default_timeout)); cmd->timeout = cmd->default_timeout; rc = pcmk__xe_get_flags(msg, PCMK__XA_ST_CALLOPT, &(cmd->options), st_opt_none); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from request: %s", pcmk_rc_str(rc)); } cmd->origin = crm_element_value_copy(msg, PCMK__XA_SRC); cmd->remote_op_id = crm_element_value_copy(msg, PCMK__XA_ST_REMOTE_OP); cmd->client_name = crm_element_value_copy(msg, PCMK__XA_ST_CLIENTNAME); cmd->target = crm_element_value_copy(op, PCMK__XA_ST_TARGET); cmd->device = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ID); cmd->done_cb = st_child_done; // Track in global command list cmd_list = g_list_append(cmd_list, cmd); return cmd; } static int -get_action_limit(stonith_device_t * device) +get_action_limit(fenced_device_t *device) { const char *value = NULL; int action_limit = 1; value = g_hash_table_lookup(device->params, PCMK_STONITH_ACTION_LIMIT); if ((value == NULL) || (pcmk__scan_min_int(value, &action_limit, INT_MIN) != pcmk_rc_ok) || (action_limit == 0)) { action_limit = 1; } return action_limit; } static int -get_active_cmds(stonith_device_t * device) +get_active_cmds(fenced_device_t *device) { int counter = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return 0); for (gIter = cmd_list; gIter != NULL; gIter = gIterNext) { async_command_t *cmd = gIter->data; gIterNext = gIter->next; if (cmd->active_on == device) { counter++; } } return counter; } static void fork_cb(int pid, void *user_data) { async_command_t *cmd = (async_command_t *) user_data; - stonith_device_t * device = - /* in case of a retry we've done the move from - activating_on to active_on already + fenced_device_t *device = cmd->activating_on; + + if (device == NULL) { + /* In case of a retry, we've done the move from activating_on to + * active_on already */ - cmd->activating_on?cmd->activating_on:cmd->active_on; + device = cmd->active_on; + } pcmk__assert(device != NULL); crm_debug("Operation '%s' [%d]%s%s using %s now running with %ds timeout", cmd->action, pid, ((cmd->target == NULL)? "" : " targeting "), pcmk__s(cmd->target, ""), device->id, cmd->timeout); cmd->active_on = device; cmd->activating_on = NULL; } static int get_agent_metadata_cb(gpointer data) { - stonith_device_t *device = data; + fenced_device_t *device = data; guint period_ms; switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } return G_SOURCE_REMOVE; case EAGAIN: period_ms = pcmk__mainloop_timer_get_period(device->timer); if (period_ms < 160 * 1000) { mainloop_timer_set_period(device->timer, 2 * period_ms); } return G_SOURCE_CONTINUE; default: return G_SOURCE_REMOVE; } } /*! * \internal * \brief Call a command's action callback for an internal (not library) result * * \param[in,out] cmd Command to report result for * \param[in] execution_status Execution status to use for result * \param[in] exit_status Exit status to use for result * \param[in] exit_reason Exit reason to use for result */ static void report_internal_result(async_command_t *cmd, int exit_status, int execution_status, const char *exit_reason) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__set_result(&result, exit_status, execution_status, exit_reason); cmd->done_cb(0, &result, cmd); pcmk__reset_result(&result); } static gboolean -stonith_device_execute(stonith_device_t * device) +stonith_device_execute(fenced_device_t *device) { int exec_rc = 0; const char *action_str = NULL; const char *host_arg = NULL; async_command_t *cmd = NULL; stonith_action_t *action = NULL; int active_cmds = 0; int action_limit = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return FALSE); active_cmds = get_active_cmds(device); action_limit = get_action_limit(device); if (action_limit > -1 && active_cmds >= action_limit) { crm_trace("%s is over its action limit of %d (%u active action%s)", device->id, action_limit, active_cmds, pcmk__plural_s(active_cmds)); return TRUE; } for (gIter = device->pending_ops; gIter != NULL; gIter = gIterNext) { async_command_t *pending_op = gIter->data; gIterNext = gIter->next; if (pending_op && pending_op->delay_id) { crm_trace("Operation '%s'%s%s using %s was asked to run too early, " "waiting for start delay of %ds", pending_op->action, ((pending_op->target == NULL)? "" : " targeting "), pcmk__s(pending_op->target, ""), device->id, pending_op->start_delay); continue; } device->pending_ops = g_list_remove_link(device->pending_ops, gIter); g_list_free_1(gIter); cmd = pending_op; break; } if (cmd == NULL) { crm_trace("No actions using %s are needed", device->id); return TRUE; } if (pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { if (pcmk__is_fencing_action(cmd->action)) { if (node_does_watchdog_fencing(fenced_get_local_node())) { pcmk__panic("Watchdog self-fencing required"); goto done; } } else { crm_info("Faking success for %s watchdog operation", cmd->action); report_internal_result(cmd, CRM_EX_OK, PCMK_EXEC_DONE, NULL); goto done; } } #if PCMK__ENABLE_CIBSECRETS exec_rc = pcmk__substitute_secrets(device->id, device->params); if (exec_rc != pcmk_rc_ok) { if (pcmk__str_eq(cmd->action, PCMK_ACTION_STOP, pcmk__str_none)) { crm_info("Proceeding with stop operation for %s " "despite being unable to load CIB secrets (%s)", device->id, pcmk_rc_str(exec_rc)); } else { crm_err("Considering %s unconfigured " "because unable to load CIB secrets: %s", device->id, pcmk_rc_str(exec_rc)); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_SECRETS, "Failed to get CIB secrets"); goto done; } } #endif action_str = cmd->action; if (pcmk__str_eq(cmd->action, PCMK_ACTION_REBOOT, pcmk__str_none) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_notice("Remapping 'reboot' action%s%s using %s to 'off' " "because agent '%s' does not support reboot", ((cmd->target == NULL)? "" : " targeting "), pcmk__s(cmd->target, ""), device->id, device->agent); action_str = PCMK_ACTION_OFF; } if (pcmk_is_set(device->flags, st_device_supports_parameter_port)) { host_arg = "port"; } else if (pcmk_is_set(device->flags, st_device_supports_parameter_plug)) { host_arg = "plug"; } action = stonith__action_create(device->agent, action_str, cmd->target, - cmd->target_nodeid, cmd->timeout, - device->params, device->aliases, host_arg); + cmd->timeout, device->params, + device->aliases, host_arg); /* for async exec, exec_rc is negative for early error exit otherwise handling of success/errors is done via callbacks */ cmd->activating_on = device; exec_rc = stonith__execute_async(action, (void *)cmd, cmd->done_cb, fork_cb); if (exec_rc < 0) { cmd->activating_on = NULL; cmd->done_cb(0, stonith__action_result(action), cmd); stonith__destroy_action(action); } done: /* Device might get triggered to work by multiple fencing commands * simultaneously. Trigger the device again to make sure any * remaining concurrent commands get executed. */ if (device->pending_ops) { mainloop_set_trigger(device->work); } return TRUE; } static gboolean stonith_device_dispatch(gpointer user_data) { return stonith_device_execute(user_data); } static gboolean start_delay_helper(gpointer data) { async_command_t *cmd = data; - stonith_device_t *device = cmd_device(cmd); + fenced_device_t *device = cmd_device(cmd); cmd->delay_id = 0; if (device) { mainloop_set_trigger(device->work); } return FALSE; } static void -schedule_stonith_command(async_command_t * cmd, stonith_device_t * device) +schedule_stonith_command(async_command_t *cmd, fenced_device_t *device) { int delay_max = 0; int delay_base = 0; int requested_delay = cmd->start_delay; CRM_CHECK(cmd != NULL, return); CRM_CHECK(device != NULL, return); if (cmd->device) { free(cmd->device); } - if (device->include_nodeid && (cmd->target != NULL)) { - pcmk__node_status_t *node = - pcmk__get_node(0, cmd->target, NULL, - pcmk__node_search_cluster_member); - - cmd->target_nodeid = node->cluster_layer_id; - } - cmd->device = pcmk__str_copy(device->id); cmd->timeout = get_action_timeout(device, cmd->action, cmd->default_timeout); if (cmd->remote_op_id) { crm_debug("Scheduling '%s' action%s%s using %s for remote peer %s " "with op id %.8s and timeout %ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->origin, cmd->remote_op_id, cmd->timeout); } else { crm_debug("Scheduling '%s' action%s%s using %s for %s with timeout %ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->client, cmd->timeout); } device->pending_ops = g_list_append(device->pending_ops, cmd); mainloop_set_trigger(device->work); // Value -1 means disable any static/random fencing delays if (requested_delay < 0) { return; } delay_max = get_action_delay_max(device, cmd->action); delay_base = get_action_delay_base(device, cmd->action, cmd->target); if (delay_max == 0) { delay_max = delay_base; } if (delay_max < delay_base) { crm_warn(PCMK_STONITH_DELAY_BASE " (%ds) is larger than " PCMK_STONITH_DELAY_MAX " (%ds) for %s using %s " "(limiting to maximum delay)", delay_base, delay_max, cmd->action, device->id); delay_base = delay_max; } if (delay_max > 0) { // coverity[dontcall] It doesn't matter here if rand() is predictable cmd->start_delay += ((delay_max != delay_base)?(rand() % (delay_max - delay_base)):0) + delay_base; } if (cmd->start_delay > 0) { crm_notice("Delaying '%s' action%s%s using %s for %ds " QB_XS " timeout=%ds requested_delay=%ds base=%ds max=%ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->start_delay, cmd->timeout, requested_delay, delay_base, delay_max); cmd->delay_id = pcmk__create_timer(cmd->start_delay * 1000, start_delay_helper, cmd); } } static void free_device(gpointer data) { GList *gIter = NULL; - stonith_device_t *device = data; + fenced_device_t *device = data; g_hash_table_destroy(device->params); g_hash_table_destroy(device->aliases); for (gIter = device->pending_ops; gIter != NULL; gIter = gIter->next) { async_command_t *cmd = gIter->data; crm_warn("Removal of device '%s' purged operation '%s'", device->id, cmd->action); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Device was removed before action could be executed"); } g_list_free(device->pending_ops); g_list_free_full(device->targets, free); if (device->timer) { mainloop_timer_stop(device->timer); mainloop_timer_del(device->timer); } mainloop_destroy_trigger(device->work); pcmk__xml_free(device->agent_metadata); free(device->namespace); if (device->on_target_actions != NULL) { g_string_free(device->on_target_actions, TRUE); } free(device->agent); free(device->id); free(device); } -void free_device_list(void) +/*! + * \internal + * \brief Initialize the table of known fence devices + */ +void +fenced_init_device_table(void) { - if (device_list != NULL) { - g_hash_table_destroy(device_list); - device_list = NULL; + if (device_table == NULL) { + device_table = pcmk__strkey_table(NULL, free_device); } } +/*! + * \internal + * \brief Free the table of known fence devices + */ void -init_device_list(void) +fenced_free_device_table(void) { - if (device_list == NULL) { - device_list = pcmk__strkey_table(NULL, free_device); + if (device_table != NULL) { + g_hash_table_destroy(device_table); + device_table = NULL; } } static GHashTable * build_port_aliases(const char *hostmap, GList ** targets) { char *name = NULL; int last = 0, lpc = 0, max = 0, added = 0; GHashTable *aliases = pcmk__strikey_table(free, free); if (hostmap == NULL) { return aliases; } max = strlen(hostmap); for (; lpc <= max; lpc++) { switch (hostmap[lpc]) { /* Skip escaped chars */ case '\\': lpc++; break; /* Assignment chars */ case '=': case ':': if (lpc > last) { free(name); name = pcmk__assert_alloc(1, 1 + lpc - last); memcpy(name, hostmap + last, lpc - last); } last = lpc + 1; break; /* Delimeter chars */ /* case ',': Potentially used to specify multiple ports */ case 0: case ';': case ' ': case '\t': if (name) { char *value = NULL; int k = 0; value = pcmk__assert_alloc(1, 1 + lpc - last); memcpy(value, hostmap + last, lpc - last); for (int i = 0; value[i] != '\0'; i++) { if (value[i] != '\\') { value[k++] = value[i]; } } value[k] = '\0'; crm_debug("Adding alias '%s'='%s'", name, value); g_hash_table_replace(aliases, name, value); if (targets) { *targets = g_list_append(*targets, pcmk__str_copy(value)); } value = NULL; name = NULL; added++; } else if (lpc > last) { crm_debug("Parse error at offset %d near '%s'", lpc - last, hostmap + last); } last = lpc + 1; break; } if (hostmap[lpc] == 0) { break; } } if (added == 0) { crm_info("No host mappings detected in '%s'", hostmap); } free(name); return aliases; } GHashTable *metadata_cache = NULL; void free_metadata_cache(void) { if (metadata_cache != NULL) { g_hash_table_destroy(metadata_cache); metadata_cache = NULL; } } static void init_metadata_cache(void) { if (metadata_cache == NULL) { metadata_cache = pcmk__strkey_table(free, free); } } int get_agent_metadata(const char *agent, xmlNode ** metadata) { char *buffer = NULL; if (metadata == NULL) { return EINVAL; } *metadata = NULL; if (pcmk__str_eq(agent, STONITH_WATCHDOG_AGENT_INTERNAL, pcmk__str_none)) { return pcmk_rc_ok; } init_metadata_cache(); buffer = g_hash_table_lookup(metadata_cache, agent); if (buffer == NULL) { stonith_t *st = stonith_api_new(); int rc; if (st == NULL) { crm_warn("Could not get agent meta-data: " "API memory allocation failed"); return EAGAIN; } rc = st->cmds->metadata(st, st_opt_sync_call, agent, NULL, &buffer, 10); stonith_api_delete(st); if (rc || !buffer) { crm_err("Could not retrieve metadata for fencing agent %s", agent); return EAGAIN; } g_hash_table_replace(metadata_cache, pcmk__str_copy(agent), buffer); } *metadata = pcmk__xml_parse(buffer); return pcmk_rc_ok; } -static gboolean -is_nodeid_required(xmlNode * xml) -{ - xmlXPathObject *xpath = NULL; - - if (!xml) { - return FALSE; - } - - xpath = pcmk__xpath_search(xml->doc, - "//" PCMK_XE_PARAMETER - "[@" PCMK_XA_NAME "='nodeid']"); - if (pcmk__xpath_num_results(xpath) == 0) { - xmlXPathFreeObject(xpath); - return FALSE; - } - - xmlXPathFreeObject(xpath); - return TRUE; -} - static void -read_action_metadata(stonith_device_t *device) +read_action_metadata(fenced_device_t *device) { xmlXPathObject *xpath = NULL; int max = 0; int lpc = 0; if (device->agent_metadata == NULL) { return; } xpath = pcmk__xpath_search(device->agent_metadata->doc, "//" PCMK_XE_ACTION); max = pcmk__xpath_num_results(xpath); if (max == 0) { xmlXPathFreeObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *action = NULL; xmlNode *match = pcmk__xpath_result(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if(match == NULL) { continue; }; action = crm_element_value(match, PCMK_XA_NAME); if (pcmk__str_eq(action, PCMK_ACTION_LIST, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_list); } else if (pcmk__str_eq(action, PCMK_ACTION_STATUS, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_status); } else if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_reboot); } else if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none)) { /* PCMK_XA_AUTOMATIC means the cluster will unfence a node when it * joins. * * @COMPAT PCMK__XA_REQUIRED is a deprecated synonym for * PCMK_XA_AUTOMATIC. */ if (pcmk__xe_attr_is_true(match, PCMK_XA_AUTOMATIC) || pcmk__xe_attr_is_true(match, PCMK__XA_REQUIRED)) { device->automatic_unfencing = TRUE; } stonith__set_device_flags(device->flags, device->id, st_device_supports_on); } if ((action != NULL) && pcmk__xe_attr_is_true(match, PCMK_XA_ON_TARGET)) { pcmk__add_word(&(device->on_target_actions), 64, action); } } xmlXPathFreeObject(xpath); } static const char * -target_list_type(stonith_device_t * dev) +target_list_type(fenced_device_t *dev) { const char *check_type = NULL; check_type = g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK); if (check_type == NULL) { if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_LIST)) { check_type = PCMK_VALUE_STATIC_LIST; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP)) { check_type = PCMK_VALUE_STATIC_LIST; } else if (pcmk_is_set(dev->flags, st_device_supports_list)) { check_type = PCMK_VALUE_DYNAMIC_LIST; } else if (pcmk_is_set(dev->flags, st_device_supports_status)) { check_type = PCMK_VALUE_STATUS; } else { check_type = PCMK_VALUE_NONE; } } return check_type; } -static stonith_device_t * -build_device_from_xml(xmlNode *dev) +static fenced_device_t * +build_device_from_xml(const xmlNode *dev) { const char *value; - stonith_device_t *device = NULL; + fenced_device_t *device = NULL; char *agent = crm_element_value_copy(dev, PCMK_XA_AGENT); CRM_CHECK(agent != NULL, return device); - device = pcmk__assert_alloc(1, sizeof(stonith_device_t)); + device = pcmk__assert_alloc(1, sizeof(fenced_device_t)); device->id = crm_element_value_copy(dev, PCMK_XA_ID); device->agent = agent; device->namespace = crm_element_value_copy(dev, PCMK__XA_NAMESPACE); device->params = xml2list(dev); value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_LIST); if (value) { device->targets = stonith__parse_targets(value); } value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_MAP); device->aliases = build_port_aliases(value, &(device->targets)); value = target_list_type(device); if (!pcmk__str_eq(value, PCMK_VALUE_STATIC_LIST, pcmk__str_casei) && (device->targets != NULL)) { // device->targets is necessary only with PCMK_VALUE_STATIC_LIST g_list_free_full(device->targets, free); device->targets = NULL; } switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } break; case EAGAIN: if (device->timer == NULL) { device->timer = mainloop_timer_add("get_agent_metadata", 10 * 1000, TRUE, get_agent_metadata_cb, device); } if (!mainloop_timer_running(device->timer)) { mainloop_timer_start(device->timer); } break; default: break; } - value = g_hash_table_lookup(device->params, "nodeid"); - if (!value) { - device->include_nodeid = is_nodeid_required(device->agent_metadata); - } - value = crm_element_value(dev, PCMK__XA_RSC_PROVIDES); if (pcmk__str_eq(value, PCMK_VALUE_UNFENCING, pcmk__str_casei)) { device->automatic_unfencing = TRUE; } if (is_action_required(PCMK_ACTION_ON, device)) { crm_info("Fencing device '%s' requires unfencing", device->id); } if (device->on_target_actions != NULL) { crm_info("Fencing device '%s' requires actions (%s) to be executed " "on target", device->id, (const char *) device->on_target_actions->str); } device->work = mainloop_add_trigger(G_PRIORITY_HIGH, stonith_device_dispatch, device); - /* TODO: Hook up priority */ return device; } static void -schedule_internal_command(const char *origin, - stonith_device_t * device, - const char *action, - const char *target, - int timeout, +schedule_internal_command(const char *origin, fenced_device_t *device, + const char *action, const char *target, int timeout, void *internal_user_data, void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data)) { async_command_t *cmd = NULL; cmd = pcmk__assert_alloc(1, sizeof(async_command_t)); cmd->id = -1; cmd->default_timeout = timeout ? timeout : 60; cmd->timeout = cmd->default_timeout; cmd->action = pcmk__str_copy(action); cmd->target = pcmk__str_copy(target); cmd->device = pcmk__str_copy(device->id); cmd->origin = pcmk__str_copy(origin); cmd->client = pcmk__str_copy(crm_system_name); cmd->client_name = pcmk__str_copy(crm_system_name); cmd->internal_user_data = internal_user_data; cmd->done_cb = done_cb; /* cmd, not internal_user_data, is passed to 'done_cb' as the userdata */ schedule_stonith_command(cmd, device); } // Fence agent status commands use custom exit status codes enum fence_status_code { fence_status_invalid = -1, fence_status_active = 0, fence_status_unknown = 1, fence_status_inactive = 2, }; static void status_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; - stonith_device_t *dev = cmd_device(cmd); + fenced_device_t *dev = cmd_device(cmd); gboolean can = FALSE; free_async_command(cmd); if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (result->execution_status != PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because status could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); search_devices_record_result(search, dev->id, FALSE); return; } switch (result->exit_status) { case fence_status_unknown: crm_trace("%s reported it cannot fence %s", dev->id, search->host); break; case fence_status_active: case fence_status_inactive: crm_trace("%s reported it can fence %s", dev->id, search->host); can = TRUE; break; default: crm_warn("Assuming %s cannot fence %s " "(status returned unknown code %d)", dev->id, search->host, result->exit_status); break; } search_devices_record_result(search, dev->id, can); } static void dynamic_list_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; - stonith_device_t *dev = cmd_device(cmd); + fenced_device_t *dev = cmd_device(cmd); gboolean can_fence = FALSE; free_async_command(cmd); /* Host/alias must be in the list output to be eligible to be fenced * * Will cause problems if down'd nodes aren't listed or (for virtual nodes) * if the guest is still listed despite being moved to another machine */ if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (pcmk__result_ok(result)) { crm_info("Refreshing target list for %s", dev->id); g_list_free_full(dev->targets, free); dev->targets = stonith__parse_targets(result->action_stdout); dev->targets_age = time(NULL); } else if (dev->targets != NULL) { if (result->execution_status == PCMK_EXEC_DONE) { crm_info("Reusing most recent target list for %s " "because list returned error code %d", dev->id, result->exit_status); } else { crm_info("Reusing most recent target list for %s " "because list could not be executed: %s%s%s%s", dev->id, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } } else { // We have never successfully executed list if (result->execution_status == PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because list returned error code %d", dev->id, search->host, result->exit_status); } else { crm_warn("Assuming %s cannot fence %s " "because list could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } /* Fall back to pcmk_host_check=PCMK_VALUE_STATUS if the user didn't * explicitly specify PCMK_VALUE_DYNAMIC_LIST */ if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK) == NULL) { crm_notice("Switching to pcmk_host_check='status' for %s", dev->id); pcmk__insert_dup(dev->params, PCMK_STONITH_HOST_CHECK, PCMK_VALUE_STATUS); } } if (dev->targets) { const char *alias = g_hash_table_lookup(dev->aliases, search->host); if (!alias) { alias = search->host; } if (pcmk__str_in_list(alias, dev->targets, pcmk__str_casei)) { can_fence = TRUE; } } search_devices_record_result(search, dev->id, can_fence); } /*! * \internal * \brief Returns true if any key in first is not in second or second has a different value for key */ static int device_params_diff(GHashTable *first, GHashTable *second) { char *key = NULL; char *value = NULL; GHashTableIter gIter; g_hash_table_iter_init(&gIter, first); while (g_hash_table_iter_next(&gIter, (void **)&key, (void **)&value)) { if(strstr(key, "CRM_meta") == key) { continue; } else if (strcmp(key, PCMK_XA_CRM_FEATURE_SET) == 0) { continue; } else { char *other_value = g_hash_table_lookup(second, key); if (!other_value || !pcmk__str_eq(other_value, value, pcmk__str_casei)) { crm_trace("Different value for %s: %s != %s", key, other_value, value); return 1; } } } return 0; } /*! * \internal - * \brief Checks to see if an identical device already exists in the device_list + * \brief Checks to see if an identical device already exists in the table */ -static stonith_device_t * -device_has_duplicate(const stonith_device_t *device) +static fenced_device_t * +device_has_duplicate(const fenced_device_t *device) { - stonith_device_t *dup = g_hash_table_lookup(device_list, device->id); + fenced_device_t *dup = g_hash_table_lookup(device_table, device->id); if (!dup) { crm_trace("No match for %s", device->id); return NULL; } else if (!pcmk__str_eq(dup->agent, device->agent, pcmk__str_casei)) { crm_trace("Different agent: %s != %s", dup->agent, device->agent); return NULL; } // Use pcmk__digest_operation() here? if (device_params_diff(device->params, dup->params) || device_params_diff(dup->params, device->params)) { return NULL; } crm_trace("Match"); return dup; } int -stonith_device_register(xmlNode *dev, gboolean from_cib) +fenced_device_register(const xmlNode *dev, bool from_cib) { - stonith_device_t *dup = NULL; - stonith_device_t *device = build_device_from_xml(dev); - guint ndevices = 0; - int rv = pcmk_ok; + const char *local_node_name = fenced_get_local_node(); + fenced_device_t *dup = NULL; + fenced_device_t *device = build_device_from_xml(dev); + int rc = pcmk_rc_ok; - CRM_CHECK(device != NULL, return -ENOMEM); + CRM_CHECK(device != NULL, return ENOMEM); /* do we have a watchdog-device? */ - if (pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none) || - pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, - STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) do { + if (pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none) + || pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, + STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { + if (stonith_watchdog_timeout_ms <= 0) { crm_err("Ignoring watchdog fence device without " - PCMK_OPT_STONITH_WATCHDOG_TIMEOUT " set."); - rv = -ENODEV; - /* fall through to cleanup & return */ - } else if (!pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, - STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { - crm_err("Ignoring watchdog fence device with unknown " - "agent '%s' unequal '" STONITH_WATCHDOG_AGENT "'.", - device->agent?device->agent:""); - rv = -ENODEV; - /* fall through to cleanup & return */ - } else if (!pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, - pcmk__str_none)) { - crm_err("Ignoring watchdog fence device " - "named %s !='"STONITH_WATCHDOG_ID"'.", - device->id?device->id:""); - rv = -ENODEV; - /* fall through to cleanup & return */ - } else { - const char *local_node_name = fenced_get_local_node(); + PCMK_OPT_STONITH_WATCHDOG_TIMEOUT " set"); + rc = ENODEV; + goto done; + } + if (!pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, + STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { + crm_err("Ignoring watchdog fence device with unknown agent '%s' " + "rather than '" STONITH_WATCHDOG_AGENT "'", + pcmk__s(device->agent, "")); + rc = ENODEV; + goto done; + } + if (!pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none)) { + crm_err("Ignoring watchdog fence device named '%s' rather than " + "'" STONITH_WATCHDOG_ID "'", + pcmk__s(device->id, "")); + rc = ENODEV; + goto done; + } - if (pcmk__str_eq(device->agent, STONITH_WATCHDOG_AGENT, - pcmk__str_none)) { - /* this either has an empty list or the targets - configured for watchdog-fencing - */ - g_list_free_full(stonith_watchdog_targets, free); - stonith_watchdog_targets = device->targets; - device->targets = NULL; - } - if (node_does_watchdog_fencing(local_node_name)) { - g_list_free_full(device->targets, free); - device->targets = stonith__parse_targets(local_node_name); - pcmk__insert_dup(device->params, - PCMK_STONITH_HOST_LIST, local_node_name); - /* proceed as with any other stonith-device */ - break; - } + if (pcmk__str_eq(device->agent, STONITH_WATCHDOG_AGENT, + pcmk__str_none)) { + /* This has either an empty list or the targets configured for + * watchdog fencing + */ + g_list_free_full(stonith_watchdog_targets, free); + stonith_watchdog_targets = device->targets; + device->targets = NULL; + } - crm_debug("Skip registration of watchdog fence device on node not in host-list."); - /* cleanup and fall through to more cleanup and return */ + if (!node_does_watchdog_fencing(local_node_name)) { + crm_debug("Skip registration of watchdog fence device on node not " + "in host list"); device->targets = NULL; stonith_device_remove(device->id, from_cib); + goto done; } - free_device(device); - return rv; - } while (0); + + // Proceed as with any other fencing device + g_list_free_full(device->targets, free); + device->targets = stonith__parse_targets(local_node_name); + pcmk__insert_dup(device->params, PCMK_STONITH_HOST_LIST, + local_node_name); + } dup = device_has_duplicate(device); - if (dup) { - ndevices = g_hash_table_size(device_list); + if (dup != NULL) { + guint ndevices = g_hash_table_size(device_table); + crm_debug("Device '%s' already in device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); free_device(device); device = dup; - dup = g_hash_table_lookup(device_list, device->id); - dup->dirty = FALSE; + device->dirty = FALSE; } else { - stonith_device_t *old = g_hash_table_lookup(device_list, device->id); - - if (from_cib && old && old->api_registered) { - /* If the cib is writing over an entry that is shared with a stonith client, - * copy any pending ops that currently exist on the old entry to the new one. - * Otherwise the pending ops will be reported as failures + guint ndevices = 0; + fenced_device_t *old = g_hash_table_lookup(device_table, device->id); + + if (from_cib && (old != NULL) && old->api_registered) { + /* If the CIB is writing over an entry that is shared with a stonith + * client, copy any pending ops that currently exist on the old + * entry to the new one. Otherwise the pending ops will be reported + * as failures. */ crm_info("Overwriting existing entry for %s from CIB", device->id); device->pending_ops = old->pending_ops; device->api_registered = TRUE; old->pending_ops = NULL; - if (device->pending_ops) { + if (device->pending_ops != NULL) { mainloop_set_trigger(device->work); } } - g_hash_table_replace(device_list, device->id, device); + g_hash_table_replace(device_table, device->id, device); - ndevices = g_hash_table_size(device_list); + ndevices = g_hash_table_size(device_table); crm_notice("Added '%s' to device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); } if (from_cib) { device->cib_registered = TRUE; } else { device->api_registered = TRUE; } - return pcmk_ok; +done: + if (rc != pcmk_rc_ok) { + free_device(device); + } + return rc; } void stonith_device_remove(const char *id, bool from_cib) { - stonith_device_t *device = g_hash_table_lookup(device_list, id); + fenced_device_t *device = g_hash_table_lookup(device_table, id); guint ndevices = 0; - if (!device) { - ndevices = g_hash_table_size(device_list); - crm_info("Device '%s' not found (%d active device%s)", - id, ndevices, pcmk__plural_s(ndevices)); + if (device == NULL) { + ndevices = g_hash_table_size(device_table); + crm_info("Device '%s' not found (%u active device%s)", id, ndevices, + pcmk__plural_s(ndevices)); return; } if (from_cib) { device->cib_registered = FALSE; } else { device->verified = FALSE; device->api_registered = FALSE; } if (!device->cib_registered && !device->api_registered) { - g_hash_table_remove(device_list, id); - ndevices = g_hash_table_size(device_list); - crm_info("Removed '%s' from device list (%d active device%s)", + g_hash_table_remove(device_table, id); + ndevices = g_hash_table_size(device_table); + crm_info("Removed '%s' from device list (%u active device%s)", id, ndevices, pcmk__plural_s(ndevices)); } else { - crm_trace("Not removing '%s' from device list (%d active) because " - "still registered via:%s%s", - id, g_hash_table_size(device_list), - (device->cib_registered? " cib" : ""), - (device->api_registered? " api" : "")); + // Exactly one is true at this point + crm_trace("Not removing '%s' from device list (%u active) because " + "still registered via %s", + id, g_hash_table_size(device_table), + (device->cib_registered? "CIB" : "API")); } } /*! * \internal * \brief Return the number of stonith levels registered for a node * * \param[in] tp Node's topology table entry * * \return Number of non-NULL levels in topology entry * \note This function is used only for log messages. */ static int count_active_levels(const stonith_topology_t *tp) { int lpc = 0; int count = 0; for (lpc = 0; lpc < ST__LEVEL_COUNT; lpc++) { if (tp->levels[lpc] != NULL) { count++; } } return count; } static void free_topology_entry(gpointer data) { stonith_topology_t *tp = data; int lpc = 0; for (lpc = 0; lpc < ST__LEVEL_COUNT; lpc++) { if (tp->levels[lpc] != NULL) { g_list_free_full(tp->levels[lpc], free); } } free(tp->target); free(tp->target_value); free(tp->target_pattern); free(tp->target_attribute); free(tp); } void free_topology_list(void) { if (topology != NULL) { g_hash_table_destroy(topology); topology = NULL; } } void init_topology_list(void) { if (topology == NULL) { topology = pcmk__strkey_table(NULL, free_topology_entry); } } char * stonith_level_key(const xmlNode *level, enum fenced_target_by mode) { if (mode == fenced_target_by_unknown) { mode = unpack_level_kind(level); } switch (mode) { case fenced_target_by_name: return crm_element_value_copy(level, PCMK_XA_TARGET); case fenced_target_by_pattern: return crm_element_value_copy(level, PCMK_XA_TARGET_PATTERN); case fenced_target_by_attribute: return crm_strdup_printf("%s=%s", crm_element_value(level, PCMK_XA_TARGET_ATTRIBUTE), crm_element_value(level, PCMK_XA_TARGET_VALUE)); default: return crm_strdup_printf("unknown-%s", pcmk__xe_id(level)); } } /*! * \internal * \brief Parse target identification from topology level XML * * \param[in] level Topology level XML to parse * * \return How to identify target of \p level */ static enum fenced_target_by unpack_level_kind(const xmlNode *level) { if (crm_element_value(level, PCMK_XA_TARGET) != NULL) { return fenced_target_by_name; } if (crm_element_value(level, PCMK_XA_TARGET_PATTERN) != NULL) { return fenced_target_by_pattern; } if ((crm_element_value(level, PCMK_XA_TARGET_ATTRIBUTE) != NULL) && (crm_element_value(level, PCMK_XA_TARGET_VALUE) != NULL)) { return fenced_target_by_attribute; } return fenced_target_by_unknown; } static stonith_key_value_t * parse_device_list(const char *devices) { int lpc = 0; int max = 0; int last = 0; stonith_key_value_t *output = NULL; if (devices == NULL) { return output; } max = strlen(devices); for (lpc = 0; lpc <= max; lpc++) { if (devices[lpc] == ',' || devices[lpc] == 0) { char *line = strndup(devices + last, lpc - last); output = stonith_key_value_add(output, NULL, line); free(line); last = lpc + 1; } } return output; } /*! * \internal * \brief Unpack essential information from topology request XML * * \param[in] xml Request XML to search * \param[out] mode If not NULL, where to store level kind * \param[out] target If not NULL, where to store representation of target * \param[out] id If not NULL, where to store level number * \param[out] desc If not NULL, where to store log-friendly level description * * \return Topology level XML from within \p xml, or NULL if not found * \note The caller is responsible for freeing \p *target and \p *desc if set. */ static xmlNode * unpack_level_request(xmlNode *xml, enum fenced_target_by *mode, char **target, int *id, char **desc) { enum fenced_target_by local_mode = fenced_target_by_unknown; char *local_target = NULL; int local_id = 0; /* The level element can be the top element or lower. If top level, don't * search by xpath, because it might give multiple hits if the XML is the * CIB. */ if ((xml != NULL) && !pcmk__xe_is(xml, PCMK_XE_FENCING_LEVEL)) { xml = pcmk__xpath_find_one(xml->doc, "//" PCMK_XE_FENCING_LEVEL, LOG_WARNING); } if (xml == NULL) { if (desc != NULL) { *desc = crm_strdup_printf("missing"); } } else { local_mode = unpack_level_kind(xml); local_target = stonith_level_key(xml, local_mode); crm_element_value_int(xml, PCMK_XA_INDEX, &local_id); if (desc != NULL) { *desc = crm_strdup_printf("%s[%d]", local_target, local_id); } } if (mode != NULL) { *mode = local_mode; } if (id != NULL) { *id = local_id; } if (target != NULL) { *target = local_target; } else { free(local_target); } return xml; } /*! * \internal * \brief Register a fencing topology level for a target * * Given an XML request specifying the target name, level index, and device IDs * for the level, this will create an entry for the target in the global topology * table if one does not already exist, then append the specified device IDs to * the entry's device list for the specified level. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of registration */ void fenced_register_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = 0; xmlNode *level; enum fenced_target_by mode; char *target; stonith_topology_t *tp; stonith_key_value_t *dIter = NULL; stonith_key_value_t *devices = NULL; CRM_CHECK((msg != NULL) && (result != NULL), return); level = unpack_level_request(msg, &mode, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } // Ensure an ID was given (even the client API adds an ID) if (pcmk__str_empty(pcmk__xe_id(level))) { crm_warn("Ignoring registration for topology level without ID"); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Topology level is invalid without ID"); return; } // Ensure a valid target was specified if (mode == fenced_target_by_unknown) { crm_warn("Ignoring registration for topology level '%s' " "without valid target", pcmk__xe_id(level)); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid target for topology level '%s'", pcmk__xe_id(level)); return; } // Ensure level ID is in allowed range if ((id < ST__LEVEL_MIN) || (id > ST__LEVEL_MAX)) { crm_warn("Ignoring topology registration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level '%s'", pcmk__s(crm_element_value(level, PCMK_XA_INDEX), ""), pcmk__xe_id(level)); return; } /* Find or create topology table entry */ tp = g_hash_table_lookup(topology, target); if (tp == NULL) { tp = pcmk__assert_alloc(1, sizeof(stonith_topology_t)); tp->kind = mode; tp->target = target; tp->target_value = crm_element_value_copy(level, PCMK_XA_TARGET_VALUE); tp->target_pattern = crm_element_value_copy(level, PCMK_XA_TARGET_PATTERN); tp->target_attribute = crm_element_value_copy(level, PCMK_XA_TARGET_ATTRIBUTE); g_hash_table_replace(topology, tp->target, tp); crm_trace("Added %s (%d) to the topology (%d active entries)", target, (int) mode, g_hash_table_size(topology)); } else { free(target); } if (tp->levels[id] != NULL) { crm_info("Adding to the existing %s[%d] topology entry", tp->target, id); } devices = parse_device_list(crm_element_value(level, PCMK_XA_DEVICES)); for (dIter = devices; dIter; dIter = dIter->next) { const char *device = dIter->value; crm_trace("Adding device '%s' for %s[%d]", device, tp->target, id); tp->levels[id] = g_list_append(tp->levels[id], pcmk__str_copy(device)); } stonith_key_value_freeall(devices, 1, 1); { int nlevels = count_active_levels(tp); crm_info("Target %s has %d active fencing level%s", tp->target, nlevels, pcmk__plural_s(nlevels)); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } /*! * \internal * \brief Unregister a fencing topology level for a target * * Given an XML request specifying the target name and level index (or 0 for all * levels), this will remove any corresponding entry for the target from the * global topology table. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of unregistration */ void fenced_unregister_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = -1; stonith_topology_t *tp; char *target; xmlNode *level = NULL; CRM_CHECK(result != NULL, return); level = unpack_level_request(msg, NULL, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } // Ensure level ID is in allowed range if ((id < 0) || (id >= ST__LEVEL_COUNT)) { crm_warn("Ignoring topology unregistration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level %s", pcmk__s(crm_element_value(level, PCMK_XA_INDEX), ""), // Client API doesn't add ID to unregistration XML pcmk__s(pcmk__xe_id(level), "")); return; } tp = g_hash_table_lookup(topology, target); if (tp == NULL) { guint nentries = g_hash_table_size(topology); crm_info("No fencing topology found for %s (%d active %s)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (id == 0 && g_hash_table_remove(topology, target)) { guint nentries = g_hash_table_size(topology); crm_info("Removed all fencing topology entries related to %s " "(%d active %s remaining)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (tp->levels[id] != NULL) { guint nlevels; g_list_free_full(tp->levels[id], free); tp->levels[id] = NULL; nlevels = count_active_levels(tp); crm_info("Removed level %d from fencing topology for %s " "(%d active level%s remaining)", id, target, nlevels, pcmk__plural_s(nlevels)); } free(target); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } static char * list_to_string(GList *list, const char *delim, gboolean terminate_with_delim) { int max = g_list_length(list); size_t delim_len = delim?strlen(delim):0; size_t alloc_size = 1 + (max?((max-1+(terminate_with_delim?1:0))*delim_len):0); char *rv; GList *gIter; char *pos = NULL; const char *lead_delim = ""; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; alloc_size += strlen(value); } rv = pcmk__assert_alloc(alloc_size, sizeof(char)); pos = rv; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; pos = &pos[sprintf(pos, "%s%s", lead_delim, value)]; lead_delim = delim; } if (max && terminate_with_delim) { sprintf(pos, "%s", delim); } return rv; } /*! * \internal * \brief Execute a fence agent action directly (and asynchronously) * * Handle a STONITH_OP_EXEC API message by scheduling a requested agent action * directly on a specified device. Only list, monitor, and status actions are * expected to use this call, though it should work with any agent command. * * \param[in] msg Request XML specifying action * \param[out] result Where to store result of action * * \note If the action is monitor, the device must be registered via the API * (CIB registration is not sufficient), because monitor should not be * possible unless the device is "started" (API registered). */ static void execute_agent_action(xmlNode *msg, pcmk__action_result_t *result) { xmlNode *dev = pcmk__xpath_find_one(msg->doc, "//" PCMK__XE_ST_DEVICE_ID, LOG_ERR); xmlNode *op = pcmk__xpath_find_one(msg->doc, "//*[@" PCMK__XA_ST_DEVICE_ACTION "]", LOG_ERR); const char *id = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); const char *action = crm_element_value(op, PCMK__XA_ST_DEVICE_ACTION); async_command_t *cmd = NULL; - stonith_device_t *device = NULL; + fenced_device_t *device = NULL; if ((id == NULL) || (action == NULL)) { crm_info("Malformed API action request: device %s, action %s", (id? id : "not specified"), (action? action : "not specified")); fenced_set_protocol_error(result); return; } if (pcmk__str_eq(id, STONITH_WATCHDOG_ID, pcmk__str_none)) { // Watchdog agent actions are implemented internally if (stonith_watchdog_timeout_ms <= 0) { pcmk__set_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Watchdog fence device not configured"); return; } else if (pcmk__str_eq(action, PCMK_ACTION_LIST, pcmk__str_none)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_result_output(result, list_to_string(stonith_watchdog_targets, "\n", TRUE), NULL); return; } else if (pcmk__str_eq(action, PCMK_ACTION_MONITOR, pcmk__str_none)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return; } } - device = g_hash_table_lookup(device_list, id); + device = g_hash_table_lookup(device_table, id); if (device == NULL) { crm_info("Ignoring API '%s' action request because device %s not found", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not found", id); return; } else if (!device->api_registered && (strcmp(action, PCMK_ACTION_MONITOR) == 0)) { // Monitors may run only on "started" (API-registered) devices crm_info("Ignoring API '%s' action request because device %s not active", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not active", id); return; } cmd = create_async_command(msg); if (cmd == NULL) { crm_log_xml_warn(msg, "invalid"); fenced_set_protocol_error(result); return; } schedule_stonith_command(cmd, device); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence) { search->replies_received++; if (can_fence && device) { if (search->support_action_only != st_device_supports_none) { - stonith_device_t *dev = g_hash_table_lookup(device_list, device); + fenced_device_t *dev = g_hash_table_lookup(device_table, device); if (dev && !pcmk_is_set(dev->flags, search->support_action_only)) { return; } } search->capable = g_list_append(search->capable, pcmk__str_copy(device)); } if (search->replies_needed == search->replies_received) { guint ndevices = g_list_length(search->capable); crm_debug("Search found %d device%s that can perform '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); search->callback(search->capable, search->user_data); free(search->host); free(search->action); free(search); } } /*! * \internal * \brief Check whether the local host is allowed to execute a fencing action * * \param[in] device Fence device to check * \param[in] action Fence action to check * \param[in] target Hostname of fence target * \param[in] allow_self Whether self-fencing is allowed for this operation * * \return TRUE if local host is allowed to execute action, FALSE otherwise */ static gboolean -localhost_is_eligible(const stonith_device_t *device, const char *action, +localhost_is_eligible(const fenced_device_t *device, const char *action, const char *target, gboolean allow_self) { gboolean localhost_is_target = pcmk__str_eq(target, fenced_get_local_node(), pcmk__str_casei); if ((device != NULL) && (action != NULL) && (device->on_target_actions != NULL) && (strstr((const char*) device->on_target_actions->str, action) != NULL)) { if (!localhost_is_target) { crm_trace("Operation '%s' using %s can only be executed for local " "host, not %s", action, device->id, target); return FALSE; } } else if (localhost_is_target && !allow_self) { crm_trace("'%s' operation does not support self-fencing", action); return FALSE; } return TRUE; } /*! * \internal * \brief Check if local node is allowed to execute (possibly remapped) action * * \param[in] device Fence device to check * \param[in] action Fence action to check * \param[in] target Node name of fence target * \param[in] allow_self Whether self-fencing is allowed for this operation * * \return true if local node is allowed to execute \p action or any actions it * might be remapped to, otherwise false */ static bool -localhost_is_eligible_with_remap(const stonith_device_t *device, +localhost_is_eligible_with_remap(const fenced_device_t *device, const char *action, const char *target, gboolean allow_self) { // Check exact action if (localhost_is_eligible(device, action, target, allow_self)) { return true; } // Check potential remaps if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none)) { /* "reboot" might get remapped to "off" then "on", so even if reboot is * disallowed, return true if either of those is allowed. We'll report * the disallowed actions with the results. We never allow self-fencing * for remapped "on" actions because the target is off at that point. */ if (localhost_is_eligible(device, PCMK_ACTION_OFF, target, allow_self) || localhost_is_eligible(device, PCMK_ACTION_ON, target, FALSE)) { return true; } } return false; } /*! * \internal * \brief Check whether we can use a device's cached target list * * \param[in] dev Fencing device to check * * \return \c true if \p dev cached its targets less than a minute ago, * otherwise \c false */ static inline bool -can_use_target_cache(const stonith_device_t *dev) +can_use_target_cache(const fenced_device_t *dev) { return (dev->targets != NULL) && (time(NULL) < (dev->targets_age + 60)); } static void -can_fence_host_with_device(stonith_device_t *dev, +can_fence_host_with_device(fenced_device_t *dev, struct device_search_s *search) { gboolean can = FALSE; const char *check_type = "Internal bug"; const char *target = NULL; const char *alias = NULL; const char *dev_id = "Unspecified device"; const char *action = (search == NULL)? NULL : search->action; CRM_CHECK((dev != NULL) && (action != NULL), goto search_report_results); if (dev->id != NULL) { dev_id = dev->id; } target = search->host; if (target == NULL) { can = TRUE; check_type = "No target"; goto search_report_results; } /* Answer immediately if the device does not support the action * or the local node is not allowed to perform it */ if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none) && !pcmk_is_set(dev->flags, st_device_supports_on)) { check_type = "Agent does not support 'on'"; goto search_report_results; } else if (!localhost_is_eligible_with_remap(dev, action, target, search->allow_self)) { check_type = "This node is not allowed to execute action"; goto search_report_results; } // Check eligibility as specified by pcmk_host_check check_type = target_list_type(dev); alias = g_hash_table_lookup(dev->aliases, target); if (pcmk__str_eq(check_type, PCMK_VALUE_NONE, pcmk__str_casei)) { can = TRUE; } else if (pcmk__str_eq(check_type, PCMK_VALUE_STATIC_LIST, pcmk__str_casei)) { if (pcmk__str_in_list(target, dev->targets, pcmk__str_casei)) { can = TRUE; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP) && g_hash_table_lookup(dev->aliases, target)) { can = TRUE; } } else if (pcmk__str_eq(check_type, PCMK_VALUE_DYNAMIC_LIST, pcmk__str_casei)) { if (!can_use_target_cache(dev)) { int device_timeout = get_action_timeout(dev, PCMK_ACTION_LIST, search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_list_timeout (%ds) parameter of %s " "is larger than " PCMK_OPT_STONITH_TIMEOUT " (%ds), timeout may occur", device_timeout, dev_id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev_id, target, action); schedule_internal_command(__func__, dev, PCMK_ACTION_LIST, NULL, search->per_device_timeout, search, dynamic_list_search_cb); /* we'll respond to this search request async in the cb */ return; } if (pcmk__str_in_list(((alias == NULL)? target : alias), dev->targets, pcmk__str_casei)) { can = TRUE; } } else if (pcmk__str_eq(check_type, PCMK_VALUE_STATUS, pcmk__str_casei)) { int device_timeout = get_action_timeout(dev, check_type, search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_status_timeout (%ds) parameter of %s is " "larger than " PCMK_OPT_STONITH_TIMEOUT " (%ds), " "timeout may occur", device_timeout, dev_id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev_id, target, action); schedule_internal_command(__func__, dev, PCMK_ACTION_STATUS, target, search->per_device_timeout, search, status_search_cb); /* we'll respond to this search request async in the cb */ return; } else { crm_err("Invalid value for " PCMK_STONITH_HOST_CHECK ": %s", check_type); check_type = "Invalid " PCMK_STONITH_HOST_CHECK; } search_report_results: crm_info("%s is%s eligible to fence (%s) %s%s%s%s: %s", dev_id, (can? "" : " not"), pcmk__s(action, "unspecified action"), pcmk__s(target, "unspecified target"), (alias == NULL)? "" : " (as '", pcmk__s(alias, ""), (alias == NULL)? "" : "')", check_type); search_devices_record_result(search, ((dev == NULL)? NULL : dev_id), can); } static void search_devices(gpointer key, gpointer value, gpointer user_data) { - stonith_device_t *dev = value; + fenced_device_t *dev = value; struct device_search_s *search = user_data; can_fence_host_with_device(dev, search); } #define DEFAULT_QUERY_TIMEOUT 20 static void get_capable_devices(const char *host, const char *action, int timeout, bool allow_self, void *user_data, void (*callback) (GList * devices, void *user_data), uint32_t support_action_only) { struct device_search_s *search; - guint ndevices = g_hash_table_size(device_list); + guint ndevices = g_hash_table_size(device_table); if (ndevices == 0) { callback(NULL, user_data); return; } search = pcmk__assert_alloc(1, sizeof(struct device_search_s)); search->host = pcmk__str_copy(host); search->action = pcmk__str_copy(action); search->per_device_timeout = timeout; search->allow_self = allow_self; search->callback = callback; search->user_data = user_data; search->support_action_only = support_action_only; /* We are guaranteed this many replies, even if a device is * unregistered while the search is in progress. */ search->replies_needed = ndevices; crm_debug("Searching %d device%s to see which can execute '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); - g_hash_table_foreach(device_list, search_devices, search); + fenced_foreach_device(search_devices, search); } struct st_query_data { xmlNode *reply; char *remote_peer; char *client_id; char *target; char *action; int call_options; }; /*! * \internal * \brief Add action-specific attributes to query reply XML * * \param[in,out] xml XML to add attributes to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target */ static void add_action_specific_attributes(xmlNode *xml, const char *action, - const stonith_device_t *device, + const fenced_device_t *device, const char *target) { int action_specific_timeout; int delay_max; int delay_base; CRM_CHECK(xml && action && device, return); // PCMK__XA_ST_REQUIRED is currently used only for unfencing if (is_action_required(action, device)) { crm_trace("Action '%s' is required using %s", action, device->id); crm_xml_add_int(xml, PCMK__XA_ST_REQUIRED, 1); } // pcmk__timeout if configured action_specific_timeout = get_action_timeout(device, action, 0); if (action_specific_timeout) { crm_trace("Action '%s' has timeout %ds using %s", action, action_specific_timeout, device->id); crm_xml_add_int(xml, PCMK__XA_ST_ACTION_TIMEOUT, action_specific_timeout); } delay_max = get_action_delay_max(device, action); if (delay_max > 0) { crm_trace("Action '%s' has maximum random delay %ds using %s", action, delay_max, device->id); crm_xml_add_int(xml, PCMK__XA_ST_DELAY_MAX, delay_max); } delay_base = get_action_delay_base(device, action, target); if (delay_base > 0) { crm_xml_add_int(xml, PCMK__XA_ST_DELAY_BASE, delay_base); } if ((delay_max > 0) && (delay_base == 0)) { crm_trace("Action '%s' has maximum random delay %ds using %s", action, delay_max, device->id); } else if ((delay_max == 0) && (delay_base > 0)) { crm_trace("Action '%s' has a static delay of %ds using %s", action, delay_base, device->id); } else if ((delay_max > 0) && (delay_base > 0)) { crm_trace("Action '%s' has a minimum delay of %ds and a randomly chosen " "maximum delay of %ds using %s", action, delay_base, delay_max, device->id); } } /*! * \internal * \brief Add "disallowed" attribute to query reply XML if appropriate * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_self Whether self-fencing is allowed */ static void -add_disallowed(xmlNode *xml, const char *action, const stonith_device_t *device, +add_disallowed(xmlNode *xml, const char *action, const fenced_device_t *device, const char *target, gboolean allow_self) { if (!localhost_is_eligible(device, action, target, allow_self)) { crm_trace("Action '%s' using %s is disallowed for local host", action, device->id); pcmk__xe_set_bool_attr(xml, PCMK__XA_ST_ACTION_DISALLOWED, true); } } /*! * \internal * \brief Add child element with action-specific values to query reply XML * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_self Whether self-fencing is allowed */ static void add_action_reply(xmlNode *xml, const char *action, - const stonith_device_t *device, const char *target, + const fenced_device_t *device, const char *target, gboolean allow_self) { xmlNode *child = pcmk__xe_create(xml, PCMK__XE_ST_DEVICE_ACTION); crm_xml_add(child, PCMK_XA_ID, action); add_action_specific_attributes(child, action, device, target); add_disallowed(child, action, device, target, allow_self); } /*! * \internal * \brief Send a reply to a CPG peer or IPC client * * \param[in] reply XML reply to send * \param[in] call_options Send synchronously if st_opt_sync_call is set * \param[in] remote_peer If not NULL, name of peer node to send CPG reply * \param[in,out] client If not NULL, client to send IPC reply */ static void stonith_send_reply(const xmlNode *reply, int call_options, const char *remote_peer, pcmk__client_t *client) { CRM_CHECK((reply != NULL) && ((remote_peer != NULL) || (client != NULL)), return); if (remote_peer == NULL) { do_local_reply(reply, client, call_options); } else { const pcmk__node_status_t *node = pcmk__get_node(0, remote_peer, NULL, pcmk__node_search_cluster_member); pcmk__cluster_send_message(node, pcmk_ipc_fenced, reply); } } static void stonith_query_capable_device_cb(GList * devices, void *user_data) { struct st_query_data *query = user_data; int available_devices = 0; xmlNode *wrapper = NULL; xmlNode *list = NULL; GList *lpc = NULL; pcmk__client_t *client = NULL; if (query->client_id != NULL) { client = pcmk__find_client_by_id(query->client_id); if ((client == NULL) && (query->remote_peer == NULL)) { crm_trace("Skipping reply to %s: no longer a client", query->client_id); goto done; } } // Pack the results into XML wrapper = pcmk__xe_create(query->reply, PCMK__XE_ST_CALLDATA); list = pcmk__xe_create(wrapper, __func__); crm_xml_add(list, PCMK__XA_ST_TARGET, query->target); for (lpc = devices; lpc != NULL; lpc = lpc->next) { - stonith_device_t *device = g_hash_table_lookup(device_list, lpc->data); + fenced_device_t *device = g_hash_table_lookup(device_table, lpc->data); const char *action = query->action; xmlNode *dev = NULL; if (!device) { /* It is possible the device got unregistered while * determining who can fence the target */ continue; } available_devices++; dev = pcmk__xe_create(list, PCMK__XE_ST_DEVICE_ID); crm_xml_add(dev, PCMK_XA_ID, device->id); crm_xml_add(dev, PCMK__XA_NAMESPACE, device->namespace); crm_xml_add(dev, PCMK_XA_AGENT, device->agent); // Has had successful monitor, list, or status on this node crm_xml_add_int(dev, PCMK__XA_ST_MONITOR_VERIFIED, device->verified); crm_xml_add_int(dev, PCMK__XA_ST_DEVICE_SUPPORT_FLAGS, device->flags); /* If the originating fencer wants to reboot the node, and we have a * capable device that doesn't support "reboot", remap to "off" instead. */ if (!pcmk_is_set(device->flags, st_device_supports_reboot) && pcmk__str_eq(query->action, PCMK_ACTION_REBOOT, pcmk__str_none)) { crm_trace("%s doesn't support reboot, using values for off instead", device->id); action = PCMK_ACTION_OFF; } /* Add action-specific values if available */ add_action_specific_attributes(dev, action, device, query->target); if (pcmk__str_eq(query->action, PCMK_ACTION_REBOOT, pcmk__str_none)) { /* A "reboot" *might* get remapped to "off" then "on", so after * sending the "reboot"-specific values in the main element, we add * sub-elements for "off" and "on" values. * * We short-circuited earlier if "reboot", "off" and "on" are all * disallowed for the local host. However if only one or two are * disallowed, we send back the results and mark which ones are * disallowed. If "reboot" is disallowed, this might cause problems * with older fencer versions, which won't check for it. Older * versions will ignore "off" and "on", so they are not a problem. */ add_disallowed(dev, action, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_self_fencing)); add_action_reply(dev, PCMK_ACTION_OFF, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_self_fencing)); add_action_reply(dev, PCMK_ACTION_ON, device, query->target, FALSE); } /* A query without a target wants device parameters */ if (query->target == NULL) { xmlNode *attrs = pcmk__xe_create(dev, PCMK__XE_ATTRIBUTES); g_hash_table_foreach(device->params, hash2field, attrs); } } crm_xml_add_int(list, PCMK__XA_ST_AVAILABLE_DEVICES, available_devices); if (query->target) { crm_debug("Found %d matching device%s for target '%s'", available_devices, pcmk__plural_s(available_devices), query->target); } else { crm_debug("%d device%s installed", available_devices, pcmk__plural_s(available_devices)); } crm_log_xml_trace(list, "query-result"); stonith_send_reply(query->reply, query->call_options, query->remote_peer, client); done: pcmk__xml_free(query->reply); free(query->remote_peer); free(query->client_id); free(query->target); free(query->action); free(query); g_list_free_full(devices, free); } /*! * \internal * \brief Log the result of an asynchronous command * * \param[in] cmd Command the result is for * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] next Alternate device that will be tried if command failed * \param[in] op_merged Whether this command was merged with an earlier one */ static void log_async_result(const async_command_t *cmd, const pcmk__action_result_t *result, int pid, const char *next, bool op_merged) { int log_level = LOG_ERR; int output_log_level = LOG_NEVER; guint devices_remaining = g_list_length(cmd->next_device_iter); GString *msg = g_string_sized_new(80); // Reasonable starting size // Choose log levels appropriately if we have a result if (pcmk__result_ok(result)) { log_level = (cmd->target == NULL)? LOG_DEBUG : LOG_NOTICE; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, PCMK_ACTION_METADATA, pcmk__str_none)) { output_log_level = LOG_DEBUG; } next = NULL; } else { log_level = (cmd->target == NULL)? LOG_NOTICE : LOG_ERR; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, PCMK_ACTION_METADATA, pcmk__str_none)) { output_log_level = LOG_WARNING; } } // Build the log message piece by piece pcmk__g_strcat(msg, "Operation '", cmd->action, "' ", NULL); if (pid != 0) { g_string_append_printf(msg, "[%d] ", pid); } if (cmd->target != NULL) { pcmk__g_strcat(msg, "targeting ", cmd->target, " ", NULL); } if (cmd->device != NULL) { pcmk__g_strcat(msg, "using ", cmd->device, " ", NULL); } // Add exit status or execution status as appropriate if (result->execution_status == PCMK_EXEC_DONE) { g_string_append_printf(msg, "returned %d", result->exit_status); } else { pcmk__g_strcat(msg, "could not be executed: ", pcmk_exec_status_str(result->execution_status), NULL); } // Add exit reason and next device if appropriate if (result->exit_reason != NULL) { pcmk__g_strcat(msg, " (", result->exit_reason, ")", NULL); } if (next != NULL) { pcmk__g_strcat(msg, ", retrying with ", next, NULL); } if (devices_remaining > 0) { g_string_append_printf(msg, " (%u device%s remaining)", (unsigned int) devices_remaining, pcmk__plural_s(devices_remaining)); } g_string_append_printf(msg, " " QB_XS " %scall %d from %s", (op_merged? "merged " : ""), cmd->id, cmd->client_name); // Log the result do_crm_log(log_level, "%s", msg->str); g_string_free(msg, TRUE); // Log the output (which may have multiple lines), if appropriate if (output_log_level != LOG_NEVER) { char *prefix = crm_strdup_printf("%s[%d]", cmd->device, pid); crm_log_output(output_log_level, prefix, result->action_stdout); free(prefix); } } /*! * \internal * \brief Reply to requester after asynchronous command completion * * \param[in] cmd Command that completed * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] merged If true, command was merged with another, not executed */ static void send_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result, int pid, bool merged) { xmlNode *reply = NULL; pcmk__client_t *client = NULL; CRM_CHECK((cmd != NULL) && (result != NULL), return); log_async_result(cmd, result, pid, NULL, merged); if (cmd->client != NULL) { client = pcmk__find_client_by_id(cmd->client); if ((client == NULL) && (cmd->origin == NULL)) { crm_trace("Skipping reply to %s: no longer a client", cmd->client); return; } } reply = construct_async_reply(cmd, result); if (merged) { pcmk__xe_set_bool_attr(reply, PCMK__XA_ST_OP_MERGED, true); } if (pcmk__is_fencing_action(cmd->action) && pcmk__str_eq(cmd->origin, cmd->target, pcmk__str_casei)) { /* The target was also the originator, so broadcast the result on its * behalf (since it will be unable to). */ crm_trace("Broadcast '%s' result for %s (target was also originator)", cmd->action, cmd->target); crm_xml_add(reply, PCMK__XA_SUBT, PCMK__VALUE_BROADCAST); crm_xml_add(reply, PCMK__XA_ST_OP, STONITH_OP_NOTIFY); pcmk__cluster_send_message(NULL, pcmk_ipc_fenced, reply); } else { // Reply only to the originator stonith_send_reply(reply, cmd->options, cmd->origin, client); } crm_log_xml_trace(reply, "Reply"); pcmk__xml_free(reply); } static void cancel_stonith_command(async_command_t * cmd) { - stonith_device_t *device = cmd_device(cmd); + fenced_device_t *device = cmd_device(cmd); if (device) { crm_trace("Cancel scheduled '%s' action using %s", cmd->action, device->id); device->pending_ops = g_list_remove(device->pending_ops, cmd); } } /*! * \internal * \brief Cancel and reply to any duplicates of a just-completed operation * * Check whether any fencing operations are scheduled to do the same thing as * one that just succeeded. If so, rather than performing the same operation * twice, return the result of this operation for all matching pending commands. * * \param[in,out] cmd Fencing operation that just succeeded * \param[in] result Result of \p cmd * \param[in] pid If nonzero, process ID of agent invocation (for logs) * * \note Duplicate merging will do the right thing for either type of remapped * reboot. If the executing fencer remapped an unsupported reboot to off, * then cmd->action will be "reboot" and will be merged with any other * reboot requests. If the originating fencer remapped a topology reboot * to off then on, we will get here once with cmd->action "off" and once * with "on", and they will be merged separately with similar requests. */ static void reply_to_duplicates(async_command_t *cmd, const pcmk__action_result_t *result, int pid) { GList *next = NULL; for (GList *iter = cmd_list; iter != NULL; iter = next) { async_command_t *cmd_other = iter->data; next = iter->next; // We might delete this entry, so grab next now if (cmd == cmd_other) { continue; } /* A pending operation matches if: * 1. The client connections are different. * 2. The target is the same. * 3. The fencing action is the same. * 4. The device scheduled to execute the action is the same. */ if (pcmk__str_eq(cmd->client, cmd_other->client, pcmk__str_casei) || !pcmk__str_eq(cmd->target, cmd_other->target, pcmk__str_casei) || !pcmk__str_eq(cmd->action, cmd_other->action, pcmk__str_none) || !pcmk__str_eq(cmd->device, cmd_other->device, pcmk__str_casei)) { continue; } crm_notice("Merging fencing action '%s'%s%s originating from " "client %s with identical fencing request from client %s", cmd_other->action, (cmd_other->target == NULL)? "" : " targeting ", pcmk__s(cmd_other->target, ""), cmd_other->client_name, cmd->client_name); // Stop tracking the duplicate, send its result, and cancel it cmd_list = g_list_remove_link(cmd_list, iter); send_async_reply(cmd_other, result, pid, true); cancel_stonith_command(cmd_other); free_async_command(cmd_other); g_list_free_1(iter); } } /*! * \internal * \brief Return the next required device (if any) for an operation * * \param[in,out] cmd Fencing operation that just succeeded * * \return Next device required for action if any, otherwise NULL */ -static stonith_device_t * +static fenced_device_t * next_required_device(async_command_t *cmd) { for (GList *iter = cmd->next_device_iter; iter != NULL; iter = iter->next) { - stonith_device_t *next_device = g_hash_table_lookup(device_list, - iter->data); + fenced_device_t *next_device = g_hash_table_lookup(device_table, + iter->data); if (is_action_required(cmd->action, next_device)) { /* This is only called for successful actions, so it's OK to skip * non-required devices. */ cmd->next_device_iter = iter->next; return next_device; } } return NULL; } static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; - stonith_device_t *device = NULL; - stonith_device_t *next_device = NULL; + fenced_device_t *device = NULL; + fenced_device_t *next_device = NULL; CRM_CHECK(cmd != NULL, return); device = cmd_device(cmd); cmd->active_on = NULL; /* The device is ready to do something else now */ if (device) { if (!device->verified && pcmk__result_ok(result) && pcmk__strcase_any_of(cmd->action, PCMK_ACTION_LIST, PCMK_ACTION_MONITOR, PCMK_ACTION_STATUS, NULL)) { device->verified = TRUE; } mainloop_set_trigger(device->work); } if (pcmk__result_ok(result)) { next_device = next_required_device(cmd); } else if ((cmd->next_device_iter != NULL) && !is_action_required(cmd->action, device)) { /* if this device didn't work out, see if there are any others we can try. * if the failed device was 'required', we can't pick another device. */ - next_device = g_hash_table_lookup(device_list, + next_device = g_hash_table_lookup(device_table, cmd->next_device_iter->data); cmd->next_device_iter = cmd->next_device_iter->next; } if (next_device == NULL) { send_async_reply(cmd, result, pid, false); if (pcmk__result_ok(result)) { reply_to_duplicates(cmd, result, pid); } free_async_command(cmd); } else { // This operation requires more fencing log_async_result(cmd, result, pid, next_device->id, false); schedule_stonith_command(cmd, next_device); } } -static gint -sort_device_priority(gconstpointer a, gconstpointer b) -{ - const stonith_device_t *dev_a = a; - const stonith_device_t *dev_b = b; - - if (dev_a->priority > dev_b->priority) { - return -1; - } else if (dev_a->priority < dev_b->priority) { - return 1; - } - return 0; -} - static void stonith_fence_get_devices_cb(GList * devices, void *user_data) { async_command_t *cmd = user_data; - stonith_device_t *device = NULL; + fenced_device_t *device = NULL; guint ndevices = g_list_length(devices); crm_info("Found %d matching device%s for target '%s'", ndevices, pcmk__plural_s(ndevices), cmd->target); if (devices != NULL) { - /* Order based on priority */ - devices = g_list_sort(devices, sort_device_priority); - device = g_hash_table_lookup(device_list, devices->data); + device = g_hash_table_lookup(device_table, devices->data); } if (device == NULL) { // No device found pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__format_result(&result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "No device configured for target '%s'", cmd->target); send_async_reply(cmd, &result, 0, false); pcmk__reset_result(&result); free_async_command(cmd); g_list_free_full(devices, free); - } else { // Device found, schedule it for fencing + } else { + /* Device found. Schedule a fencing command for it. + * + * Assign devices to device_list so that it will be freed with cmd. + */ cmd->device_list = devices; cmd->next_device_iter = devices->next; schedule_stonith_command(cmd, device); } } /*! * \internal * \brief Execute a fence action via the local node * * \param[in] msg Fencing request * \param[out] result Where to store result of fence action */ static void fence_locally(xmlNode *msg, pcmk__action_result_t *result) { const char *device_id = NULL; - stonith_device_t *device = NULL; + fenced_device_t *device = NULL; async_command_t *cmd = NULL; xmlNode *dev = NULL; CRM_CHECK((msg != NULL) && (result != NULL), return); dev = pcmk__xpath_find_one(msg->doc, "//*[@" PCMK__XA_ST_TARGET "]", LOG_ERR); cmd = create_async_command(msg); if (cmd == NULL) { crm_log_xml_warn(msg, "invalid"); fenced_set_protocol_error(result); return; } device_id = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (device_id != NULL) { - device = g_hash_table_lookup(device_list, device_id); + device = g_hash_table_lookup(device_table, device_id); if (device == NULL) { crm_err("Requested device '%s' is not available", device_id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Requested device '%s' not found", device_id); return; } schedule_stonith_command(cmd, device); } else { const char *host = crm_element_value(dev, PCMK__XA_ST_TARGET); if (pcmk_is_set(cmd->options, st_opt_cs_nodeid)) { int nodeid = 0; pcmk__node_status_t *node = NULL; pcmk__scan_min_int(host, &nodeid, 0); node = pcmk__search_node_caches(nodeid, NULL, NULL, pcmk__node_search_any |pcmk__node_search_cluster_cib); if (node != NULL) { host = node->name; } } /* If we get to here, then self-fencing is implicitly allowed */ get_capable_devices(host, cmd->action, cmd->default_timeout, TRUE, cmd, stonith_fence_get_devices_cb, fenced_support_flag(cmd->action)); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } /*! * \internal * \brief Build an XML reply for a fencing operation * * \param[in] request Request that reply is for * \param[in] data If not NULL, add to reply as call data * \param[in] result Full result of fencing operation * * \return Newly created XML reply * \note The caller is responsible for freeing the result. * \note This has some overlap with construct_async_reply(), but that copies * values from an async_command_t, whereas this one copies them from the * request. */ xmlNode * fenced_construct_reply(const xmlNode *request, xmlNode *data, const pcmk__action_result_t *result) { xmlNode *reply = NULL; reply = pcmk__xe_create(NULL, PCMK__XE_ST_REPLY); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(reply, PCMK__XA_T, PCMK__VALUE_STONITH_NG); stonith__xe_set_result(reply, result); if (request == NULL) { /* Most likely, this is the result of a stonith operation that was * initiated before we came up. Unfortunately that means we lack enough * information to provide clients with a full result. * * @TODO Maybe synchronize this information at start-up? */ crm_warn("Missing request information for client notifications for " "operation with result '%s' (initiated before we came up?)", pcmk_exec_status_str(result->execution_status)); } else { const char *name = NULL; const char *value = NULL; // Attributes to copy from request to reply const char *names[] = { PCMK__XA_ST_OP, PCMK__XA_ST_CALLID, PCMK__XA_ST_CLIENTID, PCMK__XA_ST_CLIENTNAME, PCMK__XA_ST_REMOTE_OP, PCMK__XA_ST_CALLOPT, }; for (int lpc = 0; lpc < PCMK__NELEM(names); lpc++) { name = names[lpc]; value = crm_element_value(request, name); crm_xml_add(reply, name, value); } if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(reply, PCMK__XE_ST_CALLDATA); pcmk__xml_copy(wrapper, data); } } return reply; } /*! * \internal * \brief Build an XML reply to an asynchronous fencing command * * \param[in] cmd Fencing command that reply is for * \param[in] result Command result */ static xmlNode * construct_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result) { xmlNode *reply = pcmk__xe_create(NULL, PCMK__XE_ST_REPLY); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(reply, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(reply, PCMK__XA_ST_OP, cmd->op); crm_xml_add(reply, PCMK__XA_ST_DEVICE_ID, cmd->device); crm_xml_add(reply, PCMK__XA_ST_REMOTE_OP, cmd->remote_op_id); crm_xml_add(reply, PCMK__XA_ST_CLIENTID, cmd->client); crm_xml_add(reply, PCMK__XA_ST_CLIENTNAME, cmd->client_name); crm_xml_add(reply, PCMK__XA_ST_TARGET, cmd->target); crm_xml_add(reply, PCMK__XA_ST_DEVICE_ACTION, cmd->op); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, cmd->origin); crm_xml_add_int(reply, PCMK__XA_ST_CALLID, cmd->id); crm_xml_add_int(reply, PCMK__XA_ST_CALLOPT, cmd->options); stonith__xe_set_result(reply, result); return reply; } bool fencing_peer_active(pcmk__node_status_t *peer) { return (peer != NULL) && (peer->name != NULL) && pcmk_is_set(peer->processes, crm_get_cluster_proc()); } void set_fencing_completed(remote_fencing_op_t *op) { struct timespec tv; qb_util_timespec_from_epoch_get(&tv); op->completed = tv.tv_sec; op->completed_nsec = tv.tv_nsec; } /*! * \internal * \brief Look for alternate node needed if local node shouldn't fence target * * \param[in] target Node that must be fenced * * \return Name of an alternate node that should fence \p target if any, * or NULL otherwise */ static const char * check_alternate_host(const char *target) { if (pcmk__str_eq(target, fenced_get_local_node(), pcmk__str_casei)) { GHashTableIter gIter; pcmk__node_status_t *entry = NULL; g_hash_table_iter_init(&gIter, pcmk__peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { if (fencing_peer_active(entry) && !pcmk__str_eq(entry->name, target, pcmk__str_casei)) { crm_notice("Forwarding self-fencing request to %s", entry->name); return entry->name; } } crm_warn("Will handle own fencing because no peer can"); } return NULL; } static void remove_relay_op(xmlNode * request) { xmlNode *dev = pcmk__xpath_find_one(request->doc, "//*[@" PCMK__XA_ST_DEVICE_ACTION "]", LOG_TRACE); const char *relay_op_id = NULL; const char *op_id = NULL; const char *client_name = NULL; const char *target = NULL; remote_fencing_op_t *relay_op = NULL; if (dev) { target = crm_element_value(dev, PCMK__XA_ST_TARGET); } relay_op_id = crm_element_value(request, PCMK__XA_ST_REMOTE_OP_RELAY); op_id = crm_element_value(request, PCMK__XA_ST_REMOTE_OP); client_name = crm_element_value(request, PCMK__XA_ST_CLIENTNAME); /* Delete RELAY operation. */ if ((relay_op_id != NULL) && (target != NULL) && pcmk__str_eq(target, fenced_get_local_node(), pcmk__str_casei)) { relay_op = g_hash_table_lookup(stonith_remote_op_list, relay_op_id); if (relay_op) { GHashTableIter iter; remote_fencing_op_t *list_op = NULL; g_hash_table_iter_init(&iter, stonith_remote_op_list); /* If the operation to be deleted is registered as a duplicate, delete the registration. */ while (g_hash_table_iter_next(&iter, NULL, (void **)&list_op)) { GList *dup_iter = NULL; if (list_op != relay_op) { for (dup_iter = list_op->duplicates; dup_iter != NULL; dup_iter = dup_iter->next) { remote_fencing_op_t *other = dup_iter->data; if (other == relay_op) { other->duplicates = g_list_remove(other->duplicates, relay_op); break; } } } } crm_debug("Deleting relay op %s ('%s'%s%s for %s), " "replaced by op %s ('%s'%s%s for %s)", relay_op->id, relay_op->action, (relay_op->target == NULL)? "" : " targeting ", pcmk__s(relay_op->target, ""), relay_op->client_name, op_id, relay_op->action, (target == NULL)? "" : " targeting ", pcmk__s(target, ""), client_name); g_hash_table_remove(stonith_remote_op_list, relay_op_id); } } } /*! * \internal * \brief Check whether an API request was sent by a privileged user * * API commands related to fencing configuration may be done only by privileged * IPC users (i.e. root or hacluster), because all other users should go through * the CIB to have ACLs applied. If no client was given, this is a peer request, * which is always allowed. * * \param[in] c IPC client that sent request (or NULL if sent by CPG peer) * \param[in] op Requested API operation (for logging only) * * \return true if sender is peer or privileged client, otherwise false */ static inline bool is_privileged(const pcmk__client_t *c, const char *op) { if ((c == NULL) || pcmk_is_set(c->flags, pcmk__client_privileged)) { return true; } else { crm_warn("Rejecting IPC request '%s' from unprivileged client %s", pcmk__s(op, ""), pcmk__client_name(c)); return false; } } // CRM_OP_REGISTER static xmlNode * handle_register_request(pcmk__request_t *request) { xmlNode *reply = pcmk__xe_create(NULL, "reply"); pcmk__assert(request->ipc_client != NULL); crm_xml_add(reply, PCMK__XA_ST_OP, CRM_OP_REGISTER); crm_xml_add(reply, PCMK__XA_ST_CLIENTID, request->ipc_client->id); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return reply; } // STONITH_OP_EXEC static xmlNode * handle_agent_request(pcmk__request_t *request) { execute_agent_action(request->xml, &request->result); if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_TIMEOUT_UPDATE static xmlNode * handle_update_timeout_request(pcmk__request_t *request) { const char *call_id = crm_element_value(request->xml, PCMK__XA_ST_CALLID); const char *client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); int op_timeout = 0; crm_element_value_int(request->xml, PCMK__XA_ST_TIMEOUT, &op_timeout); do_stonith_async_timeout_update(client_id, call_id, op_timeout); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } // STONITH_OP_QUERY static xmlNode * handle_query_request(pcmk__request_t *request) { int timeout = 0; xmlNode *dev = NULL; const char *action = NULL; const char *target = NULL; const char *client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); struct st_query_data *query = NULL; if (request->peer != NULL) { // Record it for the future notification create_remote_stonith_op(client_id, request->xml, TRUE); } /* Delete the DC node RELAY operation. */ remove_relay_op(request->xml); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); dev = pcmk__xpath_find_one(request->xml->doc, "//*[@" PCMK__XA_ST_DEVICE_ACTION "]", LOG_NEVER); if (dev != NULL) { const char *device = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (pcmk__str_eq(device, "manual_ack", pcmk__str_casei)) { return NULL; // No query or reply necessary } target = crm_element_value(dev, PCMK__XA_ST_TARGET); action = crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION); } crm_log_xml_trace(request->xml, "Query"); query = pcmk__assert_alloc(1, sizeof(struct st_query_data)); query->reply = fenced_construct_reply(request->xml, NULL, &request->result); query->remote_peer = pcmk__str_copy(request->peer); query->client_id = pcmk__str_copy(client_id); query->target = pcmk__str_copy(target); query->action = pcmk__str_copy(action); query->call_options = request->call_options; crm_element_value_int(request->xml, PCMK__XA_ST_TIMEOUT, &timeout); get_capable_devices(target, action, timeout, pcmk_is_set(query->call_options, st_opt_allow_self_fencing), query, stonith_query_capable_device_cb, st_device_supports_none); return NULL; } // STONITH_OP_NOTIFY static xmlNode * handle_notify_request(pcmk__request_t *request) { const char *flag_name = NULL; pcmk__assert(request->ipc_client != NULL); flag_name = crm_element_value(request->xml, PCMK__XA_ST_NOTIFY_ACTIVATE); if (flag_name != NULL) { crm_debug("Enabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__set_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } flag_name = crm_element_value(request->xml, PCMK__XA_ST_NOTIFY_DEACTIVATE); if (flag_name != NULL) { crm_debug("Disabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__clear_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return pcmk__ipc_create_ack(request->ipc_flags, PCMK__XE_ACK, NULL, CRM_EX_OK); } // STONITH_OP_RELAY static xmlNode * handle_relay_request(pcmk__request_t *request) { xmlNode *dev = pcmk__xpath_find_one(request->xml->doc, "//*[@" PCMK__XA_ST_TARGET "]", LOG_TRACE); crm_notice("Received forwarded fencing request from " "%s %s to fence (%s) peer %s", pcmk__request_origin_type(request), pcmk__request_origin(request), crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION), crm_element_value(dev, PCMK__XA_ST_TARGET)); if (initiate_remote_stonith_op(NULL, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); return fenced_construct_reply(request->xml, NULL, &request->result); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); return NULL; } // STONITH_OP_FENCE static xmlNode * handle_fence_request(pcmk__request_t *request) { if (request->peer != NULL) { fence_locally(request->xml, &request->result); } else if (pcmk_is_set(request->call_options, st_opt_manual_ack)) { switch (fenced_handle_manual_confirmation(request->ipc_client, request->xml)) { case pcmk_rc_ok: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); break; case EINPROGRESS: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); break; default: fenced_set_protocol_error(&request->result); break; } } else { const char *alternate_host = NULL; xmlNode *dev = pcmk__xpath_find_one(request->xml->doc, "//*[@" PCMK__XA_ST_TARGET "]", LOG_TRACE); const char *target = crm_element_value(dev, PCMK__XA_ST_TARGET); const char *action = crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION); const char *device = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (request->ipc_client != NULL) { int tolerance = 0; crm_notice("Client %s wants to fence (%s) %s using %s", pcmk__request_origin(request), action, target, (device? device : "any device")); crm_element_value_int(dev, PCMK__XA_ST_TOLERANCE, &tolerance); if (stonith_check_fence_tolerance(tolerance, target, action)) { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return fenced_construct_reply(request->xml, NULL, &request->result); } alternate_host = check_alternate_host(target); } else { crm_notice("Peer %s wants to fence (%s) '%s' with device '%s'", request->peer, action, target, (device == NULL)? "(any)" : device); } if (alternate_host != NULL) { const char *client_id = NULL; remote_fencing_op_t *op = NULL; pcmk__node_status_t *node = pcmk__get_node(0, alternate_host, NULL, pcmk__node_search_cluster_member); if (request->ipc_client->id == 0) { client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); } else { client_id = request->ipc_client->id; } /* Create a duplicate fencing operation to relay with the client ID. * When a query response is received, this operation should be * deleted to avoid keeping the duplicate around. */ op = create_remote_stonith_op(client_id, request->xml, FALSE); crm_xml_add(request->xml, PCMK__XA_ST_OP, STONITH_OP_RELAY); crm_xml_add(request->xml, PCMK__XA_ST_CLIENTID, request->ipc_client->id); crm_xml_add(request->xml, PCMK__XA_ST_REMOTE_OP, op->id); // @TODO On failure, fail request immediately, or maybe panic pcmk__cluster_send_message(node, pcmk_ipc_fenced, request->xml); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } else if (initiate_remote_stonith_op(request->ipc_client, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); } else { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } } if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_FENCE_HISTORY static xmlNode * handle_history_request(pcmk__request_t *request) { xmlNode *reply = NULL; xmlNode *data = NULL; stonith_fence_history(request->xml, &data, request->peer, request->call_options); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (!pcmk_is_set(request->call_options, st_opt_discard_reply)) { /* When the local node broadcasts its history, it sets * st_opt_discard_reply and doesn't need a reply. */ reply = fenced_construct_reply(request->xml, data, &request->result); } pcmk__xml_free(data); return reply; } // STONITH_OP_DEVICE_ADD static xmlNode * handle_device_add_request(pcmk__request_t *request) { const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); xmlNode *dev = pcmk__xpath_find_one(request->xml->doc, "//" PCMK__XE_ST_DEVICE_ID, LOG_ERR); if (is_privileged(request->ipc_client, op)) { - int rc = stonith_device_register(dev, FALSE); + int rc = fenced_device_register(dev, false); + rc = pcmk_rc2legacy(rc); pcmk__set_result(&request->result, ((rc == pcmk_ok)? CRM_EX_OK : CRM_EX_ERROR), stonith__legacy2status(rc), ((rc == pcmk_ok)? NULL : pcmk_strerror(rc))); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must register device via CIB"); } fenced_send_config_notification(op, &request->result, (dev == NULL)? NULL : pcmk__xe_id(dev)); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_DEVICE_DEL static xmlNode * handle_device_delete_request(pcmk__request_t *request) { xmlNode *dev = pcmk__xpath_find_one(request->xml->doc, "//" PCMK__XE_ST_DEVICE_ID, LOG_ERR); const char *device_id = crm_element_value(dev, PCMK_XA_ID); const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { stonith_device_remove(device_id, false); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete device via CIB"); } fenced_send_config_notification(op, &request->result, device_id); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_ADD static xmlNode * handle_level_add_request(pcmk__request_t *request) { char *desc = NULL; const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { fenced_register_level(request->xml, &desc, &request->result); } else { unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must add level via CIB"); } fenced_send_config_notification(op, &request->result, desc); free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_DEL static xmlNode * handle_level_delete_request(pcmk__request_t *request) { char *desc = NULL; const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { fenced_unregister_level(request->xml, &desc, &request->result); } else { unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete level via CIB"); } fenced_send_config_notification(op, &request->result, desc); free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // CRM_OP_RM_NODE_CACHE static xmlNode * handle_cache_request(pcmk__request_t *request) { int node_id = 0; const char *name = NULL; crm_element_value_int(request->xml, PCMK_XA_ID, &node_id); name = crm_element_value(request->xml, PCMK_XA_UNAME); pcmk__cluster_forget_cluster_node(node_id, name); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } static xmlNode * handle_unknown_request(pcmk__request_t *request) { crm_err("Unknown IPC request %s from %s %s", request->op, pcmk__request_origin_type(request), pcmk__request_origin(request)); pcmk__format_result(&request->result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Unknown IPC request type '%s' (bug?)", request->op); return fenced_construct_reply(request->xml, NULL, &request->result); } static void fenced_register_handlers(void) { pcmk__server_command_t handlers[] = { { CRM_OP_REGISTER, handle_register_request }, { STONITH_OP_EXEC, handle_agent_request }, { STONITH_OP_TIMEOUT_UPDATE, handle_update_timeout_request }, { STONITH_OP_QUERY, handle_query_request }, { STONITH_OP_NOTIFY, handle_notify_request }, { STONITH_OP_RELAY, handle_relay_request }, { STONITH_OP_FENCE, handle_fence_request }, { STONITH_OP_FENCE_HISTORY, handle_history_request }, { STONITH_OP_DEVICE_ADD, handle_device_add_request }, { STONITH_OP_DEVICE_DEL, handle_device_delete_request }, { STONITH_OP_LEVEL_ADD, handle_level_add_request }, { STONITH_OP_LEVEL_DEL, handle_level_delete_request }, { CRM_OP_RM_NODE_CACHE, handle_cache_request }, { NULL, handle_unknown_request }, }; fenced_handlers = pcmk__register_handlers(handlers); } void fenced_unregister_handlers(void) { if (fenced_handlers != NULL) { g_hash_table_destroy(fenced_handlers); fenced_handlers = NULL; } } static void handle_request(pcmk__request_t *request) { xmlNode *reply = NULL; const char *reason = NULL; if (fenced_handlers == NULL) { fenced_register_handlers(); } reply = pcmk__process_request(request, fenced_handlers); if (reply != NULL) { if (pcmk_is_set(request->flags, pcmk__request_reuse_options) && (request->ipc_client != NULL)) { /* Certain IPC-only commands must reuse the call options from the * original request rather than the ones set by stonith_send_reply() * -> do_local_reply(). */ pcmk__ipc_send_xml(request->ipc_client, request->ipc_id, reply, request->ipc_flags); request->ipc_client->request_id = 0; } else { stonith_send_reply(reply, request->call_options, request->peer, request->ipc_client); } pcmk__xml_free(reply); } reason = request->result.exit_reason; crm_debug("Processed %s request from %s %s: %s%s%s%s", request->op, pcmk__request_origin_type(request), pcmk__request_origin(request), pcmk_exec_status_str(request->result.execution_status), (reason == NULL)? "" : " (", (reason == NULL)? "" : reason, (reason == NULL)? "" : ")"); } static void handle_reply(pcmk__client_t *client, xmlNode *request, const char *remote_peer) { // Copy, because request might be freed before we want to log this char *op = crm_element_value_copy(request, PCMK__XA_ST_OP); if (pcmk__str_eq(op, STONITH_OP_QUERY, pcmk__str_none)) { process_remote_stonith_query(request); } else if (pcmk__str_any_of(op, STONITH_OP_NOTIFY, STONITH_OP_FENCE, NULL)) { fenced_process_fencing_reply(request); } else { crm_err("Ignoring unknown %s reply from %s %s", pcmk__s(op, "untyped"), ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); crm_log_xml_warn(request, "UnknownOp"); free(op); return; } crm_debug("Processed %s reply from %s %s", op, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); free(op); } /*! * \internal * \brief Handle a message from an IPC client or CPG peer * * \param[in,out] client If not NULL, IPC client that sent message * \param[in] id If from IPC client, IPC message ID * \param[in] flags Message flags * \param[in,out] message Message XML * \param[in] remote_peer If not NULL, CPG peer that sent message */ void stonith_command(pcmk__client_t *client, uint32_t id, uint32_t flags, xmlNode *message, const char *remote_peer) { uint32_t call_options = st_opt_none; int rc = pcmk_rc_ok; bool is_reply = false; CRM_CHECK(message != NULL, return); if (pcmk__xpath_find_one(message->doc, "//" PCMK__XE_ST_REPLY, LOG_NEVER) != NULL) { is_reply = true; } rc = pcmk__xe_get_flags(message, PCMK__XA_ST_CALLOPT, &call_options, st_opt_none); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from message: %s", pcmk_rc_str(rc)); } crm_debug("Processing %ssynchronous %s %s %u from %s %s", pcmk_is_set(call_options, st_opt_sync_call)? "" : "a", crm_element_value(message, PCMK__XA_ST_OP), (is_reply? "reply" : "request"), id, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); if (pcmk_is_set(call_options, st_opt_sync_call)) { pcmk__assert((client == NULL) || (client->request_id == id)); } if (is_reply) { handle_reply(client, message, remote_peer); } else { pcmk__request_t request = { .ipc_client = client, .ipc_id = id, .ipc_flags = flags, .peer = remote_peer, .xml = message, .call_options = call_options, .result = PCMK__UNKNOWN_RESULT, }; request.op = crm_element_value_copy(request.xml, PCMK__XA_ST_OP); CRM_CHECK(request.op != NULL, return); if (pcmk_is_set(request.call_options, st_opt_sync_call)) { pcmk__set_request_flags(&request, pcmk__request_sync); } handle_request(&request); pcmk__reset_request(&request); } } diff --git a/daemons/fenced/fenced_scheduler.c b/daemons/fenced/fenced_scheduler.c index 46d74320fc..a67fef5c3a 100644 --- a/daemons/fenced/fenced_scheduler.c +++ b/daemons/fenced/fenced_scheduler.c @@ -1,258 +1,258 @@ /* * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include // fenced_scheduler_run() assumes it's the only place scheduler->input gets set static pcmk_scheduler_t *scheduler = NULL; /*! * \internal * \brief Initialize scheduler data for fencer purposes * * \return Standard Pacemaker return code */ int fenced_scheduler_init(void) { pcmk__output_t *logger = NULL; int rc = pcmk__log_output_new(&logger); if (rc != pcmk_rc_ok) { return rc; } scheduler = pcmk_new_scheduler(); if (scheduler == NULL) { pcmk__output_free(logger); return ENOMEM; } pe__register_messages(logger); pcmk__register_lib_messages(logger); pcmk__output_set_log_level(logger, LOG_TRACE); scheduler->priv->out = logger; return pcmk_rc_ok; } /*! * \internal * \brief Set the local node name for scheduling purposes * * \param[in] node_name Name to set as local node name */ void fenced_set_local_node(const char *node_name) { pcmk__assert(scheduler != NULL); scheduler->priv->local_node_name = pcmk__str_copy(node_name); } /*! * \internal * \brief Get the local node name * * \return Local node name */ const char * fenced_get_local_node(void) { if (scheduler == NULL) { return NULL; } return scheduler->priv->local_node_name; } /*! * \internal * \brief Free all scheduler-related resources */ void fenced_scheduler_cleanup(void) { if (scheduler != NULL) { pcmk__output_t *logger = scheduler->priv->out; if (logger != NULL) { logger->finish(logger, CRM_EX_OK, true, NULL); pcmk__output_free(logger); scheduler->priv->out = NULL; } pcmk_free_scheduler(scheduler); scheduler = NULL; } } /*! * \internal * \brief Check whether the local node is in a resource's allowed node list * * \param[in] rsc Resource to check * * \return Pointer to node if found, otherwise NULL */ static pcmk_node_t * local_node_allowed_for(const pcmk_resource_t *rsc) { if ((rsc != NULL) && (scheduler->priv->local_node_name != NULL)) { GHashTableIter iter; pcmk_node_t *node = NULL; g_hash_table_iter_init(&iter, rsc->priv->allowed_nodes); while (g_hash_table_iter_next(&iter, NULL, (void **) &node)) { if (pcmk__str_eq(node->priv->name, scheduler->priv->local_node_name, pcmk__str_casei)) { return node; } } } return NULL; } /*! * \internal * \brief If a given resource or any of its children are fencing devices, * register the devices * * \param[in,out] data Resource to check * \param[in,out] user_data Ignored */ static void register_if_fencing_device(gpointer data, gpointer user_data) { pcmk_resource_t *rsc = data; const char *rsc_id = pcmk__s(rsc->priv->history_id, rsc->id); xmlNode *xml = NULL; GHashTableIter hash_iter; pcmk_node_t *node = NULL; const char *name = NULL; const char *value = NULL; const char *agent = NULL; const char *rsc_provides = NULL; stonith_key_value_t *params = NULL; // If this is a collective resource, check children instead if (rsc->priv->children != NULL) { for (GList *iter = rsc->priv->children; iter != NULL; iter = iter->next) { register_if_fencing_device(iter->data, NULL); if (pcmk__is_clone(rsc)) { return; // Only one instance needs to be checked for clones } } return; } if (!pcmk_is_set(rsc->flags, pcmk__rsc_fence_device)) { return; // Not a fencing device } if (pe__resource_is_disabled(rsc)) { crm_info("Ignoring fencing device %s because it is disabled", rsc->id); return; } if ((stonith_watchdog_timeout_ms <= 0) && pcmk__str_eq(rsc->id, STONITH_WATCHDOG_ID, pcmk__str_none)) { crm_info("Ignoring fencing device %s " "because watchdog fencing is disabled", rsc->id); return; } // Check whether local node is allowed to run resource node = local_node_allowed_for(rsc); if (node == NULL) { crm_info("Ignoring fencing device %s " "because local node is not allowed to run it", rsc->id); return; } if (node->assign->score < 0) { crm_info("Ignoring fencing device %s " "because local node has preference %s for it", rsc->id, pcmk_readable_score(node->assign->score)); return; } // If device is in a group, check whether local node is allowed for group if (pcmk__is_group(rsc->priv->parent)) { pcmk_node_t *group_node = local_node_allowed_for(rsc->priv->parent); if ((group_node != NULL) && (group_node->assign->score < 0)) { crm_info("Ignoring fencing device %s " "because local node has preference %s for its group", rsc->id, pcmk_readable_score(group_node->assign->score)); return; } } crm_debug("Reloading configuration of fencing device %s", rsc->id); agent = crm_element_value(rsc->priv->xml, PCMK_XA_TYPE); get_meta_attributes(rsc->priv->meta, rsc, NULL, scheduler); rsc_provides = g_hash_table_lookup(rsc->priv->meta, PCMK_STONITH_PROVIDES); g_hash_table_iter_init(&hash_iter, pe_rsc_params(rsc, node, scheduler)); while (g_hash_table_iter_next(&hash_iter, (gpointer *) &name, (gpointer *) &value)) { if ((name == NULL) || (value == NULL)) { continue; } params = stonith_key_value_add(params, name, value); } xml = create_device_registration_xml(rsc_id, st_namespace_any, agent, params, rsc_provides); stonith_key_value_freeall(params, 1, 1); - pcmk__assert(stonith_device_register(xml, TRUE) == pcmk_ok); + pcmk__assert(fenced_device_register(xml, true) == pcmk_rc_ok); pcmk__xml_free(xml); } /*! * \internal * \brief Run the scheduler for fencer purposes * * \param[in] cib CIB to use as scheduler input * * \note Scheduler object is reset before returning, but \p cib is not freed. */ void fenced_scheduler_run(xmlNode *cib) { CRM_CHECK((cib != NULL) && (scheduler != NULL) && (scheduler->input == NULL), return); pcmk_reset_scheduler(scheduler); scheduler->input = cib; pcmk__set_scheduler_flags(scheduler, pcmk__sched_location_only|pcmk__sched_no_counts); pcmk__schedule_actions(scheduler); g_list_foreach(scheduler->priv->resources, register_if_fencing_device, NULL); scheduler->input = NULL; // Wasn't a copy, so don't let API free it pcmk_reset_scheduler(scheduler); } diff --git a/daemons/fenced/pacemaker-fenced.c b/daemons/fenced/pacemaker-fenced.c index 32a3f2bf56..cb862b2284 100644 --- a/daemons/fenced/pacemaker-fenced.c +++ b/daemons/fenced/pacemaker-fenced.c @@ -1,666 +1,666 @@ /* - * Copyright 2009-2024 the Pacemaker project contributors + * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include // PRIu32, PRIx32 #include #include #include #include #include #include #include #include #include #include #include #include #define SUMMARY "daemon for executing fencing devices in a Pacemaker cluster" // @TODO This should be guint long long stonith_watchdog_timeout_ms = 0; GList *stonith_watchdog_targets = NULL; static GMainLoop *mainloop = NULL; gboolean stonith_shutdown_flag = FALSE; static qb_ipcs_service_t *ipcs = NULL; static pcmk__output_t *out = NULL; pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; static struct { gboolean stand_alone; gchar **log_files; } options; crm_exit_t exit_code = CRM_EX_OK; static void stonith_cleanup(void); static int32_t st_ipc_accept(qb_ipcs_connection_t * c, uid_t uid, gid_t gid) { if (stonith_shutdown_flag) { crm_info("Ignoring new client [%d] during shutdown", pcmk__client_pid(c)); return -ECONNREFUSED; } if (pcmk__new_client(c, uid, gid) == NULL) { return -ENOMEM; } return 0; } /* Exit code means? */ static int32_t st_ipc_dispatch(qb_ipcs_connection_t * qbc, void *data, size_t size) { uint32_t id = 0; uint32_t flags = 0; uint32_t call_options = st_opt_none; xmlNode *request = NULL; pcmk__client_t *c = pcmk__find_client(qbc); const char *op = NULL; int rc = pcmk_rc_ok; if (c == NULL) { crm_info("Invalid client: %p", qbc); return 0; } request = pcmk__client_data2xml(c, data, &id, &flags); if (request == NULL) { pcmk__ipc_send_ack(c, id, flags, PCMK__XE_NACK, NULL, CRM_EX_PROTOCOL); return 0; } op = crm_element_value(request, PCMK__XA_CRM_TASK); if(pcmk__str_eq(op, CRM_OP_RM_NODE_CACHE, pcmk__str_casei)) { crm_xml_add(request, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(request, PCMK__XA_ST_OP, op); crm_xml_add(request, PCMK__XA_ST_CLIENTID, c->id); crm_xml_add(request, PCMK__XA_ST_CLIENTNAME, pcmk__client_name(c)); crm_xml_add(request, PCMK__XA_ST_CLIENTNODE, fenced_get_local_node()); pcmk__cluster_send_message(NULL, pcmk_ipc_fenced, request); pcmk__xml_free(request); return 0; } if (c->name == NULL) { const char *value = crm_element_value(request, PCMK__XA_ST_CLIENTNAME); c->name = crm_strdup_printf("%s.%u", pcmk__s(value, "unknown"), c->pid); } rc = pcmk__xe_get_flags(request, PCMK__XA_ST_CALLOPT, &call_options, st_opt_none); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from IPC request: %s", pcmk_rc_str(rc)); } crm_trace("Flags %#08" PRIx32 "/%#08x for command %" PRIu32 " from client %s", flags, call_options, id, pcmk__client_name(c)); if (pcmk_is_set(call_options, st_opt_sync_call)) { pcmk__assert(pcmk_is_set(flags, crm_ipc_client_response)); CRM_LOG_ASSERT(c->request_id == 0); /* This means the client has two synchronous events in-flight */ c->request_id = id; /* Reply only to the last one */ } crm_xml_add(request, PCMK__XA_ST_CLIENTID, c->id); crm_xml_add(request, PCMK__XA_ST_CLIENTNAME, pcmk__client_name(c)); crm_xml_add(request, PCMK__XA_ST_CLIENTNODE, fenced_get_local_node()); crm_log_xml_trace(request, "ipc-received"); stonith_command(c, id, flags, request, NULL); pcmk__xml_free(request); return 0; } /* Error code means? */ static int32_t st_ipc_closed(qb_ipcs_connection_t * c) { pcmk__client_t *client = pcmk__find_client(c); if (client == NULL) { return 0; } crm_trace("Connection %p closed", c); pcmk__free_client(client); /* 0 means: yes, go ahead and destroy the connection */ return 0; } static void st_ipc_destroy(qb_ipcs_connection_t * c) { crm_trace("Connection %p destroyed", c); st_ipc_closed(c); } static void stonith_peer_callback(xmlNode * msg, void *private_data) { const char *remote_peer = crm_element_value(msg, PCMK__XA_SRC); const char *op = crm_element_value(msg, PCMK__XA_ST_OP); if (pcmk__str_eq(op, STONITH_OP_POKE, pcmk__str_none)) { return; } crm_log_xml_trace(msg, "Peer[inbound]"); stonith_command(NULL, 0, 0, msg, remote_peer); } #if SUPPORT_COROSYNC static void handle_cpg_message(cpg_handle_t handle, const struct cpg_name *groupName, uint32_t nodeid, uint32_t pid, void *msg, size_t msg_len) { xmlNode *xml = NULL; const char *from = NULL; char *data = pcmk__cpg_message_data(handle, nodeid, pid, msg, &from); if(data == NULL) { return; } xml = pcmk__xml_parse(data); if (xml == NULL) { crm_err("Invalid XML: '%.120s'", data); free(data); return; } crm_xml_add(xml, PCMK__XA_SRC, from); stonith_peer_callback(xml, NULL); pcmk__xml_free(xml); free(data); } static void stonith_peer_cs_destroy(gpointer user_data) { crm_crit("Lost connection to cluster layer, shutting down"); stonith_shutdown(0); } #endif void do_local_reply(const xmlNode *notify_src, pcmk__client_t *client, int call_options) { /* send callback to originating child */ int local_rc = pcmk_rc_ok; int rid = 0; uint32_t ipc_flags = crm_ipc_server_event; if (pcmk_is_set(call_options, st_opt_sync_call)) { CRM_LOG_ASSERT(client->request_id); rid = client->request_id; client->request_id = 0; ipc_flags = crm_ipc_flags_none; } local_rc = pcmk__ipc_send_xml(client, rid, notify_src, ipc_flags); if (local_rc == pcmk_rc_ok) { crm_trace("Sent response %d to client %s", rid, pcmk__client_name(client)); } else { crm_warn("%synchronous reply to client %s failed: %s", (pcmk_is_set(call_options, st_opt_sync_call)? "S" : "As"), pcmk__client_name(client), pcmk_rc_str(local_rc)); } } uint64_t get_stonith_flag(const char *name) { if (pcmk__str_eq(name, PCMK__VALUE_ST_NOTIFY_FENCE, pcmk__str_none)) { return st_callback_notify_fence; } else if (pcmk__str_eq(name, STONITH_OP_DEVICE_ADD, pcmk__str_casei)) { return st_callback_device_add; } else if (pcmk__str_eq(name, STONITH_OP_DEVICE_DEL, pcmk__str_casei)) { return st_callback_device_del; } else if (pcmk__str_eq(name, PCMK__VALUE_ST_NOTIFY_HISTORY, pcmk__str_none)) { return st_callback_notify_history; } else if (pcmk__str_eq(name, PCMK__VALUE_ST_NOTIFY_HISTORY_SYNCED, pcmk__str_none)) { return st_callback_notify_history_synced; } return st_callback_unknown; } static void stonith_notify_client(gpointer key, gpointer value, gpointer user_data) { const xmlNode *update_msg = user_data; pcmk__client_t *client = value; const char *type = NULL; CRM_CHECK(client != NULL, return); CRM_CHECK(update_msg != NULL, return); type = crm_element_value(update_msg, PCMK__XA_SUBT); CRM_CHECK(type != NULL, crm_log_xml_err(update_msg, "notify"); return); if (client->ipcs == NULL) { crm_trace("Skipping client with NULL channel"); return; } if (pcmk_is_set(client->flags, get_stonith_flag(type))) { int rc = pcmk__ipc_send_xml(client, 0, update_msg, crm_ipc_server_event); if (rc != pcmk_rc_ok) { crm_warn("%s notification of client %s failed: %s " QB_XS " id=%.8s rc=%d", type, pcmk__client_name(client), pcmk_rc_str(rc), client->id, rc); } else { crm_trace("Sent %s notification to client %s", type, pcmk__client_name(client)); } } } void do_stonith_async_timeout_update(const char *client_id, const char *call_id, int timeout) { pcmk__client_t *client = NULL; xmlNode *notify_data = NULL; if (!timeout || !call_id || !client_id) { return; } client = pcmk__find_client_by_id(client_id); if (!client) { return; } notify_data = pcmk__xe_create(NULL, PCMK__XE_ST_ASYNC_TIMEOUT_VALUE); crm_xml_add(notify_data, PCMK__XA_T, PCMK__VALUE_ST_ASYNC_TIMEOUT_VALUE); crm_xml_add(notify_data, PCMK__XA_ST_CALLID, call_id); crm_xml_add_int(notify_data, PCMK__XA_ST_TIMEOUT, timeout); crm_trace("timeout update is %d for client %s and call id %s", timeout, client_id, call_id); if (client) { pcmk__ipc_send_xml(client, 0, notify_data, crm_ipc_server_event); } pcmk__xml_free(notify_data); } /*! * \internal * \brief Notify relevant IPC clients of a fencing operation result * * \param[in] type Notification type * \param[in] result Result of fencing operation (assume success if NULL) * \param[in] data If not NULL, add to notification as call data */ void fenced_send_notification(const char *type, const pcmk__action_result_t *result, xmlNode *data) { /* TODO: Standardize the contents of data */ xmlNode *update_msg = pcmk__xe_create(NULL, PCMK__XE_NOTIFY); CRM_LOG_ASSERT(type != NULL); crm_xml_add(update_msg, PCMK__XA_T, PCMK__VALUE_ST_NOTIFY); crm_xml_add(update_msg, PCMK__XA_SUBT, type); crm_xml_add(update_msg, PCMK__XA_ST_OP, type); stonith__xe_set_result(update_msg, result); if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(update_msg, PCMK__XE_ST_CALLDATA); pcmk__xml_copy(wrapper, data); } crm_trace("Notifying clients"); pcmk__foreach_ipc_client(stonith_notify_client, update_msg); pcmk__xml_free(update_msg); crm_trace("Notify complete"); } /*! * \internal * \brief Send notifications for a configuration change to subscribed clients * * \param[in] op Notification type (\c STONITH_OP_DEVICE_ADD, * \c STONITH_OP_DEVICE_DEL, \c STONITH_OP_LEVEL_ADD, or * \c STONITH_OP_LEVEL_DEL) * \param[in] result Operation result * \param[in] desc Description of what changed (either device ID or string * representation of level * ([])) */ void fenced_send_config_notification(const char *op, const pcmk__action_result_t *result, const char *desc) { xmlNode *notify_data = pcmk__xe_create(NULL, op); crm_xml_add(notify_data, PCMK__XA_ST_DEVICE_ID, desc); fenced_send_notification(op, result, notify_data); pcmk__xml_free(notify_data); } /*! * \internal * \brief Check whether a node does watchdog-fencing * * \param[in] node Name of node to check * * \return TRUE if node found in stonith_watchdog_targets * or stonith_watchdog_targets is empty indicating * all nodes are doing watchdog-fencing */ gboolean node_does_watchdog_fencing(const char *node) { return ((stonith_watchdog_targets == NULL) || pcmk__str_in_list(node, stonith_watchdog_targets, pcmk__str_casei)); } void stonith_shutdown(int nsig) { crm_info("Terminating with %d clients", pcmk__ipc_client_count()); stonith_shutdown_flag = TRUE; if (mainloop != NULL && g_main_loop_is_running(mainloop)) { g_main_loop_quit(mainloop); } } static void stonith_cleanup(void) { fenced_cib_cleanup(); if (ipcs) { qb_ipcs_destroy(ipcs); } pcmk__cluster_destroy_node_caches(); pcmk__client_cleanup(); free_stonith_remote_op_list(); free_topology_list(); - free_device_list(); + fenced_free_device_table(); free_metadata_cache(); fenced_unregister_handlers(); } struct qb_ipcs_service_handlers ipc_callbacks = { .connection_accept = st_ipc_accept, .connection_created = NULL, .msg_process = st_ipc_dispatch, .connection_closed = st_ipc_closed, .connection_destroyed = st_ipc_destroy }; /*! * \internal * \brief Callback for peer status changes * * \param[in] type What changed * \param[in] node What peer had the change * \param[in] data Previous value of what changed */ static void st_peer_update_callback(enum pcmk__node_update type, pcmk__node_status_t *node, const void *data) { if ((type != pcmk__node_update_processes) && !pcmk_is_set(node->flags, pcmk__node_status_remote)) { /* * This is a hack until we can send to a nodeid and/or we fix node name lookups * These messages are ignored in stonith_peer_callback() */ xmlNode *query = pcmk__xe_create(NULL, PCMK__XE_STONITH_COMMAND); crm_xml_add(query, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(query, PCMK__XA_ST_OP, STONITH_OP_POKE); crm_debug("Broadcasting our uname because of node %" PRIu32, node->cluster_layer_id); pcmk__cluster_send_message(NULL, pcmk_ipc_fenced, query); pcmk__xml_free(query); } } /* @COMPAT Deprecated since 2.1.8. Use pcmk_list_fence_attrs() or * crm_resource --list-options=fencing instead of querying daemon metadata. * * NOTE: pcs (as of at least 0.11.8) uses this */ static int fencer_metadata(void) { const char *name = PCMK__SERVER_FENCED; const char *desc_short = N_("Instance attributes available for all " "\"stonith\"-class resources"); const char *desc_long = N_("Instance attributes available for all " "\"stonith\"-class resources and used by " "Pacemaker's fence daemon"); return pcmk__daemon_metadata(out, name, desc_short, desc_long, pcmk__opt_fencing); } static GOptionEntry entries[] = { { "stand-alone", 's', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.stand_alone, N_("Intended for use in regression testing only"), NULL }, { "logfile", 'l', G_OPTION_FLAG_NONE, G_OPTION_ARG_FILENAME_ARRAY, &options.log_files, N_("Send logs to the additional named logfile"), NULL }, { NULL } }; static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { GOptionContext *context = NULL; context = pcmk__build_arg_context(args, "text (default), xml", group, NULL); pcmk__add_main_args(context, entries); return context; } int main(int argc, char **argv) { int rc = pcmk_rc_ok; pcmk_cluster_t *cluster = NULL; crm_ipc_t *old_instance = NULL; GError *error = NULL; GOptionGroup *output_group = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, "l"); GOptionContext *context = build_arg_context(args, &output_group); crm_log_preinit(NULL, argc, argv); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_ERROR; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error creating output format %s: %s", args->output_ty, pcmk_rc_str(rc)); goto done; } if (args->version) { out->version(out, false); goto done; } if ((g_strv_length(processed_args) >= 2) && pcmk__str_eq(processed_args[1], "metadata", pcmk__str_none)) { rc = fencer_metadata(); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Unable to display metadata: %s", pcmk_rc_str(rc)); } goto done; } // Open additional log files pcmk__add_logfiles(options.log_files, out); crm_log_init(NULL, LOG_INFO + args->verbosity, TRUE, (args->verbosity > 0), argc, argv, FALSE); crm_notice("Starting Pacemaker fencer"); old_instance = crm_ipc_new("stonith-ng", 0); if (old_instance == NULL) { /* crm_ipc_new() will have already logged an error message with * crm_err() */ exit_code = CRM_EX_FATAL; goto done; } if (pcmk__connect_generic_ipc(old_instance) == pcmk_rc_ok) { // IPC endpoint already up crm_ipc_close(old_instance); crm_ipc_destroy(old_instance); crm_crit("Aborting start-up because another fencer instance is " "already active"); goto done; } else { // Not up or not authentic, we'll proceed either way crm_ipc_destroy(old_instance); old_instance = NULL; } mainloop_add_signal(SIGTERM, stonith_shutdown); pcmk__cluster_init_node_caches(); rc = fenced_scheduler_init(); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error initializing scheduler data: %s", pcmk_rc_str(rc)); goto done; } cluster = pcmk_cluster_new(); #if SUPPORT_COROSYNC if (pcmk_get_cluster_layer() == pcmk_cluster_layer_corosync) { pcmk_cluster_set_destroy_fn(cluster, stonith_peer_cs_destroy); pcmk_cpg_set_deliver_fn(cluster, handle_cpg_message); pcmk_cpg_set_confchg_fn(cluster, pcmk__cpg_confchg_cb); } #endif // SUPPORT_COROSYNC pcmk__cluster_set_status_callback(&st_peer_update_callback); if (pcmk_cluster_connect(cluster) != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; crm_crit("Cannot sign in to the cluster... terminating"); goto done; } fenced_set_local_node(cluster->priv->node_name); if (!options.stand_alone) { setup_cib(); } - init_device_list(); + fenced_init_device_table(); init_topology_list(); pcmk__serve_fenced_ipc(&ipcs, &ipc_callbacks); // Create the mainloop and run it... mainloop = g_main_loop_new(NULL, FALSE); crm_notice("Pacemaker fencer successfully started and accepting connections"); g_main_loop_run(mainloop); done: g_strfreev(processed_args); pcmk__free_arg_context(context); g_strfreev(options.log_files); stonith_cleanup(); pcmk_cluster_free(cluster); fenced_scheduler_cleanup(); pcmk__output_and_clear_error(&error, out); if (out != NULL) { out->finish(out, exit_code, true, NULL); pcmk__output_free(out); } pcmk__unregister_formats(); crm_exit(exit_code); } diff --git a/daemons/fenced/pacemaker-fenced.h b/daemons/fenced/pacemaker-fenced.h index 04aadedc39..c1590438cc 100644 --- a/daemons/fenced/pacemaker-fenced.h +++ b/daemons/fenced/pacemaker-fenced.h @@ -1,335 +1,333 @@ /* - * Copyright 2009-2024 the Pacemaker project contributors + * Copyright 2009-2025 the Pacemaker project contributors * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include // uint32_t, uint64_t #include // xmlNode #include #include #include #include /*! * \internal * \brief Check whether target has already been fenced recently * * \param[in] tolerance Number of seconds to look back in time * \param[in] target Name of node to search for * \param[in] action Action we want to match * * \return TRUE if an equivalent fencing operation took place in the last * \p tolerance seconds, FALSE otherwise */ gboolean stonith_check_fence_tolerance(int tolerance, const char *target, const char *action); -typedef struct stonith_device_s { +typedef struct { char *id; char *agent; char *namespace; /*! list of actions that must execute on the target node. Used for unfencing */ GString *on_target_actions; GList *targets; time_t targets_age; - gboolean has_attr_map; - - // Whether target's nodeid should be passed as a parameter to the agent - gboolean include_nodeid; /* whether the cluster should automatically unfence nodes with the device */ gboolean automatic_unfencing; - guint priority; uint32_t flags; // Group of enum st_device_flags GHashTable *params; GHashTable *aliases; GList *pending_ops; mainloop_timer_t *timer; crm_trigger_t *work; xmlNode *agent_metadata; /*! A verified device is one that has contacted the * agent successfully to perform a monitor operation */ gboolean verified; gboolean cib_registered; gboolean api_registered; gboolean dirty; -} stonith_device_t; +} fenced_device_t; /* These values are used to index certain arrays by "phase". Usually an * operation has only one "phase", so phase is always zero. However, some * reboots are remapped to "off" then "on", in which case "reboot" will be * phase 0, "off" will be phase 1 and "on" will be phase 2. */ enum st_remap_phase { st_phase_requested = 0, st_phase_off = 1, st_phase_on = 2, st_phase_max = 3 }; typedef struct remote_fencing_op_s { /* @TODO Abstract the overlap with async_command_t (some members have * different names for the same thing), which should allow reducing * duplication in some functions */ /* The unique id associated with this operation */ char *id; /*! The node this operation will fence */ char *target; /*! The fencing action to perform on the target. (reboot, on, off) */ char *action; /*! When was the fencing action recorded (seconds since epoch) */ time_t created; /*! Marks if the final notifications have been sent to local stonith clients. */ gboolean notify_sent; /*! The number of query replies received */ guint replies; /*! The number of query replies expected */ guint replies_expected; /*! Does this node own control of this operation */ gboolean owner; /*! After query is complete, This the high level timer that expires the entire operation */ guint op_timer_total; /*! This timer expires the current fencing request. Many fencing * requests may exist in a single operation */ guint op_timer_one; /*! This timer expires the query request sent out to determine * what nodes are contain what devices, and who those devices can fence */ guint query_timer; /*! This is the default timeout to use for each fencing device if no * custom timeout is received in the query. */ gint base_timeout; /*! This is the calculated total timeout an operation can take before * expiring. This is calculated by adding together all the timeout * values associated with the devices this fencing operation may call */ gint total_timeout; /*! * Fencing delay (in seconds) requested by API client (used by controller to * implement \c PCMK_OPT_PRIORITY_FENCING_DELAY). A value of -1 means * disable all configured delays. */ int client_delay; /*! Delegate is the node being asked to perform a fencing action * on behalf of the node that owns the remote operation. Some operations * will involve multiple delegates. This value represents the final delegate * that is used. */ char *delegate; /*! The point at which the remote operation completed */ time_t completed; //! Group of enum stonith_call_options associated with this operation uint32_t call_options; /*! The current state of the remote operation. This indicates * what stage the op is in, query, exec, done, duplicate, failed. */ enum op_state state; /*! The node that owns the remote operation */ char *originator; /*! The local client id that initiated the fencing request */ char *client_id; /*! The client's call_id that initiated the fencing request */ int client_callid; /*! The name of client that initiated the fencing request */ char *client_name; /*! List of the received query results for all the nodes in the cpg group */ GList *query_results; /*! The original request that initiated the remote stonith operation */ xmlNode *request; /*! The current topology level being executed */ guint level; /*! The current operation phase being executed */ enum st_remap_phase phase; /*! Devices with automatic unfencing (always run if "on" requested, never if remapped) */ GList *automatic_list; /*! List of all devices at the currently executing topology level */ GList *devices_list; /*! Current entry in the topology device list */ GList *devices; /*! List of duplicate operations attached to this operation. Once this operation * completes, the duplicate operations will be closed out as well. */ GList *duplicates; /*! The point at which the remote operation completed(nsec) */ long long completed_nsec; /*! The (potentially intermediate) result of the operation */ pcmk__action_result_t result; } remote_fencing_op_t; void fenced_broadcast_op_result(const remote_fencing_op_t *op, bool op_merged); // Fencer-specific client flags enum st_client_flags { st_callback_unknown = UINT64_C(0), st_callback_notify_fence = (UINT64_C(1) << 0), st_callback_device_add = (UINT64_C(1) << 2), st_callback_device_del = (UINT64_C(1) << 4), st_callback_notify_history = (UINT64_C(1) << 5), st_callback_notify_history_synced = (UINT64_C(1) << 6) }; // How the user specified the target of a topology level enum fenced_target_by { fenced_target_by_unknown = -1, // Invalid or not yet parsed fenced_target_by_name, // By target name fenced_target_by_pattern, // By a pattern matching target names fenced_target_by_attribute, // By a node attribute/value on target }; /* * Complex fencing requirements are specified via fencing topologies. * A topology consists of levels; each level is a list of fencing devices. * Topologies are stored in a hash table by node name. When a node needs to be * fenced, if it has an entry in the topology table, the levels are tried * sequentially, and the devices in each level are tried sequentially. * Fencing is considered successful as soon as any level succeeds; * a level is considered successful if all its devices succeed. * Essentially, all devices at a given level are "and-ed" and the * levels are "or-ed". * * This structure is used for the topology table entries. * Topology levels start from 1, so levels[0] is unused and always NULL. */ typedef struct stonith_topology_s { enum fenced_target_by kind; // How target was specified /*! Node name regex or attribute name=value for which topology applies */ char *target; char *target_value; char *target_pattern; char *target_attribute; /*! Names of fencing devices at each topology level */ GList *levels[ST__LEVEL_COUNT]; } stonith_topology_t; void stonith_shutdown(int nsig); -void init_device_list(void); -void free_device_list(void); +void fenced_init_device_table(void); +void fenced_free_device_table(void); +bool fenced_has_watchdog_device(void); +void fenced_foreach_device(GHFunc fn, gpointer user_data); +void fenced_foreach_device_remove(GHRFunc fn); + void init_topology_list(void); void free_topology_list(void); void free_stonith_remote_op_list(void); void init_stonith_remote_op_hash_table(GHashTable **table); void free_metadata_cache(void); void fenced_unregister_handlers(void); uint64_t get_stonith_flag(const char *name); void stonith_command(pcmk__client_t *client, uint32_t id, uint32_t flags, xmlNode *op_request, const char *remote_peer); -int stonith_device_register(xmlNode *msg, gboolean from_cib); +int fenced_device_register(const xmlNode *dev, bool from_cib); void stonith_device_remove(const char *id, bool from_cib); char *stonith_level_key(const xmlNode *msg, enum fenced_target_by); void fenced_register_level(xmlNode *msg, char **desc, pcmk__action_result_t *result); void fenced_unregister_level(xmlNode *msg, char **desc, pcmk__action_result_t *result); stonith_topology_t *find_topology_for_host(const char *host); void do_local_reply(const xmlNode *notify_src, pcmk__client_t *client, int call_options); xmlNode *fenced_construct_reply(const xmlNode *request, xmlNode *data, const pcmk__action_result_t *result); void do_stonith_async_timeout_update(const char *client, const char *call_id, int timeout); void fenced_send_notification(const char *type, const pcmk__action_result_t *result, xmlNode *data); void fenced_send_config_notification(const char *op, const pcmk__action_result_t *result, const char *desc); remote_fencing_op_t *initiate_remote_stonith_op(const pcmk__client_t *client, xmlNode *request, gboolean manual_ack); void fenced_process_fencing_reply(xmlNode *msg); int process_remote_stonith_query(xmlNode * msg); void *create_remote_stonith_op(const char *client, xmlNode * request, gboolean peer); void stonith_fence_history(xmlNode *msg, xmlNode **output, const char *remote_peer, int options); void stonith_fence_history_trim(void); bool fencing_peer_active(pcmk__node_status_t *peer); void set_fencing_completed(remote_fencing_op_t * op); int fenced_handle_manual_confirmation(const pcmk__client_t *client, xmlNode *msg); const char *fenced_device_reboot_action(const char *device_id); bool fenced_device_supports_on(const char *device_id); gboolean node_has_attr(const char *node, const char *name, const char *value); gboolean node_does_watchdog_fencing(const char *node); void fencing_topology_init(void); void setup_cib(void); void fenced_cib_cleanup(void); int fenced_scheduler_init(void); void fenced_set_local_node(const char *node_name); const char *fenced_get_local_node(void); void fenced_scheduler_cleanup(void); void fenced_scheduler_run(xmlNode *cib); static inline void fenced_set_protocol_error(pcmk__action_result_t *result) { pcmk__set_result(result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Fencer API request missing required information (bug?)"); } /*! * \internal * \brief Get the device flag to use with a given action when searching devices * * \param[in] action Action to check * * \return st_device_supports_on if \p action is "on", otherwise * st_device_supports_none */ static inline uint32_t fenced_support_flag(const char *action) { if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none)) { return st_device_supports_on; } return st_device_supports_none; } -extern GHashTable *device_list; extern GHashTable *topology; extern long long stonith_watchdog_timeout_ms; extern GList *stonith_watchdog_targets; extern GHashTable *stonith_remote_op_list; extern crm_exit_t exit_code; extern gboolean stonith_shutdown_flag; diff --git a/doc/sphinx/Pacemaker_Explained/fencing.rst b/doc/sphinx/Pacemaker_Explained/fencing.rst index 915f69fd0b..dce479e3c6 100644 --- a/doc/sphinx/Pacemaker_Explained/fencing.rst +++ b/doc/sphinx/Pacemaker_Explained/fencing.rst @@ -1,1292 +1,1281 @@ .. index:: single: fencing single: STONITH .. _fencing: Fencing ------- What Is Fencing? ################ *Fencing* is the ability to make a node unable to run resources, even when that node is unresponsive to cluster commands. Fencing is also known as *STONITH*, an acronym for "Shoot The Other Node In The Head", since the most common fencing method is cutting power to the node. Another method is "fabric fencing", cutting the node's access to some capability required to run resources (such as network access or a shared disk). .. index:: single: fencing; why necessary Why Is Fencing Necessary? ######################### Fencing protects your data from being corrupted by malfunctioning nodes or unintentional concurrent access to shared resources. Fencing protects against the "split brain" failure scenario, where cluster nodes have lost the ability to reliably communicate with each other but are still able to run resources. If the cluster just assumed that uncommunicative nodes were down, then multiple instances of a resource could be started on different nodes. The effect of split brain depends on the resource type. For example, an IP address brought up on two hosts on a network will cause packets to randomly be sent to one or the other host, rendering the IP useless. For a database or clustered file system, the effect could be much more severe, causing data corruption or divergence. Fencing is also used when a resource cannot otherwise be stopped. If a resource fails to stop on a node, it cannot be started on a different node without risking the same type of conflict as split-brain. Fencing the original node ensures the resource can be safely started elsewhere. Users may also configure the ``on-fail`` property of :ref:`operation` or the ``loss-policy`` property of :ref:`ticket constraints ` to ``fence``, in which case the cluster will fence the resource's node if the operation fails or the ticket is lost. .. index:: single: fencing; device Fence Devices ############# A *fence device* or *fencing device* is a special type of resource that provides the means to fence a node. Examples of fencing devices include intelligent power switches and IPMI devices that accept SNMP commands to cut power to a node, and iSCSI controllers that allow SCSI reservations to be used to cut a node's access to a shared disk. Since fencing devices will be used to recover from loss of networking connectivity to other nodes, it is essential that they do not rely on the same network as the cluster itself, otherwise that network becomes a single point of failure. Since loss of a node due to power outage is indistinguishable from loss of network connectivity to that node, it is also essential that at least one fence device for a node does not share power with that node. For example, an on-board IPMI controller that shares power with its host should not be used as the sole fencing device for that host. Since fencing is used to isolate malfunctioning nodes, no fence device should rely on its target functioning properly. This includes, for example, devices that ssh into a node and issue a shutdown command (such devices might be suitable for testing, but never for production). .. index:: single: fencing; agent Fence Agents ############ A *fence agent* or *fencing agent* is a ``stonith``-class resource agent. The fence agent standard provides commands (such as ``off`` and ``reboot``) that the cluster can use to fence nodes. As with other resource agent classes, this allows a layer of abstraction so that Pacemaker doesn't need any knowledge about specific fencing technologies -- that knowledge is isolated in the agent. Pacemaker supports two fence agent standards, both inherited from no-longer-active projects: * Red Hat Cluster Suite (RHCS) style: These are typically installed in ``/usr/sbin`` with names starting with ``fence_``. * Linux-HA style: These typically have names starting with ``external/``. Pacemaker can support these agents using the **fence_legacy** RHCS-style agent as a wrapper, *if* support was enabled when Pacemaker was built, which requires the ``cluster-glue`` library. When a Fence Device Can Be Used ############################### Fencing devices do not actually "run" like most services. Typically, they just provide an interface for sending commands to an external device. Additionally, fencing may be initiated by Pacemaker, by other cluster-aware software such as DRBD or DLM, or manually by an administrator, at any point in the cluster life cycle, including before any resources have been started. To accommodate this, Pacemaker does not require the fence device resource to be "started" in order to be used. Whether a fence device is started or not determines whether a node runs any recurring monitor for the device, and gives the node a slight preference for being chosen to execute fencing using that device. By default, any node can execute any fencing device. If a fence device is disabled by setting its ``target-role`` to ``Stopped``, then no node can use that device. If a location constraint with a negative score prevents a specific node from "running" a fence device, then that node will never be chosen to execute fencing using the device. A node may fence itself, but the cluster will choose that only if no other nodes can do the fencing. A common configuration scenario is to have one fence device per target node. In such a case, users often configure anti-location constraints so that the target node does not monitor its own device. Limitations of Fencing Resources ################################ Fencing resources have certain limitations that other resource classes don't: * They may have only one set of meta-attributes and one set of instance attributes. * If :ref:`rules` are used to determine fencing resource options, these might be evaluated only when first read, meaning that later changes to the rules will have no effect. Therefore, it is better to avoid confusion and not use rules at all with fencing resources. These limitations could be revisited if there is sufficient user demand. .. index:: single: fencing; special instance attributes Special Meta-Attributes for Fencing Resources ############################################# The table below lists special resource meta-attributes that may be set for any fencing resource. .. table:: **Additional Properties of Fencing Resources** :widths: 2 1 2 4 +----------------------+---------+--------------------+----------------------------------------+ | Field | Type | Default | Description | +======================+=========+====================+========================================+ | provides | string | | .. index:: | | | | | single: provides | | | | | | | | | | Any special capability provided by the | | | | | fence device. Currently, only one such | | | | | capability is meaningful: | | | | | :ref:`unfencing `. | +----------------------+---------+--------------------+----------------------------------------+ .. _fencing-attributes: Special Instance Attributes for Fencing Resources ################################################# The table below lists special instance attributes that may be set for any fencing resource (*not* meta-attributes, even though they are interpreted by Pacemaker rather than the fence agent). These are also listed in the man page for ``pacemaker-fenced``. -.. Not_Yet_Implemented: - - +----------------------+---------+--------------------+----------------------------------------+ - | priority | integer | 0 | .. index:: | - | | | | single: priority | - | | | | | - | | | | The priority of the fence device. | - | | | | Devices are tried in order of highest | - | | | | priority to lowest. | - +----------------------+---------+--------------------+----------------------------------------+ - .. list-table:: **Additional Properties of Fencing Resources** :class: longtable :widths: 2 1 2 4 :header-rows: 1 * - Name - Type - Default - Description * - .. _primitive_stonith_timeout: .. index:: single: stonith-timeout (primitive instance attribute) stonith-timeout - :ref:`timeout ` - - This is not used by Pacemaker (see the ``pcmk_reboot_timeout``, ``pcmk_off_timeout``, etc., properties instead), but it may be used by Linux-HA fence agents. * - .. _pcmk_host_map: .. index:: single: pcmk_host_map pcmk_host_map - :ref:`text ` - - A mapping of node names to ports for devices that do not understand the node names. For example, ``node1:1;node2:2,3`` tells the cluster to use port 1 for ``node1`` and ports 2 and 3 for ``node2``. If ``pcmk_host_check`` is explicitly set to ``static-list``, either this or ``pcmk_host_list`` must be set. The port portion of the map may contain special characters such as spaces if preceded by a backslash *(since 2.1.2)*. * - .. _pcmk_host_list: .. index:: single: pcmk_host_list pcmk_host_list - :ref:`text ` - - Comma-separated list of nodes that can be targeted by this device (for example, ``node1,node2,node3``). If pcmk_host_check is ``static-list``, either this or ``pcmk_host_map`` must be set. * - .. _pcmk_host_check: .. index:: single: pcmk_host_check pcmk_host_check - :ref:`text ` - See :ref:`pcmk_host_check_default` - The method Pacemaker should use to determine which nodes can be targeted by this device. Allowed values: * ``static-list:`` targets are listed in the ``pcmk_host_list`` or ``pcmk_host_map`` attribute * ``dynamic-list:`` query the device via the agent's ``list`` action * ``status:`` query the device via the agent's ``status`` action * ``none:`` assume the device can fence any node * - .. _pcmk_delay_max: .. index:: single: pcmk_delay_max pcmk_delay_max - :ref:`duration ` - 0s - Enable a delay of no more than the time specified before executing fencing actions. Pacemaker derives the overall delay by taking the value of pcmk_delay_base and adding a random delay value such that the sum is kept below this maximum. This is sometimes used in two-node clusters to ensure that the nodes don't fence each other at the same time. * - .. _pcmk_delay_base: .. index:: single: pcmk_delay_base pcmk_delay_base - :ref:`text ` - 0s - Enable a static delay before executing fencing actions. This can be used, for example, in two-node clusters to ensure that the nodes don't fence each other, by having separate fencing resources with different values. The node that is fenced with the shorter delay will lose a fencing race. The overall delay introduced by pacemaker is derived from this value plus a random delay such that the sum is kept below the maximum delay. A single device can have different delays per node using a host map *(since 2.1.2)*, for example ``node1:0s;node2:5s.`` * - .. _pcmk_action_limit: .. index:: single: pcmk_action_limit pcmk_action_limit - :ref:`integer ` - 1 - The maximum number of actions that can be performed in parallel on this device. A value of -1 means unlimited. Node fencing actions initiated by the cluster (as opposed to an administrator running the ``stonith_admin`` tool or the fencer running recurring device monitors and ``status`` and ``list`` commands) are additionally subject to the ``concurrent-fencing`` cluster property. * - .. _pcmk_host_argument: .. index:: single: pcmk_host_argument pcmk_host_argument - :ref:`text ` - ``port`` if the fence agent metadata advertises support for it, otherwise ``plug`` if supported, otherwise ``none`` - *Advanced use only.* Which parameter should be supplied to the fence agent to identify the node to be fenced. A value of ``none`` tells the cluster not to supply any additional parameters. * - .. _pcmk_reboot_action: .. index:: single: pcmk_reboot_action pcmk_reboot_action - :ref:`text ` - ``reboot`` - *Advanced use only.* The command to send to the resource agent in order to reboot a node. Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific command. * - .. _pcmk_reboot_timeout: .. index:: single: pcmk_reboot_timeout pcmk_reboot_timeout - :ref:`timeout ` - 60s - *Advanced use only.* Specify an alternate timeout (in seconds) to use for ``reboot`` actions instead of the value of ``stonith-timeout``. Some devices need much more or less time to complete than normal. Use this to specify an alternate, device-specific timeout. * - .. _pcmk_reboot_retries: .. index:: single: pcmk_reboot_retries pcmk_reboot_retries - :ref:`integer ` - 2 - *Advanced use only.* The maximum number of times to retry the ``reboot`` command within the timeout period. Some devices do not support multiple connections, and operations may fail if the device is busy with another task, so Pacemaker will automatically retry the operation, if there is time remaining. Use this option to alter the number of times Pacemaker retries before giving up. * - .. _pcmk_off_action: .. index:: single: pcmk_off_action pcmk_off_action - :ref:`text ` - ``off`` - *Advanced use only.* The command to send to the resource agent in order to shut down a node. Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific command. * - .. _pcmk_off_timeout: .. index:: single: pcmk_off_timeout pcmk_off_timeout - :ref:`timeout ` - 60s - *Advanced use only.* Specify an alternate timeout (in seconds) to use for ``off`` actions instead of the value of ``stonith-timeout``. Some devices need much more or less time to complete than normal. Use this to specify an alternate, device-specific timeout. * - .. _pcmk_off_retries: .. index:: single: pcmk_off_retries pcmk_off_retries - :ref:`integer ` - 2 - *Advanced use only.* The maximum number of times to retry the ``off`` command within the timeout period. Some devices do not support multiple connections, and operations may fail if the device is busy with another task, so Pacemaker will automatically retry the operation, if there is time remaining. Use this option to alter the number of times Pacemaker retries before giving up. * - .. _pcmk_list_action: .. index:: single: pcmk_list_action pcmk_list_action - :ref:`text ` - ``list`` - *Advanced use only.* The command to send to the resource agent in order to list nodes. Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific command. * - .. _pcmk_list_timeout: .. index:: single: pcmk_list_timeout pcmk_list_timeout - :ref:`timeout ` - 60s - *Advanced use only.* Specify an alternate timeout (in seconds) to use for ``list`` actions instead of the value of ``stonith-timeout``. Some devices need much more or less time to complete than normal. Use this to specify an alternate, device-specific timeout. * - .. _pcmk_list_retries: .. index:: single: pcmk_list_retries pcmk_list_retries - :ref:`integer ` - 2 - *Advanced use only.* The maximum number of times to retry the ``list`` command within the timeout period. Some devices do not support multiple connections, and operations may fail if the device is busy with another task, so Pacemaker will automatically retry the operation, if there is time remaining. Use this option to alter the number of times Pacemaker retries before giving up. * - .. _pcmk_monitor_action: .. index:: single: pcmk_monitor_action pcmk_monitor_action - :ref:`text ` - ``monitor`` - *Advanced use only.* The command to send to the resource agent in order to report extended status. Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific command. * - .. _pcmk_monitor_timeout: .. index:: single: pcmk_monitor_timeout pcmk_monitor_timeout - :ref:`timeout ` - 60s - *Advanced use only.* Specify an alternate timeout (in seconds) to use for ``monitor`` actions instead of the value of ``stonith-timeout``. Some devices need much more or less time to complete than normal. Use this to specify an alternate, device-specific timeout. * - .. _pcmk_monitor_retries: .. index:: single: pcmk_monitor_retries pcmk_monitor_retries - :ref:`integer ` - 2 - *Advanced use only.* The maximum number of times to retry the ``monitor`` command within the timeout period. Some devices do not support multiple connections, and operations may fail if the device is busy with another task, so Pacemaker will automatically retry the operation, if there is time remaining. Use this option to alter the number of times Pacemaker retries before giving up. * - .. _pcmk_status_action: .. index:: single: pcmk_status_action pcmk_status_action - :ref:`text ` - ``status`` - *Advanced use only.* The command to send to the resource agent in order to report status. Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific command. * - .. _pcmk_status_timeout: .. index:: single: pcmk_status_timeout pcmk_status_timeout - :ref:`timeout ` - 60s - *Advanced use only.* Specify an alternate timeout (in seconds) to use for ``status`` actions instead of the value of ``stonith-timeout``. Some devices need much more or less time to complete than normal. Use this to specify an alternate, device-specific timeout. * - .. _pcmk_status_retries: .. index:: single: pcmk_status_retries pcmk_status_retries - :ref:`integer ` - 2 - *Advanced use only.* The maximum number of times to retry the ``status`` command within the timeout period. Some devices do not support multiple connections, and operations may fail if the device is busy with another task, so Pacemaker will automatically retry the operation, if there is time remaining. Use this option to alter the number of times Pacemaker retries before giving up. .. _pcmk_host_check_default: Default Check Type ################## If the user does not explicitly configure ``pcmk_host_check`` for a fence device, a default value appropriate to other configured parameters will be used: * If either ``pcmk_host_list`` or ``pcmk_host_map`` is configured, ``static-list`` will be used; * otherwise, if the fence device supports the ``list`` action, and the first attempt at using ``list`` succeeds, ``dynamic-list`` will be used; * otherwise, if the fence device supports the ``status`` action, ``status`` will be used; * otherwise, ``none`` will be used. .. index:: single: unfencing single: fencing; unfencing .. _unfencing: Unfencing ######### With fabric fencing (such as cutting network or shared disk access rather than power), it is expected that the cluster will fence the node, and then a system administrator must manually investigate what went wrong, correct any issues found, then reboot (or restart the cluster services on) the node. Once the node reboots and rejoins the cluster, some fabric fencing devices require an explicit command to restore the node's access. This capability is called *unfencing* and is typically implemented as the fence agent's ``on`` command. If any cluster resource has ``requires`` set to ``unfencing``, then that resource will not be probed or started on a node until that node has been unfenced. Fencing and Quorum ################## In general, a cluster partition may execute fencing only if the partition has quorum, and the ``stonith-enabled`` cluster property is set to true. However, there are exceptions: * The requirements apply only to fencing initiated by Pacemaker. If an administrator initiates fencing using the ``stonith_admin`` command, or an external application such as DLM initiates fencing using Pacemaker's C API, the requirements do not apply. * A cluster partition without quorum is allowed to fence any active member of that partition. As a corollary, this allows a ``no-quorum-policy`` of ``suicide`` to work. * If the ``no-quorum-policy`` cluster property is set to ``ignore``, then quorum is not required to execute fencing of any node. Fencing Timeouts ################ Fencing timeouts are complicated, since a single fencing operation can involve many steps, each of which may have a separate timeout. Fencing may be initiated in one of several ways: * An administrator may initiate fencing using the ``stonith_admin`` tool, which has a ``--timeout`` option (defaulting to 2 minutes) that will be used as the fence operation timeout. * An external application such as DLM may initiate fencing using the Pacemaker C API. The application will specify the fence operation timeout in this case, which might or might not be configurable by the user. * The cluster may initiate fencing itself. In this case, the ``stonith-timeout`` cluster property (defaulting to 1 minute) will be used as the fence operation timeout. However fencing is initiated, the initiator contacts Pacemaker's fencer (``pacemaker-fenced``) to request fencing. This connection and request has its own timeout, separate from the fencing operation timeout, but usually happens very quickly. The fencer will contact all fencers in the cluster to ask what devices they have available to fence the target node. The fence operation timeout will be used as the timeout for each of these queries. Once a fencing device has been selected, the fencer will check whether any action-specific timeout has been configured for the device, to use instead of the fence operation timeout. For example, if ``stonith-timeout`` is 60 seconds, but the fencing device has ``pcmk_reboot_timeout`` configured as 90 seconds, then a timeout of 90 seconds will be used for reboot actions using that device. A device may have retries configured, in which case the timeout applies across all attempts. For example, if a device has ``pcmk_reboot_retries`` configured as 2, and the first reboot attempt fails, the second attempt will only have whatever time is remaining in the action timeout after subtracting how much time the first attempt used. This means that if the first attempt fails due to using the entire timeout, no further attempts will be made. There is currently no way to configure a per-attempt timeout. If more than one device is required to fence a target, whether due to failure of the first device or a fencing topology with multiple devices configured for the target, each device will have its own separate action timeout. For all of the above timeouts, the fencer will generally multiply the configured value by 1.2 to get an actual value to use, to account for time needed by the fencer's own processing. Separate from the fencer's timeouts, some fence agents have internal timeouts for individual steps of their fencing process. These agents often have parameters to configure these timeouts, such as ``login-timeout``, ``shell-timeout``, or ``power-timeout``. Many such agents also have a ``disable-timeout`` parameter to ignore their internal timeouts and just let Pacemaker handle the timeout. This causes a difference in retry behavior. If ``disable-timeout`` is not set, and the agent hits one of its internal timeouts, it will report that as a failure to Pacemaker, which can then retry. If ``disable-timeout`` is set, and Pacemaker hits a timeout for the agent, then there will be no time remaining, and no retry will be done. Fence Devices Dependent on Other Resources ########################################## In some cases, a fence device may require some other cluster resource (such as an IP address) to be active in order to function properly. This is obviously undesirable in general: fencing may be required when the depended-on resource is not active, or fencing may be required because the node running the depended-on resource is no longer responding. However, this may be acceptable under certain conditions: * The dependent fence device should not be able to target any node that is allowed to run the depended-on resource. * The depended-on resource should not be disabled during production operation. * The ``concurrent-fencing`` cluster property should be set to ``true``. Otherwise, if both the node running the depended-on resource and some node targeted by the dependent fence device need to be fenced, the fencing of the node running the depended-on resource might be ordered first, making the second fencing impossible and blocking further recovery. With concurrent fencing, the dependent fence device might fail at first due to the depended-on resource being unavailable, but it will be retried and eventually succeed once the resource is brought back up. Even under those conditions, there is one unlikely problem scenario. The DC always schedules fencing of itself after any other fencing needed, to avoid unnecessary repeated DC elections. If the dependent fence device targets the DC, and both the DC and a different node running the depended-on resource need to be fenced, the DC fencing will always fail and block further recovery. Note, however, that losing a DC node entirely causes some other node to become DC and schedule the fencing, so this is only a risk when a stop or other operation with ``on-fail`` set to ``fencing`` fails on the DC. .. index:: single: fencing; configuration Configuring Fencing ################### Higher-level tools can provide simpler interfaces to this process, but using Pacemaker command-line tools, this is how you could configure a fence device. #. Find the correct driver: .. code-block:: none # stonith_admin --list-installed .. note:: You may have to install packages to make fence agents available on your host. Searching your available packages for ``fence-`` is usually helpful. Ensure the packages providing the fence agents you require are installed on every cluster node. #. Find the required parameters associated with the device (replacing ``$AGENT_NAME`` with the name obtained from the previous step): .. code-block:: none # stonith_admin --metadata --agent $AGENT_NAME #. Create a file called ``stonith.xml`` containing a primitive resource with a class of ``stonith``, a type equal to the agent name obtained earlier, and a parameter for each of the values returned in the previous step. #. If the device does not know how to fence nodes based on their uname, you may also need to set the special ``pcmk_host_map`` parameter. See :ref:`fencing-attributes` for details. #. If the device does not support the ``list`` command, you may also need to set the special ``pcmk_host_list`` and/or ``pcmk_host_check`` parameters. See :ref:`fencing-attributes` for details. #. If the device does not expect the target to be specified with the ``port`` parameter, you may also need to set the special ``pcmk_host_argument`` parameter. See :ref:`fencing-attributes` for details. #. Upload it into the CIB using cibadmin: .. code-block:: none # cibadmin --create --scope resources --xml-file stonith.xml #. Set ``stonith-enabled`` to true: .. code-block:: none # crm_attribute --type crm_config --name stonith-enabled --update true #. Once the stonith resource is running, you can test it by executing the following, replacing ``$NODE_NAME`` with the name of the node to fence (although you might want to stop the cluster on that machine first): .. code-block:: none # stonith_admin --reboot $NODE_NAME Example Fencing Configuration _____________________________ For this example, we assume we have a cluster node, ``pcmk-1``, whose IPMI controller is reachable at the IP address 192.0.2.1. The IPMI controller uses the username ``testuser`` and the password ``abc123``. #. Looking at what's installed, we may see a variety of available agents: .. code-block:: none # stonith_admin --list-installed .. code-block:: none (... some output omitted ...) fence_idrac fence_ilo3 fence_ilo4 fence_ilo5 fence_imm fence_ipmilan (... some output omitted ...) Perhaps after some reading some man pages and doing some Internet searches, we might decide ``fence_ipmilan`` is our best choice. #. Next, we would check what parameters ``fence_ipmilan`` provides: .. code-block:: none # stonith_admin --metadata -a fence_ipmilan .. code-block:: xml fence_ipmilan is an I/O Fencing agentwhich can be used with machines controlled by IPMI.This agent calls support software ipmitool (http://ipmitool.sf.net/). WARNING! This fence agent might report success before the node is powered off. You should use -m/method onoff if your fence device works correctly with that option. Fencing action IPMI Lan Auth type. Ciphersuite to use (same as ipmitool -C parameter) Hexadecimal-encoded Kg key for IPMIv2 authentication IP address or hostname of fencing device IP address or hostname of fencing device TCP/UDP port to use for connection with device Use Lanplus to improve security of connection Login name Method to fence Login password or passphrase Script to run to retrieve password Login password or passphrase Script to run to retrieve password IP address or hostname of fencing device (together with --port-as-ip) IP address or hostname of fencing device (together with --port-as-ip) Privilege level on IPMI device Bridge IPMI requests to the remote target address Login name Disable logging to stderr. Does not affect --verbose or --debug-file or logging to syslog. Verbose mode Write debug information to given file Write debug information to given file Display version information and exit Display help and exit Wait X seconds before fencing is started Path to ipmitool binary Wait X seconds for cmd prompt after login Make "port/plug" to be an alias to IP address Test X seconds for status change after ON/OFF Wait X seconds after issuing ON/OFF Wait X seconds for cmd prompt after issuing command Count of attempts to retry power on Use sudo (without password) when calling 3rd party software Use sudo (without password) when calling 3rd party software Path to sudo binary Once we've decided what parameter values we think we need, it is a good idea to run the fence agent's status action manually, to verify that our values work correctly: .. code-block:: none # fence_ipmilan --lanplus -a 192.0.2.1 -l testuser -p abc123 -o status Chassis Power is on #. Based on that, we might create a fencing resource configuration like this in ``stonith.xml`` (or any file name, just use the same name with ``cibadmin`` later): .. code-block:: xml .. note:: Even though the man page shows that the ``action`` parameter is supported, we do not provide that in the resource configuration. Pacemaker will supply an appropriate action whenever the fence device must be used. #. In this case, we don't need to configure ``pcmk_host_map`` because ``fence_ipmilan`` ignores the target node name and instead uses its ``ip`` parameter to know how to contact the IPMI controller. #. We do need to let Pacemaker know which cluster node can be fenced by this device, since ``fence_ipmilan`` doesn't support the ``list`` action. Add a line like this to the agent's instance attributes: .. code-block:: xml #. We don't need to configure ``pcmk_host_argument`` since ``ip`` is all the fence agent needs (it ignores the target name). #. Make the configuration active: .. code-block:: none # cibadmin --create --scope resources --xml-file stonith.xml #. Set ``stonith-enabled`` to true (this only has to be done once): .. code-block:: none # crm_attribute --type crm_config --name stonith-enabled --update true #. Since our cluster is still in testing, we can reboot ``pcmk-1`` without bothering anyone, so we'll test our fencing configuration by running this from one of the other cluster nodes: .. code-block:: none # stonith_admin --reboot pcmk-1 Then we will verify that the node did, in fact, reboot. We can repeat that process to create a separate fencing resource for each node. With some other fence device types, a single fencing resource is able to be used for all nodes. In fact, we could do that with ``fence_ipmilan``, using the ``port-as-ip`` parameter along with ``pcmk_host_map``. Either approach is fine. .. index:: single: fencing; topology single: fencing-topology single: fencing-level Fencing Topologies ################## Pacemaker supports fencing nodes with multiple devices through a feature called *fencing topologies*. Fencing topologies may be used to provide alternative devices in case one fails, or to require multiple devices to all be executed successfully in order to consider the node successfully fenced, or even a combination of the two. Create the individual devices as you normally would, then define one or more ``fencing-level`` entries in the ``fencing-topology`` section of the configuration. * Each fencing level is attempted in order of ascending ``index``. Allowed values are 1 through 9. * If a device fails, processing terminates for the current level. No further devices in that level are exercised, and the next level is attempted instead. * If the operation succeeds for all the listed devices in a level, the level is deemed to have passed. * The operation is finished when a level has passed (success), or all levels have been attempted (failed). * If the operation failed, the next step is determined by the scheduler and/or the controller. Some possible uses of topologies include: * Try on-board IPMI, then an intelligent power switch if that fails * Try fabric fencing of both disk and network, then fall back to power fencing if either fails * Wait up to a certain time for a kernel dump to complete, then cut power to the node .. table:: **Attributes of a fencing-level Element** :class: longtable :widths: 1 4 +------------------+-----------------------------------------------------------------------------------------+ | Attribute | Description | +==================+=========================================================================================+ | id | .. index:: | | | pair: fencing-level; id | | | | | | A unique name for this element (required) | +------------------+-----------------------------------------------------------------------------------------+ | target | .. index:: | | | pair: fencing-level; target | | | | | | The name of a single node to which this level applies | +------------------+-----------------------------------------------------------------------------------------+ | target-pattern | .. index:: | | | pair: fencing-level; target-pattern | | | | | | An extended regular expression (as defined in `POSIX | | | `_) | | | matching the names of nodes to which this level applies | +------------------+-----------------------------------------------------------------------------------------+ | target-attribute | .. index:: | | | pair: fencing-level; target-attribute | | | | | | The name of a node attribute that is set (to ``target-value``) for nodes to which this | | | level applies | +------------------+-----------------------------------------------------------------------------------------+ | target-value | .. index:: | | | pair: fencing-level; target-value | | | | | | The node attribute value (of ``target-attribute``) that is set for nodes to which this | | | level applies | +------------------+-----------------------------------------------------------------------------------------+ | index | .. index:: | | | pair: fencing-level; index | | | | | | The order in which to attempt the levels. Levels are attempted in ascending order | | | *until one succeeds*. Valid values are 1 through 9. | +------------------+-----------------------------------------------------------------------------------------+ | devices | .. index:: | | | pair: fencing-level; devices | | | | | | A comma-separated list of devices that must all be tried for this level | +------------------+-----------------------------------------------------------------------------------------+ .. note:: **Fencing topology with different devices for different nodes** .. code-block:: xml ... ... Example Dual-Layer, Dual-Device Fencing Topologies __________________________________________________ The following example illustrates an advanced use of ``fencing-topology`` in a cluster with the following properties: * 2 nodes (prod-mysql1 and prod-mysql2) * the nodes have IPMI controllers reachable at 192.0.2.1 and 192.0.2.2 * the nodes each have two independent Power Supply Units (PSUs) connected to two independent Power Distribution Units (PDUs) reachable at 198.51.100.1 (port 10 and port 11) and 203.0.113.1 (port 10 and port 11) * fencing via the IPMI controller uses the ``fence_ipmilan`` agent (1 fence device per controller, with each device targeting a separate node) * fencing via the PDUs uses the ``fence_apc_snmp`` agent (1 fence device per PDU, with both devices targeting both nodes) * a random delay is used to lessen the chance of a "death match" * fencing topology is set to try IPMI fencing first then dual PDU fencing if that fails In a node failure scenario, Pacemaker will first select ``fence_ipmilan`` to try to kill the faulty node. Using the fencing topology, if that method fails, it will then move on to selecting ``fence_apc_snmp`` twice (once for the first PDU, then again for the second PDU). The fence action is considered successful only if both PDUs report the required status. If any of them fails, fencing loops back to the first fencing method, ``fence_ipmilan``, and so on, until the node is fenced or the fencing action is cancelled. .. note:: **First fencing method: single IPMI device per target** Each cluster node has it own dedicated IPMI controller that can be contacted for fencing using the following primitives: .. code-block:: xml .. note:: **Second fencing method: dual PDU devices** Each cluster node also has 2 distinct power supplies controlled by 2 distinct PDUs: * Node 1: PDU 1 port 10 and PDU 2 port 10 * Node 2: PDU 1 port 11 and PDU 2 port 11 The matching fencing agents are configured as follows: .. code-block:: xml .. note:: **Fencing topology** Now that all the fencing resources are defined, it's time to create the right topology. We want to first fence using IPMI and if that does not work, fence both PDUs to effectively and surely kill the node. .. code-block:: xml In ``fencing-topology``, the lowest ``index`` value for a target determines its first fencing method. Remapping Reboots ################# When the cluster needs to reboot a node, whether because ``stonith-action`` is ``reboot`` or because a reboot was requested externally (such as by ``stonith_admin --reboot``), it will remap that to other commands in two cases: * If the chosen fencing device does not support the ``reboot`` command, the cluster will ask it to perform ``off`` instead. * If a fencing topology level with multiple devices must be executed, the cluster will ask all the devices to perform ``off``, then ask the devices to perform ``on``. To understand the second case, consider the example of a node with redundant power supplies connected to intelligent power switches. Rebooting one switch and then the other would have no effect on the node. Turning both switches off, and then on, actually reboots the node. In such a case, the fencing operation will be treated as successful as long as the ``off`` commands succeed, because then it is safe for the cluster to recover any resources that were on the node. Timeouts and errors in the ``on`` phase will be logged but ignored. When a reboot operation is remapped, any action-specific timeout for the remapped action will be used (for example, ``pcmk_off_timeout`` will be used when executing the ``off`` command, not ``pcmk_reboot_timeout``). diff --git a/include/crm/fencing/internal.h b/include/crm/fencing/internal.h index d46dab7ab9..4615993645 100644 --- a/include/crm/fencing/internal.h +++ b/include/crm/fencing/internal.h @@ -1,199 +1,198 @@ /* - * Copyright 2011-2024 the Pacemaker project contributors + * Copyright 2011-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #ifndef PCMK__CRM_FENCING_INTERNAL__H #define PCMK__CRM_FENCING_INTERNAL__H #include #include #include #include #include #include #ifdef __cplusplus extern "C" { #endif enum st_device_flags { st_device_supports_none = (0 << 0), st_device_supports_list = (1 << 0), st_device_supports_status = (1 << 1), st_device_supports_reboot = (1 << 2), st_device_supports_parameter_plug = (1 << 3), st_device_supports_parameter_port = (1 << 4), st_device_supports_on = (1 << 5), }; #define stonith__set_device_flags(device_flags, device_id, flags_to_set) do { \ device_flags = pcmk__set_flags_as(__func__, __LINE__, LOG_TRACE, \ "Fence device", device_id, \ (device_flags), (flags_to_set), \ #flags_to_set); \ } while (0) #define stonith__set_call_options(st_call_opts, call_for, flags_to_set) do { \ st_call_opts = pcmk__set_flags_as(__func__, __LINE__, LOG_TRACE, \ "Fencer call", (call_for), \ (st_call_opts), (flags_to_set), \ #flags_to_set); \ } while (0) #define stonith__clear_call_options(st_call_opts, call_for, flags_to_clear) do { \ st_call_opts = pcmk__clear_flags_as(__func__, __LINE__, LOG_TRACE, \ "Fencer call", (call_for), \ (st_call_opts), (flags_to_clear), \ #flags_to_clear); \ } while (0) struct stonith_action_s; typedef struct stonith_action_s stonith_action_t; stonith_action_t *stonith__action_create(const char *agent, const char *action_name, const char *target, - uint32_t target_nodeid, int timeout_sec, GHashTable *device_args, GHashTable *port_map, const char *host_arg); void stonith__destroy_action(stonith_action_t *action); pcmk__action_result_t *stonith__action_result(stonith_action_t *action); int stonith__result2rc(const pcmk__action_result_t *result); void stonith__xe_set_result(xmlNode *xml, const pcmk__action_result_t *result); void stonith__xe_get_result(const xmlNode *xml, pcmk__action_result_t *result); xmlNode *stonith__find_xe_with_result(xmlNode *xml); int stonith__execute_async(stonith_action_t *action, void *userdata, void (*done) (int pid, const pcmk__action_result_t *result, void *user_data), void (*fork_cb) (int pid, void *user_data)); int stonith__metadata_async(const char *agent, int timeout_sec, void (*callback)(int pid, const pcmk__action_result_t *result, void *user_data), void *user_data); xmlNode *create_level_registration_xml(const char *node, const char *pattern, const char *attr, const char *value, int level, const stonith_key_value_t *device_list); xmlNode *create_device_registration_xml(const char *id, enum stonith_namespace standard, const char *agent, const stonith_key_value_t *params, const char *rsc_provides); void stonith__register_messages(pcmk__output_t *out); GList *stonith__parse_targets(const char *hosts); const char *stonith__later_succeeded(const stonith_history_t *event, const stonith_history_t *top_history); stonith_history_t *stonith__sort_history(stonith_history_t *history); void stonith__device_parameter_flags(uint32_t *device_flags, const char *device_name, xmlNode *metadata); /* Only 1-9 is allowed for fencing topology levels, * however, 0 is used to unregister all levels in * unregister requests. */ # define ST__LEVEL_COUNT 10 # define STONITH_ATTR_ACTION_OP "action" # define STONITH_OP_EXEC "st_execute" # define STONITH_OP_TIMEOUT_UPDATE "st_timeout_update" # define STONITH_OP_QUERY "st_query" # define STONITH_OP_FENCE "st_fence" # define STONITH_OP_RELAY "st_relay" # define STONITH_OP_DEVICE_ADD "st_device_register" # define STONITH_OP_DEVICE_DEL "st_device_remove" # define STONITH_OP_FENCE_HISTORY "st_fence_history" # define STONITH_OP_LEVEL_ADD "st_level_add" # define STONITH_OP_LEVEL_DEL "st_level_remove" # define STONITH_OP_NOTIFY "st_notify" # define STONITH_OP_POKE "poke" # define STONITH_WATCHDOG_AGENT "fence_watchdog" /* Don't change 2 below as it would break rolling upgrade */ # define STONITH_WATCHDOG_AGENT_INTERNAL "#watchdog" # define STONITH_WATCHDOG_ID "watchdog" stonith_history_t *stonith__first_matching_event(stonith_history_t *history, bool (*matching_fn)(stonith_history_t *, void *), void *user_data); bool stonith__event_state_pending(stonith_history_t *history, void *user_data); bool stonith__event_state_eq(stonith_history_t *history, void *user_data); bool stonith__event_state_neq(stonith_history_t *history, void *user_data); int stonith__legacy2status(int rc); int stonith__exit_status(const stonith_callback_data_t *data); int stonith__execution_status(const stonith_callback_data_t *data); const char *stonith__exit_reason(const stonith_callback_data_t *data); int stonith__event_exit_status(const stonith_event_t *event); int stonith__event_execution_status(const stonith_event_t *event); const char *stonith__event_exit_reason(const stonith_event_t *event); char *stonith__event_description(const stonith_event_t *event); gchar *stonith__history_description(const stonith_history_t *event, bool full_history, const char *later_succeeded, uint32_t show_opts); /*! * \internal * \brief Is a fencing operation in pending state? * * \param[in] state State as enum op_state value * * \return A boolean */ static inline bool stonith__op_state_pending(enum op_state state) { return state != st_failed && state != st_done; } gboolean stonith__watchdog_fencing_enabled_for_node(const char *node); gboolean stonith__watchdog_fencing_enabled_for_node_api(stonith_t *st, const char *node); /*! * \internal * \brief Validate a fencing configuration * * \param[in,out] st Fencer connection to use * \param[in] call_options Group of enum stonith_call_options * \param[in] rsc_id Resource to validate * \param[in] namespace_s Type of fence agent to search for * \param[in] agent Fence agent to validate * \param[in,out] params Fence device configuration parameters * \param[in] timeout_sec How long to wait for operation to complete * \param[in,out] output If non-NULL, where to store any agent output * \param[in,out] error_output If non-NULL, where to store agent error output * * \return Standard Pacemaker return code */ int stonith__validate(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, GHashTable *params, int timeout_sec, char **output, char **error_output); #ifdef __cplusplus } #endif #endif // PCMK__CRM_FENCING_INTERNAL__H diff --git a/lib/fencing/st_actions.c b/lib/fencing/st_actions.c index d085ff8ab3..5da7acf026 100644 --- a/lib/fencing/st_actions.c +++ b/lib/fencing/st_actions.c @@ -1,730 +1,717 @@ /* * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include // xmlNode #include #include #include #include #include #include "fencing_private.h" struct stonith_action_s { /*! user defined data */ char *agent; char *action; GHashTable *args; int timeout; bool async; void *userdata; void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data); void (*fork_cb) (int pid, void *user_data); svc_action_t *svc_action; /*! internal timing information */ time_t initial_start_time; int tries; int remaining_timeout; int max_retries; int pid; pcmk__action_result_t result; }; static int internal_stonith_action_execute(stonith_action_t *action); static void log_action(stonith_action_t *action, pid_t pid); /*! * \internal * \brief Set an action's result based on services library result * * \param[in,out] action Fence action to set result for * \param[in,out] svc_action Service action to get result from */ static void set_result_from_svc_action(stonith_action_t *action, svc_action_t *svc_action) { services__copy_result(svc_action, &(action->result)); pcmk__set_result_output(&(action->result), services__grab_stdout(svc_action), services__grab_stderr(svc_action)); } static void log_action(stonith_action_t *action, pid_t pid) { /* The services library has already logged the output at info or debug * level, so just raise to warning for stderr. */ if (action->result.action_stderr != NULL) { /* Logging the whole string confuses syslog when the string is xml */ char *prefix = crm_strdup_printf("%s[%d] stderr:", action->agent, pid); crm_log_output(LOG_WARNING, prefix, action->result.action_stderr); free(prefix); } } static void append_config_arg(gpointer key, gpointer value, gpointer user_data) { /* Filter out parameters handled directly by Pacemaker. * * STONITH_ATTR_ACTION_OP is added elsewhere and should never be part of the * fencing resource's parameter list. We should ignore its value if it is * configured there. */ if (!pcmk__str_eq(key, STONITH_ATTR_ACTION_OP, pcmk__str_casei) && !pcmk_stonith_param(key) && (strstr(key, CRM_META) == NULL) && !pcmk__str_eq(key, PCMK_XA_CRM_FEATURE_SET, pcmk__str_none)) { crm_trace("Passing %s=%s with fence action", (const char *) key, (const char *) (value? value : "")); pcmk__insert_dup((GHashTable *) user_data, key, pcmk__s(value, "")); } } /*! * \internal * \brief Create a table of arguments for a fencing action * * \param[in] agent Fencing agent name * \param[in] action Name of fencing action * \param[in] target Name of target node for fencing action - * \param[in] target_nodeid Node ID of target node for fencing action * \param[in] device_args Fence device parameters * \param[in] port_map Target node-to-port mapping for fence device * \param[in] host_arg Argument name for passing target * * \return Newly created hash table of arguments for fencing action */ static GHashTable * make_args(const char *agent, const char *action, const char *target, - uint32_t target_nodeid, GHashTable *device_args, - GHashTable *port_map, const char *host_arg) + GHashTable *device_args, GHashTable *port_map, const char *host_arg) { GHashTable *arg_list = NULL; const char *value = NULL; CRM_CHECK(action != NULL, return NULL); arg_list = pcmk__strkey_table(free, free); // Add action to arguments (using an alias if requested) if (device_args) { char buffer[512]; snprintf(buffer, sizeof(buffer), "pcmk_%s_action", action); value = g_hash_table_lookup(device_args, buffer); if (value) { crm_debug("Substituting '%s' for fence action %s targeting %s", value, action, pcmk__s(target, "no node")); action = value; } } // Tell the fence agent what action to perform pcmk__insert_dup(arg_list, STONITH_ATTR_ACTION_OP, action); /* If this is a fencing operation against another node, add more standard * arguments. */ if ((target != NULL) && (device_args != NULL)) { const char *param = NULL; /* Always pass the target's name, per * https://github.com/ClusterLabs/fence-agents/blob/main/doc/FenceAgentAPI.md */ pcmk__insert_dup(arg_list, "nodename", target); - // If the target's node ID was specified, pass it, too - if (target_nodeid != 0) { - char *nodeid = crm_strdup_printf("%" PRIu32, target_nodeid); - - // cts-fencing looks for this log message - crm_info("Passing '%s' as nodeid with fence action '%s' targeting %s", - nodeid, action, pcmk__s(target, "no node")); - g_hash_table_insert(arg_list, strdup("nodeid"), nodeid); - } - // Check whether target should be specified as some other argument param = g_hash_table_lookup(device_args, PCMK_STONITH_HOST_ARGUMENT); if (param == NULL) { // Use caller's default (likely from agent metadata) param = host_arg; } if ((param != NULL) && !pcmk__str_eq(agent, "fence_legacy", pcmk__str_none) && !pcmk__str_eq(param, PCMK_VALUE_NONE, pcmk__str_casei)) { value = g_hash_table_lookup(device_args, param); if (pcmk__str_eq(value, "dynamic", pcmk__str_casei|pcmk__str_null_matches)) { /* If the host argument is "dynamic" or not configured, * reset it to the target */ const char *alias = NULL; if (port_map) { alias = g_hash_table_lookup(port_map, target); } if (alias == NULL) { alias = target; } crm_debug("Passing %s='%s' with fence action %s targeting %s", param, alias, action, pcmk__s(target, "no node")); pcmk__insert_dup(arg_list, param, alias); } } } if (device_args) { g_hash_table_foreach(device_args, append_config_arg, arg_list); } return arg_list; } /*! * \internal * \brief Free all memory used by a stonith action * * \param[in,out] action Action to free */ void stonith__destroy_action(stonith_action_t *action) { if (action) { free(action->agent); if (action->args) { g_hash_table_destroy(action->args); } free(action->action); if (action->svc_action) { services_action_free(action->svc_action); } pcmk__reset_result(&(action->result)); free(action); } } /*! * \internal * \brief Get the result of an executed stonith action * * \param[in] action Executed action * * \return Pointer to action's result (or NULL if \p action is NULL) */ pcmk__action_result_t * stonith__action_result(stonith_action_t *action) { return (action == NULL)? NULL : &(action->result); } #define FAILURE_MAX_RETRIES 2 /*! * \internal * \brief Create a new fencing action to be executed * * \param[in] agent Fence agent to use * \param[in] action_name Fencing action to be executed * \param[in] target Name of target of fencing action (if known) - * \param[in] target_nodeid Node ID of target of fencing action (if known) * \param[in] timeout_sec Timeout to be used when executing action * \param[in] device_args Parameters to pass to fence agent * \param[in] port_map Mapping of target names to device ports * \param[in] host_arg Agent parameter used to pass target name * * \return Newly created fencing action (asserts on error, never NULL) */ stonith_action_t * stonith__action_create(const char *agent, const char *action_name, - const char *target, uint32_t target_nodeid, - int timeout_sec, GHashTable *device_args, - GHashTable *port_map, const char *host_arg) + const char *target, int timeout_sec, + GHashTable *device_args, GHashTable *port_map, + const char *host_arg) { stonith_action_t *action = pcmk__assert_alloc(1, sizeof(stonith_action_t)); - action->args = make_args(agent, action_name, target, target_nodeid, - device_args, port_map, host_arg); + action->args = make_args(agent, action_name, target, device_args, port_map, + host_arg); crm_debug("Preparing '%s' action targeting %s using agent %s", action_name, pcmk__s(target, "no node"), agent); action->agent = strdup(agent); action->action = strdup(action_name); action->timeout = action->remaining_timeout = timeout_sec; action->max_retries = FAILURE_MAX_RETRIES; pcmk__set_result(&(action->result), PCMK_OCF_UNKNOWN, PCMK_EXEC_UNKNOWN, "Initialization bug in fencing library"); if (device_args) { char buffer[512]; const char *value = NULL; snprintf(buffer, sizeof(buffer), "pcmk_%s_retries", action_name); value = g_hash_table_lookup(device_args, buffer); if (value) { action->max_retries = atoi(value); } } return action; } static gboolean update_remaining_timeout(stonith_action_t * action) { int diff = time(NULL) - action->initial_start_time; if (action->tries >= action->max_retries) { crm_info("Attempted to execute agent %s (%s) the maximum number of times (%d) allowed", action->agent, action->action, action->max_retries); action->remaining_timeout = 0; } else if ((action->result.execution_status != PCMK_EXEC_TIMEOUT) && (diff < (action->timeout * 0.7))) { /* only set remaining timeout period if there is 30% * or greater of the original timeout period left */ action->remaining_timeout = action->timeout - diff; } else { action->remaining_timeout = 0; } return action->remaining_timeout ? TRUE : FALSE; } /*! * \internal * \brief Map a fencing action result to a standard return code * * \param[in] result Fencing action result to map * * \return Standard Pacemaker return code that best corresponds to \p result */ int stonith__result2rc(const pcmk__action_result_t *result) { if (pcmk__result_ok(result)) { return pcmk_rc_ok; } switch (result->execution_status) { case PCMK_EXEC_PENDING: return EINPROGRESS; case PCMK_EXEC_CANCELLED: return ECANCELED; case PCMK_EXEC_TIMEOUT: return ETIME; case PCMK_EXEC_NOT_INSTALLED: return ENOENT; case PCMK_EXEC_NOT_SUPPORTED: return EOPNOTSUPP; case PCMK_EXEC_NOT_CONNECTED: return ENOTCONN; case PCMK_EXEC_NO_FENCE_DEVICE: return ENODEV; case PCMK_EXEC_NO_SECRETS: return EACCES; /* For the fencing API, PCMK_EXEC_INVALID is used with fencer API * operations that don't involve executing an agent (for example, * registering devices). This allows us to use the CRM_EX_* codes in the * exit status for finer-grained responses. */ case PCMK_EXEC_INVALID: switch (result->exit_status) { case CRM_EX_INVALID_PARAM: return EINVAL; case CRM_EX_INSUFFICIENT_PRIV: return EACCES; case CRM_EX_PROTOCOL: return EPROTO; /* CRM_EX_EXPIRED is used for orphaned fencing operations left * over from a previous instance of the fencer. For API backward * compatibility, this is mapped to the previously used code for * this case, EHOSTUNREACH. */ case CRM_EX_EXPIRED: return EHOSTUNREACH; default: break; } break; default: break; } // Try to provide useful error code based on result's error output if (result->action_stderr == NULL) { return ENODATA; } else if (strcasestr(result->action_stderr, "timed out") || strcasestr(result->action_stderr, "timeout")) { return ETIME; } else if (strcasestr(result->action_stderr, "unrecognised action") || strcasestr(result->action_stderr, "unrecognized action") || strcasestr(result->action_stderr, "unsupported action")) { return EOPNOTSUPP; } // Oh well, we tried return pcmk_rc_error; } /*! * \internal * \brief Determine execution status equivalent of legacy fencer return code * * Fence action notifications, and fence action callbacks from older fencers * (<=2.1.2) in a rolling upgrade, will have only a legacy return code. Map this * to an execution status as best as possible (essentially, the inverse of * stonith__result2rc()). * * \param[in] rc Legacy return code from fencer * * \return Execution status best corresponding to \p rc */ int stonith__legacy2status(int rc) { if (rc >= 0) { return PCMK_EXEC_DONE; } switch (-rc) { case EACCES: return PCMK_EXEC_NO_SECRETS; case ECANCELED: return PCMK_EXEC_CANCELLED; case EHOSTUNREACH: return PCMK_EXEC_INVALID; case EINPROGRESS: return PCMK_EXEC_PENDING; case ENODEV: return PCMK_EXEC_NO_FENCE_DEVICE; case ENOENT: return PCMK_EXEC_NOT_INSTALLED; case ENOTCONN: return PCMK_EXEC_NOT_CONNECTED; case EOPNOTSUPP: return PCMK_EXEC_NOT_SUPPORTED; case EPROTO: return PCMK_EXEC_INVALID; case EPROTONOSUPPORT: return PCMK_EXEC_NOT_SUPPORTED; case ETIME: return PCMK_EXEC_TIMEOUT; case ETIMEDOUT: return PCMK_EXEC_TIMEOUT; default: return PCMK_EXEC_ERROR; } } /*! * \internal * \brief Add a fencing result to an XML element as attributes * * \param[in,out] xml XML element to add result to * \param[in] result Fencing result to add (assume success if NULL) */ void stonith__xe_set_result(xmlNode *xml, const pcmk__action_result_t *result) { int exit_status = CRM_EX_OK; enum pcmk_exec_status execution_status = PCMK_EXEC_DONE; const char *exit_reason = NULL; const char *action_stdout = NULL; int rc = pcmk_ok; CRM_CHECK(xml != NULL, return); if (result != NULL) { exit_status = result->exit_status; execution_status = result->execution_status; exit_reason = result->exit_reason; action_stdout = result->action_stdout; rc = pcmk_rc2legacy(stonith__result2rc(result)); } crm_xml_add_int(xml, PCMK__XA_OP_STATUS, (int) execution_status); crm_xml_add_int(xml, PCMK__XA_RC_CODE, exit_status); crm_xml_add(xml, PCMK_XA_EXIT_REASON, exit_reason); crm_xml_add(xml, PCMK__XA_ST_OUTPUT, action_stdout); /* @COMPAT Peers in rolling upgrades, Pacemaker Remote nodes, and external * code that use libstonithd <=2.1.2 don't check for the full result, and * need a legacy return code instead. */ crm_xml_add_int(xml, PCMK__XA_ST_RC, rc); } /*! * \internal * \brief Find a fencing result beneath an XML element * * \param[in] xml XML element to search * * \return \p xml or descendant of it that contains a fencing result, else NULL */ xmlNode * stonith__find_xe_with_result(xmlNode *xml) { xmlNode *match = pcmk__xpath_find_one(xml->doc, "//*[@" PCMK__XA_RC_CODE "]", LOG_NEVER); if (match == NULL) { /* @COMPAT Peers <=2.1.2 in a rolling upgrade provide only a legacy * return code, not a full result, so check for that. */ match = pcmk__xpath_find_one(xml->doc, "//*[@" PCMK__XA_ST_RC "]", LOG_ERR); } return match; } /*! * \internal * \brief Get a fencing result from an XML element's attributes * * \param[in] xml XML element with fencing result * \param[out] result Where to store fencing result */ void stonith__xe_get_result(const xmlNode *xml, pcmk__action_result_t *result) { int exit_status = CRM_EX_OK; int execution_status = PCMK_EXEC_DONE; const char *exit_reason = NULL; char *action_stdout = NULL; CRM_CHECK((xml != NULL) && (result != NULL), return); exit_reason = crm_element_value(xml, PCMK_XA_EXIT_REASON); action_stdout = crm_element_value_copy(xml, PCMK__XA_ST_OUTPUT); // A result must include an exit status and execution status if ((crm_element_value_int(xml, PCMK__XA_RC_CODE, &exit_status) < 0) || (crm_element_value_int(xml, PCMK__XA_OP_STATUS, &execution_status) < 0)) { int rc = pcmk_ok; exit_status = CRM_EX_ERROR; /* @COMPAT Peers <=2.1.2 in rolling upgrades provide only a legacy * return code, not a full result, so check for that. */ if (crm_element_value_int(xml, PCMK__XA_ST_RC, &rc) == 0) { if ((rc == pcmk_ok) || (rc == -EINPROGRESS)) { exit_status = CRM_EX_OK; } execution_status = stonith__legacy2status(rc); exit_reason = pcmk_strerror(rc); } else { execution_status = PCMK_EXEC_ERROR; exit_reason = "Fencer reply contained neither a full result " "nor a legacy return code (bug?)"; } } pcmk__set_result(result, exit_status, execution_status, exit_reason); pcmk__set_result_output(result, action_stdout, NULL); } static void stonith_action_async_done(svc_action_t *svc_action) { stonith_action_t *action = (stonith_action_t *) svc_action->cb_data; set_result_from_svc_action(action, svc_action); svc_action->params = NULL; log_action(action, action->pid); if (!pcmk__result_ok(&(action->result)) && update_remaining_timeout(action)) { int rc = internal_stonith_action_execute(action); if (rc == pcmk_ok) { return; } } if (action->done_cb) { action->done_cb(action->pid, &(action->result), action->userdata); } action->svc_action = NULL; // don't remove our caller stonith__destroy_action(action); } static void stonith_action_async_forked(svc_action_t *svc_action) { stonith_action_t *action = (stonith_action_t *) svc_action->cb_data; action->pid = svc_action->pid; action->svc_action = svc_action; if (action->fork_cb) { (action->fork_cb) (svc_action->pid, action->userdata); } pcmk__set_result(&(action->result), PCMK_OCF_UNKNOWN, PCMK_EXEC_PENDING, NULL); crm_trace("Child process %d performing action '%s' successfully forked", action->pid, action->action); } /*! * \internal * \brief Convert a fencing library action to a services library action * * \param[in,out] action Fencing library action to convert * * \return Services library action equivalent to \p action on success; on error, * NULL will be returned and \p action's result will be set */ static svc_action_t * stonith_action_to_svc(stonith_action_t *action) { static int stonith_sequence = 0; char *path = crm_strdup_printf(PCMK__FENCE_BINDIR "/%s", action->agent); svc_action_t *svc_action = services_action_create_generic(path, NULL); free(path); if (svc_action->rc != PCMK_OCF_UNKNOWN) { set_result_from_svc_action(action, svc_action); services_action_free(svc_action); return NULL; } svc_action->timeout = action->remaining_timeout * 1000; svc_action->standard = pcmk__str_copy(PCMK_RESOURCE_CLASS_STONITH); svc_action->id = crm_strdup_printf("%s_%s_%dof%d", action->agent, action->action, action->tries, action->max_retries); svc_action->agent = pcmk__str_copy(action->agent); svc_action->sequence = stonith_sequence++; svc_action->params = action->args; svc_action->cb_data = (void *) action; svc_action->flags = pcmk__set_flags_as(__func__, __LINE__, LOG_TRACE, "Action", svc_action->id, svc_action->flags, SVC_ACTION_NON_BLOCKED, "SVC_ACTION_NON_BLOCKED"); return svc_action; } static int internal_stonith_action_execute(stonith_action_t * action) { int rc = pcmk_ok; int is_retry = 0; svc_action_t *svc_action = NULL; CRM_CHECK(action != NULL, return -EINVAL); if ((action->action == NULL) || (action->args == NULL) || (action->agent == NULL)) { pcmk__set_result(&(action->result), PCMK_OCF_UNKNOWN_ERROR, PCMK_EXEC_ERROR_FATAL, "Bug in fencing library"); return -EINVAL; } if (action->tries++ == 0) { // First attempt of the desired action action->initial_start_time = time(NULL); } else { // Later attempt after earlier failure crm_info("Attempt %d to execute '%s' action of agent %s " "(%ds timeout remaining)", action->tries, action->action, action->agent, action->remaining_timeout); is_retry = 1; } svc_action = stonith_action_to_svc(action); if (svc_action == NULL) { // The only possible errors are out-of-memory and too many arguments return -E2BIG; } /* keep retries from executing out of control and free previous results */ if (is_retry) { pcmk__reset_result(&(action->result)); // @TODO This should be nonblocking via timer if mainloop is used sleep(1); } if (action->async) { // We never create a recurring action, so this should always return TRUE CRM_LOG_ASSERT(services_action_async_fork_notify(svc_action, &stonith_action_async_done, &stonith_action_async_forked)); return pcmk_ok; } else if (!services_action_sync(svc_action)) { rc = -ECONNABORTED; // @TODO Update API to return more useful error } set_result_from_svc_action(action, svc_action); svc_action->params = NULL; services_action_free(svc_action); return rc; } /*! * \internal * \brief Kick off execution of an async stonith action * * \param[in,out] action Action to be executed * \param[in,out] userdata Datapointer to be passed to callbacks * \param[in] done Callback to notify action has failed/succeeded * \param[in] fork_callback Callback to notify successful fork of child * * \return pcmk_ok if ownership of action has been taken, -errno otherwise */ int stonith__execute_async(stonith_action_t * action, void *userdata, void (*done) (int pid, const pcmk__action_result_t *result, void *user_data), void (*fork_cb) (int pid, void *user_data)) { if (!action) { return -EINVAL; } action->userdata = userdata; action->done_cb = done; action->fork_cb = fork_cb; action->async = true; return internal_stonith_action_execute(action); } /*! * \internal * \brief Execute a stonith action * * \param[in,out] action Action to execute * * \return pcmk_ok on success, -errno otherwise */ int stonith__execute(stonith_action_t *action) { int rc = pcmk_ok; CRM_CHECK(action != NULL, return -EINVAL); // Keep trying until success, max retries, or timeout do { rc = internal_stonith_action_execute(action); } while ((rc != pcmk_ok) && update_remaining_timeout(action)); return rc; } diff --git a/lib/fencing/st_client.c b/lib/fencing/st_client.c index a2c541c366..52d348d27d 100644 --- a/lib/fencing/st_client.c +++ b/lib/fencing/st_client.c @@ -1,2733 +1,2733 @@ /* * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include // xmlNode #include // xmlXPathObject, etc. #include #include #include #include #include #include "fencing_private.h" CRM_TRACE_INIT_DATA(stonith); // Used as stonith_t:st_private typedef struct stonith_private_s { char *token; crm_ipc_t *ipc; mainloop_io_t *source; GHashTable *stonith_op_callback_table; GList *notify_list; int notify_refcnt; bool notify_deletes; void (*op_callback) (stonith_t * st, stonith_callback_data_t * data); } stonith_private_t; // Used as stonith_event_t:opaque struct event_private { pcmk__action_result_t result; }; typedef struct stonith_notify_client_s { const char *event; const char *obj_id; /* implement one day */ const char *obj_type; /* implement one day */ void (*notify) (stonith_t * st, stonith_event_t * e); bool delete; } stonith_notify_client_t; typedef struct stonith_callback_client_s { void (*callback) (stonith_t * st, stonith_callback_data_t * data); const char *id; void *user_data; gboolean only_success; gboolean allow_timeout_updates; struct timer_rec_s *timer; } stonith_callback_client_t; struct notify_blob_s { stonith_t *stonith; xmlNode *xml; }; struct timer_rec_s { int call_id; int timeout; guint ref; stonith_t *stonith; }; typedef int (*stonith_op_t) (const char *, int, const char *, xmlNode *, xmlNode *, xmlNode *, xmlNode **, xmlNode **); bool stonith_dispatch(stonith_t * st); xmlNode *stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options); static int stonith_send_command(stonith_t *stonith, const char *op, xmlNode *data, xmlNode **output_data, int call_options, int timeout); static void stonith_connection_destroy(gpointer user_data); static void stonith_send_notification(gpointer data, gpointer user_data); static int stonith_api_del_notification(stonith_t *stonith, const char *event); /*! * \brief Get agent namespace by name * * \param[in] namespace_s Name of namespace as string * * \return Namespace as enum value */ enum stonith_namespace stonith_text2namespace(const char *namespace_s) { if (pcmk__str_eq(namespace_s, "any", pcmk__str_null_matches)) { return st_namespace_any; } else if (!strcmp(namespace_s, "redhat") || !strcmp(namespace_s, "stonith-ng")) { return st_namespace_rhcs; } else if (!strcmp(namespace_s, "internal")) { return st_namespace_internal; } else if (!strcmp(namespace_s, "heartbeat")) { return st_namespace_lha; } return st_namespace_invalid; } /*! * \brief Get agent namespace name * * \param[in] namespace Namespace as enum value * * \return Namespace name as string */ const char * stonith_namespace2text(enum stonith_namespace st_namespace) { switch (st_namespace) { case st_namespace_any: return "any"; case st_namespace_rhcs: return "stonith-ng"; case st_namespace_internal: return "internal"; case st_namespace_lha: return "heartbeat"; default: break; } return "unsupported"; } /*! * \brief Determine namespace of a fence agent * * \param[in] agent Fence agent type * \param[in] namespace_s Name of agent namespace as string, if known * * \return Namespace of specified agent, as enum value */ enum stonith_namespace stonith_get_namespace(const char *agent, const char *namespace_s) { if (pcmk__str_eq(namespace_s, "internal", pcmk__str_none)) { return st_namespace_internal; } if (stonith__agent_is_rhcs(agent)) { return st_namespace_rhcs; } #if HAVE_STONITH_STONITH_H if (stonith__agent_is_lha(agent)) { return st_namespace_lha; } #endif return st_namespace_invalid; } gboolean stonith__watchdog_fencing_enabled_for_node_api(stonith_t *st, const char *node) { gboolean rv = FALSE; stonith_t *stonith_api = st?st:stonith_api_new(); char *list = NULL; if(stonith_api) { if (stonith_api->state == stonith_disconnected) { int rc = stonith_api->cmds->connect(stonith_api, "stonith-api", NULL); if (rc != pcmk_ok) { crm_err("Failed connecting to Stonith-API for watchdog-fencing-query."); } } if (stonith_api->state != stonith_disconnected) { /* caveat!!! * this might fail when when stonithd is just updating the device-list * probably something we should fix as well for other api-calls */ int rc = stonith_api->cmds->list(stonith_api, st_opt_sync_call, STONITH_WATCHDOG_ID, &list, 0); if ((rc != pcmk_ok) || (list == NULL)) { /* due to the race described above it can happen that * we drop in here - so as not to make remote nodes * panic on that answer */ if (rc == -ENODEV) { crm_notice("Cluster does not have watchdog fencing device"); } else { crm_warn("Could not check for watchdog fencing device: %s", pcmk_strerror(rc)); } } else if (list[0] == '\0') { rv = TRUE; } else { GList *targets = stonith__parse_targets(list); rv = pcmk__str_in_list(node, targets, pcmk__str_casei); g_list_free_full(targets, free); } free(list); if (!st) { /* if we're provided the api we still might have done the * connection - but let's assume the caller won't bother */ stonith_api->cmds->disconnect(stonith_api); } } if (!st) { stonith_api_delete(stonith_api); } } else { crm_err("Stonith-API for watchdog-fencing-query couldn't be created."); } crm_trace("Pacemaker assumes node %s %sto do watchdog-fencing.", node, rv?"":"not "); return rv; } gboolean stonith__watchdog_fencing_enabled_for_node(const char *node) { return stonith__watchdog_fencing_enabled_for_node_api(NULL, node); } /* when cycling through the list we don't want to delete items so just mark them and when we know nobody is using the list loop over it to remove the marked items */ static void foreach_notify_entry (stonith_private_t *private, GFunc func, gpointer user_data) { private->notify_refcnt++; g_list_foreach(private->notify_list, func, user_data); private->notify_refcnt--; if ((private->notify_refcnt == 0) && private->notify_deletes) { GList *list_item = private->notify_list; private->notify_deletes = FALSE; while (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; GList *next = g_list_next(list_item); if (list_client->delete) { free(list_client); private->notify_list = g_list_delete_link(private->notify_list, list_item); } list_item = next; } } } static void stonith_connection_destroy(gpointer user_data) { stonith_t *stonith = user_data; stonith_private_t *native = NULL; struct notify_blob_s blob; crm_trace("Sending destroyed notification"); blob.stonith = stonith; blob.xml = pcmk__xe_create(NULL, PCMK__XE_NOTIFY); native = stonith->st_private; native->ipc = NULL; native->source = NULL; free(native->token); native->token = NULL; stonith->state = stonith_disconnected; crm_xml_add(blob.xml, PCMK__XA_T, PCMK__VALUE_ST_NOTIFY); crm_xml_add(blob.xml, PCMK__XA_SUBT, PCMK__VALUE_ST_NOTIFY_DISCONNECT); foreach_notify_entry(native, stonith_send_notification, &blob); pcmk__xml_free(blob.xml); } xmlNode * create_device_registration_xml(const char *id, enum stonith_namespace standard, const char *agent, const stonith_key_value_t *params, const char *rsc_provides) { xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); xmlNode *args = pcmk__xe_create(data, PCMK__XE_ATTRIBUTES); #if HAVE_STONITH_STONITH_H if (standard == st_namespace_any) { standard = stonith_get_namespace(agent, NULL); } if (standard == st_namespace_lha) { hash2field((gpointer) "plugin", (gpointer) agent, args); agent = "fence_legacy"; } #endif crm_xml_add(data, PCMK_XA_ID, id); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK_XA_AGENT, agent); if ((standard != st_namespace_any) && (standard != st_namespace_invalid)) { crm_xml_add(data, PCMK__XA_NAMESPACE, stonith_namespace2text(standard)); } if (rsc_provides) { crm_xml_add(data, PCMK__XA_RSC_PROVIDES, rsc_provides); } for (; params; params = params->next) { hash2field((gpointer) params->key, (gpointer) params->value, args); } return data; } static int stonith_api_register_device(stonith_t *st, int call_options, const char *id, const char *namespace_s, const char *agent, const stonith_key_value_t *params) { int rc = 0; xmlNode *data = NULL; data = create_device_registration_xml(id, stonith_text2namespace(namespace_s), agent, params, NULL); rc = stonith_send_command(st, STONITH_OP_DEVICE_ADD, data, NULL, call_options, 0); pcmk__xml_free(data); return rc; } static int stonith_api_remove_device(stonith_t * st, int call_options, const char *name) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK_XA_ID, name); rc = stonith_send_command(st, STONITH_OP_DEVICE_DEL, data, NULL, call_options, 0); pcmk__xml_free(data); return rc; } static int stonith_api_remove_level_full(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level) { int rc = 0; xmlNode *data = NULL; CRM_CHECK(node || pattern || (attr && value), return -EINVAL); data = pcmk__xe_create(NULL, PCMK_XE_FENCING_LEVEL); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); if (node) { crm_xml_add(data, PCMK_XA_TARGET, node); } else if (pattern) { crm_xml_add(data, PCMK_XA_TARGET_PATTERN, pattern); } else { crm_xml_add(data, PCMK_XA_TARGET_ATTRIBUTE, attr); crm_xml_add(data, PCMK_XA_TARGET_VALUE, value); } crm_xml_add_int(data, PCMK_XA_INDEX, level); rc = stonith_send_command(st, STONITH_OP_LEVEL_DEL, data, NULL, options, 0); pcmk__xml_free(data); return rc; } static int stonith_api_remove_level(stonith_t * st, int options, const char *node, int level) { return stonith_api_remove_level_full(st, options, node, NULL, NULL, NULL, level); } /*! * \internal * \brief Create XML for fence topology level registration request * * \param[in] node If not NULL, target level by this node name * \param[in] pattern If not NULL, target by node name using this regex * \param[in] attr If not NULL, target by this node attribute * \param[in] value If not NULL, target by this node attribute value * \param[in] level Index number of level to register * \param[in] device_list List of devices in level * * \return Newly allocated XML tree on success, NULL otherwise * * \note The caller should set only one of node, pattern or attr/value. */ xmlNode * create_level_registration_xml(const char *node, const char *pattern, const char *attr, const char *value, int level, const stonith_key_value_t *device_list) { GString *list = NULL; xmlNode *data; CRM_CHECK(node || pattern || (attr && value), return NULL); data = pcmk__xe_create(NULL, PCMK_XE_FENCING_LEVEL); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add_int(data, PCMK_XA_ID, level); crm_xml_add_int(data, PCMK_XA_INDEX, level); if (node) { crm_xml_add(data, PCMK_XA_TARGET, node); } else if (pattern) { crm_xml_add(data, PCMK_XA_TARGET_PATTERN, pattern); } else { crm_xml_add(data, PCMK_XA_TARGET_ATTRIBUTE, attr); crm_xml_add(data, PCMK_XA_TARGET_VALUE, value); } for (; device_list; device_list = device_list->next) { pcmk__add_separated_word(&list, 1024, device_list->value, ","); } if (list != NULL) { crm_xml_add(data, PCMK_XA_DEVICES, (const char *) list->str); g_string_free(list, TRUE); } return data; } static int stonith_api_register_level_full(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level, const stonith_key_value_t *device_list) { int rc = 0; xmlNode *data = create_level_registration_xml(node, pattern, attr, value, level, device_list); CRM_CHECK(data != NULL, return -EINVAL); rc = stonith_send_command(st, STONITH_OP_LEVEL_ADD, data, NULL, options, 0); pcmk__xml_free(data); return rc; } static int stonith_api_register_level(stonith_t * st, int options, const char *node, int level, const stonith_key_value_t * device_list) { return stonith_api_register_level_full(st, options, node, NULL, NULL, NULL, level, device_list); } static int stonith_api_device_list(stonith_t *stonith, int call_options, const char *namespace_s, stonith_key_value_t **devices, int timeout) { int count = 0; enum stonith_namespace ns = stonith_text2namespace(namespace_s); if (devices == NULL) { crm_err("Parameter error: stonith_api_device_list"); return -EFAULT; } #if HAVE_STONITH_STONITH_H // Include Linux-HA agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_lha)) { count += stonith__list_lha_agents(devices); } #endif // Include Red Hat agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_rhcs)) { count += stonith__list_rhcs_agents(devices); } return count; } // See stonith_api_operations_t:metadata() documentation static int stonith_api_device_metadata(stonith_t *stonith, int call_options, const char *agent, const char *namespace_s, char **output, int timeout_sec) { /* By executing meta-data directly, we can get it from stonith_admin when * the cluster is not running, which is important for higher-level tools. */ enum stonith_namespace ns = stonith_get_namespace(agent, namespace_s); if (timeout_sec <= 0) { timeout_sec = PCMK_DEFAULT_ACTION_TIMEOUT_MS; } crm_trace("Looking up metadata for %s agent %s", stonith_namespace2text(ns), agent); switch (ns) { case st_namespace_rhcs: return stonith__rhcs_metadata(agent, timeout_sec, output); #if HAVE_STONITH_STONITH_H case st_namespace_lha: return stonith__lha_metadata(agent, timeout_sec, output); #endif default: crm_err("Can't get fence agent '%s' meta-data: No such agent", agent); break; } return -ENODEV; } static int stonith_api_query(stonith_t * stonith, int call_options, const char *target, stonith_key_value_t ** devices, int timeout) { int rc = 0, lpc = 0, max = 0; xmlNode *data = NULL; xmlNode *output = NULL; xmlXPathObject *xpathObj = NULL; CRM_CHECK(devices != NULL, return -EINVAL); data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, target); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, PCMK_ACTION_OFF); rc = stonith_send_command(stonith, STONITH_OP_QUERY, data, &output, call_options, timeout); if (rc < 0) { return rc; } xpathObj = pcmk__xpath_search(output->doc, "//*[@" PCMK_XA_AGENT "]"); if (xpathObj) { max = pcmk__xpath_num_results(xpathObj); for (lpc = 0; lpc < max; lpc++) { xmlNode *match = pcmk__xpath_result(xpathObj, lpc); CRM_LOG_ASSERT(match != NULL); if(match != NULL) { xmlChar *match_path = xmlGetNodePath(match); crm_info("//*[@" PCMK_XA_AGENT "][%d] = %s", lpc, match_path); free(match_path); *devices = stonith_key_value_add(*devices, NULL, crm_element_value(match, PCMK_XA_ID)); } } xmlXPathFreeObject(xpathObj); } pcmk__xml_free(output); pcmk__xml_free(data); return max; } /*! * \internal * \brief Make a STONITH_OP_EXEC request * * \param[in,out] stonith Fencer connection * \param[in] call_options Bitmask of \c stonith_call_options * \param[in] id Fence device ID that request is for * \param[in] action Agent action to request (list, status, monitor) * \param[in] target Name of target node for requested action * \param[in] timeout_sec Error if not completed within this many seconds * \param[out] output Where to set agent output */ static int stonith_api_call(stonith_t *stonith, int call_options, const char *id, const char *action, const char *target, int timeout_sec, xmlNode **output) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_ST_DEVICE_ID, id); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, action); crm_xml_add(data, PCMK__XA_ST_TARGET, target); rc = stonith_send_command(stonith, STONITH_OP_EXEC, data, output, call_options, timeout_sec); pcmk__xml_free(data); return rc; } static int stonith_api_list(stonith_t * stonith, int call_options, const char *id, char **list_info, int timeout) { int rc; xmlNode *output = NULL; rc = stonith_api_call(stonith, call_options, id, PCMK_ACTION_LIST, NULL, timeout, &output); if (output && list_info) { const char *list_str; list_str = crm_element_value(output, PCMK__XA_ST_OUTPUT); if (list_str) { *list_info = strdup(list_str); } } if (output) { pcmk__xml_free(output); } return rc; } static int stonith_api_monitor(stonith_t * stonith, int call_options, const char *id, int timeout) { return stonith_api_call(stonith, call_options, id, PCMK_ACTION_MONITOR, NULL, timeout, NULL); } static int stonith_api_status(stonith_t * stonith, int call_options, const char *id, const char *port, int timeout) { return stonith_api_call(stonith, call_options, id, PCMK_ACTION_STATUS, port, timeout, NULL); } static int stonith_api_fence_with_delay(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance, int delay) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, node); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, action); crm_xml_add_int(data, PCMK__XA_ST_TIMEOUT, timeout); crm_xml_add_int(data, PCMK__XA_ST_TOLERANCE, tolerance); crm_xml_add_int(data, PCMK__XA_ST_DELAY, delay); rc = stonith_send_command(stonith, STONITH_OP_FENCE, data, NULL, call_options, timeout); pcmk__xml_free(data); return rc; } static int stonith_api_fence(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance) { return stonith_api_fence_with_delay(stonith, call_options, node, action, timeout, tolerance, 0); } static int stonith_api_confirm(stonith_t * stonith, int call_options, const char *target) { stonith__set_call_options(call_options, target, st_opt_manual_ack); return stonith_api_fence(stonith, call_options, target, PCMK_ACTION_OFF, 0, 0); } static int stonith_api_history(stonith_t * stonith, int call_options, const char *node, stonith_history_t ** history, int timeout) { int rc = 0; xmlNode *data = NULL; xmlNode *output = NULL; stonith_history_t *last = NULL; *history = NULL; if (node) { data = pcmk__xe_create(NULL, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, node); } stonith__set_call_options(call_options, node, st_opt_sync_call); rc = stonith_send_command(stonith, STONITH_OP_FENCE_HISTORY, data, &output, call_options, timeout); pcmk__xml_free(data); if (rc == 0) { xmlNode *op = NULL; xmlNode *reply = pcmk__xpath_find_one(output->doc, "//" PCMK__XE_ST_HISTORY, LOG_NEVER); for (op = pcmk__xe_first_child(reply, NULL, NULL, NULL); op != NULL; op = pcmk__xe_next(op, NULL)) { stonith_history_t *kvp; long long completed; long long completed_nsec = 0L; kvp = pcmk__assert_alloc(1, sizeof(stonith_history_t)); kvp->target = crm_element_value_copy(op, PCMK__XA_ST_TARGET); kvp->action = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ACTION); kvp->origin = crm_element_value_copy(op, PCMK__XA_ST_ORIGIN); kvp->delegate = crm_element_value_copy(op, PCMK__XA_ST_DELEGATE); kvp->client = crm_element_value_copy(op, PCMK__XA_ST_CLIENTNAME); crm_element_value_ll(op, PCMK__XA_ST_DATE, &completed); kvp->completed = (time_t) completed; crm_element_value_ll(op, PCMK__XA_ST_DATE_NSEC, &completed_nsec); kvp->completed_nsec = completed_nsec; crm_element_value_int(op, PCMK__XA_ST_STATE, &kvp->state); kvp->exit_reason = crm_element_value_copy(op, PCMK_XA_EXIT_REASON); if (last) { last->next = kvp; } else { *history = kvp; } last = kvp; } } pcmk__xml_free(output); return rc; } void stonith_history_free(stonith_history_t *history) { stonith_history_t *hp, *hp_old; for (hp = history; hp; hp_old = hp, hp = hp->next, free(hp_old)) { free(hp->target); free(hp->action); free(hp->origin); free(hp->delegate); free(hp->client); free(hp->exit_reason); } } static gint stonithlib_GCompareFunc(gconstpointer a, gconstpointer b) { int rc = 0; const stonith_notify_client_t *a_client = a; const stonith_notify_client_t *b_client = b; if (a_client->delete || b_client->delete) { /* make entries marked for deletion not findable */ return -1; } CRM_CHECK(a_client->event != NULL && b_client->event != NULL, return 0); rc = strcmp(a_client->event, b_client->event); if (rc == 0) { if (a_client->notify == NULL || b_client->notify == NULL) { return 0; } else if (a_client->notify == b_client->notify) { return 0; } else if (((long)a_client->notify) < ((long)b_client->notify)) { crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return -1; } crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return 1; } return rc; } xmlNode * stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options) { xmlNode *op_msg = NULL; CRM_CHECK(token != NULL, return NULL); op_msg = pcmk__xe_create(NULL, PCMK__XE_STONITH_COMMAND); crm_xml_add(op_msg, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(op_msg, PCMK__XA_ST_OP, op); crm_xml_add_int(op_msg, PCMK__XA_ST_CALLID, call_id); crm_trace("Sending call options: %.8lx, %d", (long)call_options, call_options); crm_xml_add_int(op_msg, PCMK__XA_ST_CALLOPT, call_options); if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(op_msg, PCMK__XE_ST_CALLDATA); pcmk__xml_copy(wrapper, data); } return op_msg; } static void stonith_destroy_op_callback(gpointer data) { stonith_callback_client_t *blob = data; if (blob->timer && blob->timer->ref > 0) { g_source_remove(blob->timer->ref); } free(blob->timer); free(blob); } static int stonith_api_signoff(stonith_t * stonith) { stonith_private_t *native = stonith->st_private; crm_debug("Disconnecting from the fencer"); if (native->source != NULL) { /* Attached to mainloop */ mainloop_del_ipc_client(native->source); native->source = NULL; native->ipc = NULL; } else if (native->ipc) { /* Not attached to mainloop */ crm_ipc_t *ipc = native->ipc; native->ipc = NULL; crm_ipc_close(ipc); crm_ipc_destroy(ipc); } free(native->token); native->token = NULL; stonith->state = stonith_disconnected; return pcmk_ok; } static int stonith_api_del_callback(stonith_t * stonith, int call_id, bool all_callbacks) { stonith_private_t *private = stonith->st_private; if (all_callbacks) { private->op_callback = NULL; g_hash_table_destroy(private->stonith_op_callback_table); private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); } else if (call_id == 0) { private->op_callback = NULL; } else { pcmk__intkey_table_remove(private->stonith_op_callback_table, call_id); } return pcmk_ok; } /*! * \internal * \brief Invoke a (single) specified fence action callback * * \param[in,out] st Fencer API connection * \param[in] call_id If positive, call ID of completed fence action, * otherwise legacy return code for early failure * \param[in,out] result Full result for action * \param[in,out] userdata User data to pass to callback * \param[in] callback Fence action callback to invoke */ static void invoke_fence_action_callback(stonith_t *st, int call_id, pcmk__action_result_t *result, void *userdata, void (*callback) (stonith_t *st, stonith_callback_data_t *data)) { stonith_callback_data_t data = { 0, }; data.call_id = call_id; data.rc = pcmk_rc2legacy(stonith__result2rc(result)); data.userdata = userdata; data.opaque = (void *) result; callback(st, &data); } /*! * \internal * \brief Invoke any callbacks registered for a specified fence action result * * Given a fence action result from the fencer, invoke any callback registered * for that action, as well as any global callback registered. * * \param[in,out] stonith Fencer API connection * \param[in] msg If non-NULL, fencer reply * \param[in] call_id If \p msg is NULL, call ID of action that timed out */ static void invoke_registered_callbacks(stonith_t *stonith, const xmlNode *msg, int call_id) { stonith_private_t *private = NULL; stonith_callback_client_t *cb_info = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(stonith != NULL, return); CRM_CHECK(stonith->st_private != NULL, return); private = stonith->st_private; if (msg == NULL) { // Fencer didn't reply in time pcmk__set_result(&result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, "Fencer accepted request but did not reply in time"); CRM_LOG_ASSERT(call_id > 0); } else { // We have the fencer reply if ((crm_element_value_int(msg, PCMK__XA_ST_CALLID, &call_id) != 0) || (call_id <= 0)) { crm_log_xml_warn(msg, "Bad fencer reply"); } stonith__xe_get_result(msg, &result); } if (call_id > 0) { cb_info = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); } if ((cb_info != NULL) && (cb_info->callback != NULL) && (pcmk__result_ok(&result) || !(cb_info->only_success))) { crm_trace("Invoking callback %s for call %d", pcmk__s(cb_info->id, "without ID"), call_id); invoke_fence_action_callback(stonith, call_id, &result, cb_info->user_data, cb_info->callback); } else if ((private->op_callback == NULL) && !pcmk__result_ok(&result)) { crm_warn("Fencing action without registered callback failed: %d (%s%s%s)", result.exit_status, pcmk_exec_status_str(result.execution_status), ((result.exit_reason == NULL)? "" : ": "), ((result.exit_reason == NULL)? "" : result.exit_reason)); crm_log_xml_debug(msg, "Failed fence update"); } if (private->op_callback != NULL) { crm_trace("Invoking global callback for call %d", call_id); invoke_fence_action_callback(stonith, call_id, &result, NULL, private->op_callback); } if (cb_info != NULL) { stonith_api_del_callback(stonith, call_id, FALSE); } pcmk__reset_result(&result); } static gboolean stonith_async_timeout_handler(gpointer data) { struct timer_rec_s *timer = data; crm_err("Async call %d timed out after %dms", timer->call_id, timer->timeout); invoke_registered_callbacks(timer->stonith, NULL, timer->call_id); /* Always return TRUE, never remove the handler * We do that in stonith_del_callback() */ return TRUE; } static void set_callback_timeout(stonith_callback_client_t * callback, stonith_t * stonith, int call_id, int timeout) { struct timer_rec_s *async_timer = callback->timer; if (timeout <= 0) { return; } if (!async_timer) { async_timer = pcmk__assert_alloc(1, sizeof(struct timer_rec_s)); callback->timer = async_timer; } async_timer->stonith = stonith; async_timer->call_id = call_id; /* Allow a fair bit of grace to allow the server to tell us of a timeout * This is only a fallback */ async_timer->timeout = (timeout + 60) * 1000; if (async_timer->ref) { g_source_remove(async_timer->ref); } async_timer->ref = pcmk__create_timer(async_timer->timeout, stonith_async_timeout_handler, async_timer); } static void update_callback_timeout(int call_id, int timeout, stonith_t * st) { stonith_callback_client_t *callback = NULL; stonith_private_t *private = st->st_private; callback = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); if (!callback || !callback->allow_timeout_updates) { return; } set_callback_timeout(callback, st, call_id, timeout); } static int stonith_dispatch_internal(const char *buffer, ssize_t length, gpointer userdata) { const char *type = NULL; struct notify_blob_s blob; stonith_t *st = userdata; stonith_private_t *private = NULL; pcmk__assert(st != NULL); private = st->st_private; blob.stonith = st; blob.xml = pcmk__xml_parse(buffer); if (blob.xml == NULL) { crm_warn("Received malformed message from fencer: %s", buffer); return 0; } /* do callbacks */ type = crm_element_value(blob.xml, PCMK__XA_T); crm_trace("Activating %s callbacks...", type); if (pcmk__str_eq(type, PCMK__VALUE_STONITH_NG, pcmk__str_none)) { invoke_registered_callbacks(st, blob.xml, 0); } else if (pcmk__str_eq(type, PCMK__VALUE_ST_NOTIFY, pcmk__str_none)) { foreach_notify_entry(private, stonith_send_notification, &blob); } else if (pcmk__str_eq(type, PCMK__VALUE_ST_ASYNC_TIMEOUT_VALUE, pcmk__str_none)) { int call_id = 0; int timeout = 0; crm_element_value_int(blob.xml, PCMK__XA_ST_TIMEOUT, &timeout); crm_element_value_int(blob.xml, PCMK__XA_ST_CALLID, &call_id); update_callback_timeout(call_id, timeout, st); } else { crm_err("Unknown message type: %s", type); crm_log_xml_warn(blob.xml, "BadReply"); } pcmk__xml_free(blob.xml); return 1; } static int stonith_api_signon(stonith_t * stonith, const char *name, int *stonith_fd) { int rc = pcmk_ok; stonith_private_t *native = NULL; const char *display_name = name? name : "client"; struct ipc_client_callbacks st_callbacks = { .dispatch = stonith_dispatch_internal, .destroy = stonith_connection_destroy }; CRM_CHECK(stonith != NULL, return -EINVAL); native = stonith->st_private; pcmk__assert(native != NULL); crm_debug("Attempting fencer connection by %s with%s mainloop", display_name, (stonith_fd? "out" : "")); stonith->state = stonith_connected_command; if (stonith_fd) { /* No mainloop */ native->ipc = crm_ipc_new("stonith-ng", 0); if (native->ipc != NULL) { rc = pcmk__connect_generic_ipc(native->ipc); if (rc == pcmk_rc_ok) { rc = pcmk__ipc_fd(native->ipc, stonith_fd); if (rc != pcmk_rc_ok) { crm_debug("Couldn't get file descriptor for IPC: %s", pcmk_rc_str(rc)); } } if (rc != pcmk_rc_ok) { crm_ipc_close(native->ipc); crm_ipc_destroy(native->ipc); native->ipc = NULL; } } } else { /* With mainloop */ native->source = mainloop_add_ipc_client("stonith-ng", G_PRIORITY_MEDIUM, 0, stonith, &st_callbacks); native->ipc = mainloop_get_ipc_client(native->source); } if (native->ipc == NULL) { rc = -ENOTCONN; } else { xmlNode *reply = NULL; xmlNode *hello = pcmk__xe_create(NULL, PCMK__XE_STONITH_COMMAND); crm_xml_add(hello, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(hello, PCMK__XA_ST_OP, CRM_OP_REGISTER); crm_xml_add(hello, PCMK__XA_ST_CLIENTNAME, name); rc = crm_ipc_send(native->ipc, hello, crm_ipc_client_response, -1, &reply); if (rc < 0) { crm_debug("Couldn't register with the fencer: %s " QB_XS " rc=%d", pcmk_strerror(rc), rc); rc = -ECOMM; } else if (reply == NULL) { crm_debug("Couldn't register with the fencer: no reply"); rc = -EPROTO; } else { const char *msg_type = crm_element_value(reply, PCMK__XA_ST_OP); native->token = crm_element_value_copy(reply, PCMK__XA_ST_CLIENTID); if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_none)) { crm_debug("Couldn't register with the fencer: invalid reply type '%s'", (msg_type? msg_type : "(missing)")); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else if (native->token == NULL) { crm_debug("Couldn't register with the fencer: no token in reply"); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else { crm_debug("Connection to fencer by %s succeeded (registration token: %s)", display_name, native->token); rc = pcmk_ok; } } pcmk__xml_free(reply); pcmk__xml_free(hello); } if (rc != pcmk_ok) { crm_debug("Connection attempt to fencer by %s failed: %s " QB_XS " rc=%d", display_name, pcmk_strerror(rc), rc); stonith->cmds->disconnect(stonith); } return rc; } static int stonith_set_notification(stonith_t * stonith, const char *callback, int enabled) { int rc = pcmk_ok; xmlNode *notify_msg = pcmk__xe_create(NULL, __func__); stonith_private_t *native = stonith->st_private; if (stonith->state != stonith_disconnected) { crm_xml_add(notify_msg, PCMK__XA_ST_OP, STONITH_OP_NOTIFY); if (enabled) { crm_xml_add(notify_msg, PCMK__XA_ST_NOTIFY_ACTIVATE, callback); } else { crm_xml_add(notify_msg, PCMK__XA_ST_NOTIFY_DEACTIVATE, callback); } rc = crm_ipc_send(native->ipc, notify_msg, crm_ipc_client_response, -1, NULL); if (rc < 0) { crm_perror(LOG_DEBUG, "Couldn't register for fencing notifications: %d", rc); rc = -ECOMM; } else { rc = pcmk_ok; } } pcmk__xml_free(notify_msg); return rc; } static int stonith_api_add_notification(stonith_t * stonith, const char *event, void (*callback) (stonith_t * stonith, stonith_event_t * e)) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = NULL; private = stonith->st_private; crm_trace("Adding callback for %s events (%d)", event, g_list_length(private->notify_list)); new_client = pcmk__assert_alloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = callback; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); if (list_item != NULL) { crm_warn("Callback already present"); free(new_client); return -ENOTUNIQ; } else { private->notify_list = g_list_append(private->notify_list, new_client); stonith_set_notification(stonith, event, 1); crm_trace("Callback added (%d)", g_list_length(private->notify_list)); } return pcmk_ok; } static void del_notify_entry(gpointer data, gpointer user_data) { stonith_notify_client_t *entry = data; stonith_t * stonith = user_data; if (!entry->delete) { crm_debug("Removing callback for %s events", entry->event); stonith_api_del_notification(stonith, entry->event); } } static int stonith_api_del_notification(stonith_t * stonith, const char *event) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = stonith->st_private; if (event == NULL) { foreach_notify_entry(private, del_notify_entry, stonith); crm_trace("Removed callback"); return pcmk_ok; } crm_debug("Removing callback for %s events", event); new_client = pcmk__assert_alloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = NULL; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); stonith_set_notification(stonith, event, 0); if (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; if (private->notify_refcnt) { list_client->delete = TRUE; private->notify_deletes = TRUE; } else { private->notify_list = g_list_remove(private->notify_list, list_client); free(list_client); } crm_trace("Removed callback"); } else { crm_trace("Callback not present"); } free(new_client); return pcmk_ok; } static int stonith_api_add_callback(stonith_t * stonith, int call_id, int timeout, int options, void *user_data, const char *callback_name, void (*callback) (stonith_t * st, stonith_callback_data_t * data)) { stonith_callback_client_t *blob = NULL; stonith_private_t *private = NULL; CRM_CHECK(stonith != NULL, return -EINVAL); CRM_CHECK(stonith->st_private != NULL, return -EINVAL); private = stonith->st_private; if (call_id == 0) { // Add global callback private->op_callback = callback; } else if (call_id < 0) { // Call failed immediately, so call callback now if (!(options & st_opt_report_only_success)) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Call failed, calling %s: %s", callback_name, pcmk_strerror(call_id)); pcmk__set_result(&result, CRM_EX_ERROR, stonith__legacy2status(call_id), NULL); invoke_fence_action_callback(stonith, call_id, &result, user_data, callback); } else { crm_warn("Fencer call failed: %s", pcmk_strerror(call_id)); } return FALSE; } blob = pcmk__assert_alloc(1, sizeof(stonith_callback_client_t)); blob->id = callback_name; blob->only_success = (options & st_opt_report_only_success) ? TRUE : FALSE; blob->user_data = user_data; blob->callback = callback; blob->allow_timeout_updates = (options & st_opt_timeout_updates) ? TRUE : FALSE; if (timeout > 0) { set_callback_timeout(blob, stonith, call_id, timeout); } pcmk__intkey_table_insert(private->stonith_op_callback_table, call_id, blob); crm_trace("Added callback to %s for call %d", callback_name, call_id); return TRUE; } static void stonith_dump_pending_op(gpointer key, gpointer value, gpointer user_data) { int call = GPOINTER_TO_INT(key); stonith_callback_client_t *blob = value; crm_debug("Call %d (%s): pending", call, pcmk__s(blob->id, "no ID")); } void stonith_dump_pending_callbacks(stonith_t * stonith) { stonith_private_t *private = stonith->st_private; if (private->stonith_op_callback_table == NULL) { return; } return g_hash_table_foreach(private->stonith_op_callback_table, stonith_dump_pending_op, NULL); } /*! * \internal * \brief Get the data section of a fencer notification * * \param[in] msg Notification XML * \param[in] ntype Notification type */ static xmlNode * get_event_data_xml(xmlNode *msg, const char *ntype) { char *data_addr = crm_strdup_printf("//%s", ntype); xmlNode *data = pcmk__xpath_find_one(msg->doc, data_addr, LOG_DEBUG); free(data_addr); return data; } /* */ static stonith_event_t * xml_to_event(xmlNode *msg) { stonith_event_t *event = pcmk__assert_alloc(1, sizeof(stonith_event_t)); struct event_private *event_private = NULL; event->opaque = pcmk__assert_alloc(1, sizeof(struct event_private)); event_private = (struct event_private *) event->opaque; crm_log_xml_trace(msg, "stonith_notify"); // All notification types have the operation result and notification subtype stonith__xe_get_result(msg, &event_private->result); event->operation = crm_element_value_copy(msg, PCMK__XA_ST_OP); // @COMPAT The API originally provided the result as a legacy return code event->result = pcmk_rc2legacy(stonith__result2rc(&event_private->result)); // Some notification subtypes have additional information if (pcmk__str_eq(event->operation, PCMK__VALUE_ST_NOTIFY_FENCE, pcmk__str_none)) { xmlNode *data = get_event_data_xml(msg, event->operation); if (data == NULL) { crm_err("No data for %s event", event->operation); crm_log_xml_notice(msg, "BadEvent"); } else { event->origin = crm_element_value_copy(data, PCMK__XA_ST_ORIGIN); event->action = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ACTION); event->target = crm_element_value_copy(data, PCMK__XA_ST_TARGET); event->executioner = crm_element_value_copy(data, PCMK__XA_ST_DELEGATE); event->id = crm_element_value_copy(data, PCMK__XA_ST_REMOTE_OP); event->client_origin = crm_element_value_copy(data, PCMK__XA_ST_CLIENTNAME); event->device = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ID); } } else if (pcmk__str_any_of(event->operation, STONITH_OP_DEVICE_ADD, STONITH_OP_DEVICE_DEL, STONITH_OP_LEVEL_ADD, STONITH_OP_LEVEL_DEL, NULL)) { xmlNode *data = get_event_data_xml(msg, event->operation); if (data == NULL) { crm_err("No data for %s event", event->operation); crm_log_xml_notice(msg, "BadEvent"); } else { event->device = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ID); } } return event; } static void event_free(stonith_event_t * event) { struct event_private *event_private = event->opaque; free(event->id); free(event->operation); free(event->origin); free(event->action); free(event->target); free(event->executioner); free(event->device); free(event->client_origin); pcmk__reset_result(&event_private->result); free(event->opaque); free(event); } static void stonith_send_notification(gpointer data, gpointer user_data) { struct notify_blob_s *blob = user_data; stonith_notify_client_t *entry = data; stonith_event_t *st_event = NULL; const char *event = NULL; if (blob->xml == NULL) { crm_warn("Skipping callback - NULL message"); return; } event = crm_element_value(blob->xml, PCMK__XA_SUBT); if (entry == NULL) { crm_warn("Skipping callback - NULL callback client"); return; } else if (entry->delete) { crm_trace("Skipping callback - marked for deletion"); return; } else if (entry->notify == NULL) { crm_warn("Skipping callback - NULL callback"); return; } else if (!pcmk__str_eq(entry->event, event, pcmk__str_none)) { crm_trace("Skipping callback - event mismatch %p/%s vs. %s", entry, entry->event, event); return; } st_event = xml_to_event(blob->xml); crm_trace("Invoking callback for %p/%s event...", entry, event); entry->notify(blob->stonith, st_event); crm_trace("Callback invoked..."); event_free(st_event); } /*! * \internal * \brief Create and send an API request * * \param[in,out] stonith Stonith connection * \param[in] op API operation to request * \param[in] data Data to attach to request * \param[out] output_data If not NULL, will be set to reply if synchronous * \param[in] call_options Bitmask of stonith_call_options to use * \param[in] timeout Error if not completed within this many seconds * * \return pcmk_ok (for synchronous requests) or positive call ID * (for asynchronous requests) on success, -errno otherwise */ static int stonith_send_command(stonith_t * stonith, const char *op, xmlNode * data, xmlNode ** output_data, int call_options, int timeout) { int rc = 0; int reply_id = -1; xmlNode *op_msg = NULL; xmlNode *op_reply = NULL; stonith_private_t *native = NULL; pcmk__assert((stonith != NULL) && (stonith->st_private != NULL) && (op != NULL)); native = stonith->st_private; if (output_data != NULL) { *output_data = NULL; } if ((stonith->state == stonith_disconnected) || (native->token == NULL)) { return -ENOTCONN; } /* Increment the call ID, which must be positive to avoid conflicting with * error codes. This shouldn't be a problem unless the client mucked with * it or the counter wrapped around. */ stonith->call_id++; if (stonith->call_id < 1) { stonith->call_id = 1; } op_msg = stonith_create_op(stonith->call_id, native->token, op, data, call_options); if (op_msg == NULL) { return -EINVAL; } crm_xml_add_int(op_msg, PCMK__XA_ST_TIMEOUT, timeout); crm_trace("Sending %s message to fencer with timeout %ds", op, timeout); if (data) { const char *delay_s = crm_element_value(data, PCMK__XA_ST_DELAY); if (delay_s) { crm_xml_add(op_msg, PCMK__XA_ST_DELAY, delay_s); } } { enum crm_ipc_flags ipc_flags = crm_ipc_flags_none; if (call_options & st_opt_sync_call) { pcmk__set_ipc_flags(ipc_flags, "stonith command", crm_ipc_client_response); } rc = crm_ipc_send(native->ipc, op_msg, ipc_flags, 1000 * (timeout + 60), &op_reply); } pcmk__xml_free(op_msg); if (rc < 0) { crm_perror(LOG_ERR, "Couldn't perform %s operation (timeout=%ds): %d", op, timeout, rc); rc = -ECOMM; goto done; } crm_log_xml_trace(op_reply, "Reply"); if (!(call_options & st_opt_sync_call)) { crm_trace("Async call %d, returning", stonith->call_id); pcmk__xml_free(op_reply); return stonith->call_id; } crm_element_value_int(op_reply, PCMK__XA_ST_CALLID, &reply_id); if (reply_id == stonith->call_id) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Synchronous reply %d received", reply_id); stonith__xe_get_result(op_reply, &result); rc = pcmk_rc2legacy(stonith__result2rc(&result)); pcmk__reset_result(&result); if ((call_options & st_opt_discard_reply) || output_data == NULL) { crm_trace("Discarding reply"); } else { *output_data = op_reply; op_reply = NULL; /* Prevent subsequent free */ } } else if (reply_id <= 0) { crm_err("Received bad reply: No id set"); crm_log_xml_err(op_reply, "Bad reply"); pcmk__xml_free(op_reply); op_reply = NULL; rc = -ENOMSG; } else { crm_err("Received bad reply: %d (wanted %d)", reply_id, stonith->call_id); crm_log_xml_err(op_reply, "Old reply"); pcmk__xml_free(op_reply); op_reply = NULL; rc = -ENOMSG; } done: if (!crm_ipc_connected(native->ipc)) { crm_err("Fencer disconnected"); free(native->token); native->token = NULL; stonith->state = stonith_disconnected; } pcmk__xml_free(op_reply); return rc; } /* Not used with mainloop */ bool stonith_dispatch(stonith_t * st) { gboolean stay_connected = TRUE; stonith_private_t *private = NULL; pcmk__assert(st != NULL); private = st->st_private; while (crm_ipc_ready(private->ipc)) { if (crm_ipc_read(private->ipc) > 0) { const char *msg = crm_ipc_buffer(private->ipc); stonith_dispatch_internal(msg, strlen(msg), st); } if (!crm_ipc_connected(private->ipc)) { crm_err("Connection closed"); stay_connected = FALSE; } } return stay_connected; } static int stonith_api_free(stonith_t * stonith) { int rc = pcmk_ok; crm_trace("Destroying %p", stonith); if (stonith->state != stonith_disconnected) { crm_trace("Unregistering notifications and disconnecting %p first", stonith); stonith->cmds->remove_notification(stonith, NULL); rc = stonith->cmds->disconnect(stonith); } if (stonith->state == stonith_disconnected) { stonith_private_t *private = stonith->st_private; crm_trace("Removing %d callbacks", g_hash_table_size(private->stonith_op_callback_table)); g_hash_table_destroy(private->stonith_op_callback_table); crm_trace("Destroying %d notification clients", g_list_length(private->notify_list)); g_list_free_full(private->notify_list, free); free(stonith->st_private); free(stonith->cmds); free(stonith); } else { crm_err("Not free'ing active connection: %s (%d)", pcmk_strerror(rc), rc); } return rc; } void stonith_api_delete(stonith_t * stonith) { crm_trace("Destroying %p", stonith); if(stonith) { stonith->cmds->free(stonith); } } static gboolean is_stonith_param(gpointer key, gpointer value, gpointer user_data) { return pcmk_stonith_param(key); } int stonith__validate(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, GHashTable *params, int timeout_sec, char **output, char **error_output) { int rc = pcmk_rc_ok; /* Use a dummy node name in case the agent requires a target. We assume the * actual target doesn't matter for validation purposes (if in practice, * that is incorrect, we will need to allow the caller to pass the target). */ const char *target = "node1"; char *host_arg = NULL; if (params != NULL) { host_arg = pcmk__str_copy(g_hash_table_lookup(params, PCMK_STONITH_HOST_ARGUMENT)); /* Remove special stonith params from the table before doing anything else */ g_hash_table_foreach_remove(params, is_stonith_param, NULL); } #if PCMK__ENABLE_CIBSECRETS rc = pcmk__substitute_secrets(rsc_id, params); if (rc != pcmk_rc_ok) { crm_warn("Could not replace secret parameters for validation of %s: %s", agent, pcmk_rc_str(rc)); // rc is standard return value, don't return it in this function } #endif if (output) { *output = NULL; } if (error_output) { *error_output = NULL; } if (timeout_sec <= 0) { timeout_sec = PCMK_DEFAULT_ACTION_TIMEOUT_MS; } switch (stonith_get_namespace(agent, namespace_s)) { case st_namespace_rhcs: rc = stonith__rhcs_validate(st, call_options, target, agent, params, host_arg, timeout_sec, output, error_output); rc = pcmk_legacy2rc(rc); break; #if HAVE_STONITH_STONITH_H case st_namespace_lha: rc = stonith__lha_validate(st, call_options, target, agent, params, timeout_sec, output, error_output); rc = pcmk_legacy2rc(rc); break; #endif case st_namespace_invalid: errno = ENOENT; rc = errno; if (error_output) { *error_output = crm_strdup_printf("Agent %s not found", agent); } else { crm_err("Agent %s not found", agent); } break; default: errno = EOPNOTSUPP; rc = errno; if (error_output) { *error_output = crm_strdup_printf("Agent %s does not support validation", agent); } else { crm_err("Agent %s does not support validation", agent); } break; } free(host_arg); return rc; } static int stonith_api_validate(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, const stonith_key_value_t *params, int timeout_sec, char **output, char **error_output) { /* Validation should be done directly via the agent, so we can get it from * stonith_admin when the cluster is not running, which is important for * higher-level tools. */ int rc = pcmk_ok; GHashTable *params_table = pcmk__strkey_table(free, free); // Convert parameter list to a hash table for (; params; params = params->next) { if (!pcmk_stonith_param(params->key)) { pcmk__insert_dup(params_table, params->key, params->value); } } rc = stonith__validate(st, call_options, rsc_id, namespace_s, agent, params_table, timeout_sec, output, error_output); g_hash_table_destroy(params_table); return rc; } stonith_t * stonith_api_new(void) { stonith_t *new_stonith = NULL; stonith_private_t *private = NULL; new_stonith = calloc(1, sizeof(stonith_t)); if (new_stonith == NULL) { return NULL; } private = calloc(1, sizeof(stonith_private_t)); if (private == NULL) { free(new_stonith); return NULL; } new_stonith->st_private = private; private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); private->notify_list = NULL; private->notify_refcnt = 0; private->notify_deletes = FALSE; new_stonith->call_id = 1; new_stonith->state = stonith_disconnected; new_stonith->cmds = calloc(1, sizeof(stonith_api_operations_t)); if (new_stonith->cmds == NULL) { free(new_stonith->st_private); free(new_stonith); return NULL; } /* *INDENT-OFF* */ new_stonith->cmds->free = stonith_api_free; new_stonith->cmds->connect = stonith_api_signon; new_stonith->cmds->disconnect = stonith_api_signoff; new_stonith->cmds->list = stonith_api_list; new_stonith->cmds->monitor = stonith_api_monitor; new_stonith->cmds->status = stonith_api_status; new_stonith->cmds->fence = stonith_api_fence; new_stonith->cmds->fence_with_delay = stonith_api_fence_with_delay; new_stonith->cmds->confirm = stonith_api_confirm; new_stonith->cmds->history = stonith_api_history; new_stonith->cmds->list_agents = stonith_api_device_list; new_stonith->cmds->metadata = stonith_api_device_metadata; new_stonith->cmds->query = stonith_api_query; new_stonith->cmds->remove_device = stonith_api_remove_device; new_stonith->cmds->register_device = stonith_api_register_device; new_stonith->cmds->remove_level = stonith_api_remove_level; new_stonith->cmds->remove_level_full = stonith_api_remove_level_full; new_stonith->cmds->register_level = stonith_api_register_level; new_stonith->cmds->register_level_full = stonith_api_register_level_full; new_stonith->cmds->remove_callback = stonith_api_del_callback; new_stonith->cmds->register_callback = stonith_api_add_callback; new_stonith->cmds->remove_notification = stonith_api_del_notification; new_stonith->cmds->register_notification = stonith_api_add_notification; new_stonith->cmds->validate = stonith_api_validate; /* *INDENT-ON* */ return new_stonith; } /*! * \brief Make a blocking connection attempt to the fencer * * \param[in,out] st Fencer API object * \param[in] name Client name to use with fencer * \param[in] max_attempts Return error if this many attempts fail * * \return pcmk_ok on success, result of last attempt otherwise */ int stonith_api_connect_retry(stonith_t *st, const char *name, int max_attempts) { int rc = -EINVAL; // if max_attempts is not positive for (int attempt = 1; attempt <= max_attempts; attempt++) { rc = st->cmds->connect(st, name, NULL); if (rc == pcmk_ok) { return pcmk_ok; } else if (attempt < max_attempts) { crm_notice("Fencer connection attempt %d of %d failed (retrying in 2s): %s " QB_XS " rc=%d", attempt, max_attempts, pcmk_strerror(rc), rc); sleep(2); } } crm_notice("Could not connect to fencer: %s " QB_XS " rc=%d", pcmk_strerror(rc), rc); return rc; } stonith_key_value_t * stonith_key_value_add(stonith_key_value_t * head, const char *key, const char *value) { stonith_key_value_t *p, *end; p = pcmk__assert_alloc(1, sizeof(stonith_key_value_t)); p->key = pcmk__str_copy(key); p->value = pcmk__str_copy(value); end = head; while (end && end->next) { end = end->next; } if (end) { end->next = p; } else { head = p; } return head; } void stonith_key_value_freeall(stonith_key_value_t * head, int keys, int values) { stonith_key_value_t *p; while (head) { p = head->next; if (keys) { free(head->key); } if (values) { free(head->value); } free(head); head = p; } } #define api_log_open() openlog("stonith-api", LOG_CONS | LOG_NDELAY | LOG_PID, LOG_DAEMON) #define api_log(level, fmt, args...) syslog(level, "%s: "fmt, __func__, args) int stonith_api_kick(uint32_t nodeid, const char *uname, int timeout, bool off) { int rc = pcmk_ok; stonith_t *st = stonith_api_new(); const char *action = off? PCMK_ACTION_OFF : PCMK_ACTION_REBOOT; api_log_open(); if (st == NULL) { api_log(LOG_ERR, "API initialization failed, could not kick (%s) node %u/%s", action, nodeid, uname); return -EPROTO; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_ERR, "Connection failed, could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); int opts = 0; stonith__set_call_options(opts, name, st_opt_sync_call|st_opt_allow_self_fencing); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->fence(st, opts, name, action, timeout, 0); free(name); if (rc != pcmk_ok) { api_log(LOG_ERR, "Could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { api_log(LOG_NOTICE, "Node %u/%s kicked: %s", nodeid, uname, action); } } stonith_api_delete(st); return rc; } time_t stonith_api_time(uint32_t nodeid, const char *uname, bool in_progress) { int rc = pcmk_ok; time_t when = 0; stonith_t *st = stonith_api_new(); stonith_history_t *history = NULL, *hp = NULL; if (st == NULL) { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: " "API initialization failed", nodeid, uname); return when; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_NOTICE, "Connection failed: %s (%d)", pcmk_strerror(rc), rc); } else { int entries = 0; int progress = 0; int completed = 0; int opts = 0; char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); stonith__set_call_options(opts, name, st_opt_sync_call); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->history(st, opts, name, &history, 120); free(name); for (hp = history; hp; hp = hp->next) { entries++; if (in_progress) { progress++; if (hp->state != st_done && hp->state != st_failed) { when = time(NULL); } } else if (hp->state == st_done) { completed++; if (hp->completed > when) { when = hp->completed; } } } stonith_history_free(history); if(rc == pcmk_ok) { api_log(LOG_INFO, "Found %d entries for %u/%s: %d in progress, %d completed", entries, nodeid, uname, progress, completed); } else { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: %s (%d)", nodeid, uname, pcmk_strerror(rc), rc); } } stonith_api_delete(st); if(when) { api_log(LOG_INFO, "Node %u/%s last kicked at: %ld", nodeid, uname, (long int)when); } return when; } bool stonith_agent_exists(const char *agent, int timeout) { stonith_t *st = NULL; stonith_key_value_t *devices = NULL; stonith_key_value_t *dIter = NULL; bool rc = FALSE; if (agent == NULL) { return rc; } st = stonith_api_new(); if (st == NULL) { crm_err("Could not list fence agents: API memory allocation failed"); return FALSE; } st->cmds->list_agents(st, st_opt_sync_call, NULL, &devices, timeout == 0 ? 120 : timeout); for (dIter = devices; dIter != NULL; dIter = dIter->next) { if (pcmk__str_eq(dIter->value, agent, pcmk__str_none)) { rc = TRUE; break; } } stonith_key_value_freeall(devices, 1, 1); stonith_api_delete(st); return rc; } const char * stonith_action_str(const char *action) { if (action == NULL) { return "fencing"; } else if (strcmp(action, PCMK_ACTION_ON) == 0) { return "unfencing"; } else if (strcmp(action, PCMK_ACTION_OFF) == 0) { return "turning off"; } else { return action; } } /*! * \internal * \brief Parse a target name from one line of a target list string * * \param[in] line One line of a target list string * \param[in] len String length of line * \param[in,out] output List to add newly allocated target name to */ static void parse_list_line(const char *line, int len, GList **output) { size_t i = 0; size_t entry_start = 0; /* Skip complaints about additional parameters device doesn't understand * * @TODO Document or eliminate the implied restriction of target names */ if (strstr(line, "invalid") || strstr(line, "variable")) { crm_debug("Skipping list output line: %s", line); return; } // Process line content, character by character for (i = 0; i <= len; i++) { if (isspace(line[i]) || (line[i] == ',') || (line[i] == ';') || (line[i] == '\0')) { // We've found a separator (i.e. the end of an entry) int rc = 0; char *entry = NULL; if (i == entry_start) { // Skip leading and sequential separators entry_start = i + 1; continue; } entry = pcmk__assert_alloc(i - entry_start + 1, sizeof(char)); /* Read entry, stopping at first separator * * @TODO Document or eliminate these character restrictions */ rc = sscanf(line + entry_start, "%[a-zA-Z0-9_-.]", entry); if (rc != 1) { crm_warn("Could not parse list output entry: %s " QB_XS " entry_start=%d position=%d", line + entry_start, entry_start, i); free(entry); } else if (pcmk__strcase_any_of(entry, PCMK_ACTION_ON, PCMK_ACTION_OFF, NULL)) { /* Some agents print the target status in the list output, * though none are known now (the separate list-status command * is used for this, but it can also print "UNKNOWN"). To handle * this possibility, skip such entries. * * @TODO Document or eliminate the implied restriction of target * names. */ free(entry); } else { // We have a valid entry *output = g_list_append(*output, entry); } entry_start = i + 1; } } } /*! * \internal * \brief Parse a list of targets from a string * * \param[in] list_output Target list as a string * * \return List of target names * \note The target list string format is flexible, to allow for user-specified * lists such pcmk_host_list and the output of an agent's list action * (whether direct or via the API, which escapes newlines). There may be * multiple lines, separated by either a newline or an escaped newline * (backslash n). Each line may have one or more target names, separated * by any combination of whitespace, commas, and semi-colons. Lines * containing "invalid" or "variable" will be ignored entirely. Target * names "on" or "off" (case-insensitive) will be ignored. Target names * may contain only alphanumeric characters, underbars (_), dashes (-), * and dots (.) (if any other character occurs in the name, it and all * subsequent characters in the name will be ignored). * \note The caller is responsible for freeing the result with * g_list_free_full(result, free). */ GList * stonith__parse_targets(const char *target_spec) { GList *targets = NULL; if (target_spec != NULL) { size_t out_len = strlen(target_spec); size_t line_start = 0; // Starting index of line being processed for (size_t i = 0; i <= out_len; ++i) { if ((target_spec[i] == '\n') || (target_spec[i] == '\0') || ((target_spec[i] == '\\') && (target_spec[i + 1] == 'n'))) { // We've reached the end of one line of output int len = i - line_start; if (len > 0) { char *line = strndup(target_spec + line_start, len); line[len] = '\0'; // Because it might be a newline parse_list_line(line, len, &targets); free(line); } if (target_spec[i] == '\\') { ++i; // backslash-n takes up two positions } line_start = i + 1; } } } return targets; } /*! * \internal * \brief Check whether a fencing failure was followed by an equivalent success * * \param[in] event Fencing failure * \param[in] top_history Complete fencing history (must be sorted by * stonith__sort_history() beforehand) * * \return The name of the node that executed the fencing if a later successful * event exists, or NULL if no such event exists */ const char * stonith__later_succeeded(const stonith_history_t *event, const stonith_history_t *top_history) { const char *other = NULL; for (const stonith_history_t *prev_hp = top_history; prev_hp != NULL; prev_hp = prev_hp->next) { if (prev_hp == event) { break; } if ((prev_hp->state == st_done) && pcmk__str_eq(event->target, prev_hp->target, pcmk__str_casei) && pcmk__str_eq(event->action, prev_hp->action, pcmk__str_none) && ((event->completed < prev_hp->completed) || ((event->completed == prev_hp->completed) && (event->completed_nsec < prev_hp->completed_nsec)))) { if ((event->delegate == NULL) || pcmk__str_eq(event->delegate, prev_hp->delegate, pcmk__str_casei)) { // Prefer equivalent fencing by same executioner return prev_hp->delegate; } else if (other == NULL) { // Otherwise remember first successful executioner other = (prev_hp->delegate == NULL)? "some node" : prev_hp->delegate; } } } return other; } /*! * \internal * \brief Sort fencing history, pending first then by most recently completed * * \param[in,out] history List of stonith actions * * \return New head of sorted \p history */ stonith_history_t * stonith__sort_history(stonith_history_t *history) { stonith_history_t *new = NULL, *pending = NULL, *hp, *np, *tmp; for (hp = history; hp; ) { tmp = hp->next; if ((hp->state == st_done) || (hp->state == st_failed)) { /* sort into new */ if ((!new) || (hp->completed > new->completed) || ((hp->completed == new->completed) && (hp->completed_nsec > new->completed_nsec))) { hp->next = new; new = hp; } else { np = new; do { if ((!np->next) || (hp->completed > np->next->completed) || ((hp->completed == np->next->completed) && (hp->completed_nsec > np->next->completed_nsec))) { hp->next = np->next; np->next = hp; break; } np = np->next; } while (1); } } else { /* put into pending */ hp->next = pending; pending = hp; } hp = tmp; } /* pending actions don't have a completed-stamp so make them go front */ if (pending) { stonith_history_t *last_pending = pending; while (last_pending->next) { last_pending = last_pending->next; } last_pending->next = new; new = pending; } return new; } /*! * \brief Return string equivalent of an operation state value * * \param[in] state Fencing operation state value * * \return Human-friendly string equivalent of state */ const char * stonith_op_state_str(enum op_state state) { switch (state) { case st_query: return "querying"; case st_exec: return "executing"; case st_done: return "completed"; case st_duplicate: return "duplicate"; case st_failed: return "failed"; } return "unknown"; } stonith_history_t * stonith__first_matching_event(stonith_history_t *history, bool (*matching_fn)(stonith_history_t *, void *), void *user_data) { for (stonith_history_t *hp = history; hp; hp = hp->next) { if (matching_fn(hp, user_data)) { return hp; } } return NULL; } bool stonith__event_state_pending(stonith_history_t *history, void *user_data) { return history->state != st_failed && history->state != st_done; } bool stonith__event_state_eq(stonith_history_t *history, void *user_data) { return history->state == GPOINTER_TO_INT(user_data); } bool stonith__event_state_neq(stonith_history_t *history, void *user_data) { return history->state != GPOINTER_TO_INT(user_data); } void stonith__device_parameter_flags(uint32_t *device_flags, const char *device_name, xmlNode *metadata) { xmlXPathObject *xpath = NULL; int max = 0; int lpc = 0; CRM_CHECK((device_flags != NULL) && (metadata != NULL), return); xpath = pcmk__xpath_search(metadata->doc, "//" PCMK_XE_PARAMETER); max = pcmk__xpath_num_results(xpath); if (max == 0) { xmlXPathFreeObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *parameter = NULL; xmlNode *match = pcmk__xpath_result(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if (match == NULL) { continue; } parameter = crm_element_value(match, PCMK_XA_NAME); if (pcmk__str_eq(parameter, "plug", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_plug); } else if (pcmk__str_eq(parameter, "port", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_port); } } xmlXPathFreeObject(xpath); } /*! * \internal * \brief Retrieve fence agent meta-data asynchronously * * \param[in] agent Agent to execute * \param[in] timeout_sec Error if not complete within this time * \param[in] callback Function to call with result (this will always be * called, whether by this function directly or * later via the main loop, and on success the * metadata will be in its result argument's * action_stdout) * \param[in,out] user_data User data to pass to callback * * \return Standard Pacemaker return code * \note The caller must use a main loop. This function is not a * stonith_api_operations_t method because it does not need a stonith_t * object and does not go through the fencer, but executes the agent * directly. */ int stonith__metadata_async(const char *agent, int timeout_sec, void (*callback)(int pid, const pcmk__action_result_t *result, void *user_data), void *user_data) { switch (stonith_get_namespace(agent, NULL)) { case st_namespace_rhcs: { stonith_action_t *action = NULL; int rc = pcmk_ok; action = stonith__action_create(agent, PCMK_ACTION_METADATA, - NULL, 0, timeout_sec, NULL, - NULL, NULL); + NULL, timeout_sec, NULL, NULL, + NULL); rc = stonith__execute_async(action, user_data, callback, NULL); if (rc != pcmk_ok) { callback(0, stonith__action_result(action), user_data); stonith__destroy_action(action); } return pcmk_legacy2rc(rc); } #if HAVE_STONITH_STONITH_H case st_namespace_lha: // LHA metadata is simply synthesized, so simulate async { pcmk__action_result_t result = { .exit_status = CRM_EX_OK, .execution_status = PCMK_EXEC_DONE, .exit_reason = NULL, .action_stdout = NULL, .action_stderr = NULL, }; stonith__lha_metadata(agent, timeout_sec, &result.action_stdout); callback(0, &result, user_data); pcmk__reset_result(&result); return pcmk_rc_ok; } #endif default: { pcmk__action_result_t result = { .exit_status = CRM_EX_NOSUCH, .execution_status = PCMK_EXEC_ERROR_HARD, .exit_reason = crm_strdup_printf("No such agent '%s'", agent), .action_stdout = NULL, .action_stderr = NULL, }; callback(0, &result, user_data); pcmk__reset_result(&result); return ENOENT; } } } /*! * \internal * \brief Return the exit status from an async action callback * * \param[in] data Callback data * * \return Exit status from callback data */ int stonith__exit_status(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return CRM_EX_ERROR; } return ((pcmk__action_result_t *) data->opaque)->exit_status; } /*! * \internal * \brief Return the execution status from an async action callback * * \param[in] data Callback data * * \return Execution status from callback data */ int stonith__execution_status(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } return ((pcmk__action_result_t *) data->opaque)->execution_status; } /*! * \internal * \brief Return the exit reason from an async action callback * * \param[in] data Callback data * * \return Exit reason from callback data */ const char * stonith__exit_reason(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return NULL; } return ((pcmk__action_result_t *) data->opaque)->exit_reason; } /*! * \internal * \brief Return the exit status from an event notification * * \param[in] event Event * * \return Exit status from event */ int stonith__event_exit_status(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return CRM_EX_ERROR; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_status; } } /*! * \internal * \brief Return the execution status from an event notification * * \param[in] event Event * * \return Execution status from event */ int stonith__event_execution_status(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } else { struct event_private *event_private = event->opaque; return event_private->result.execution_status; } } /*! * \internal * \brief Return the exit reason from an event notification * * \param[in] event Event * * \return Exit reason from event */ const char * stonith__event_exit_reason(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return NULL; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_reason; } } /*! * \internal * \brief Return a human-friendly description of a fencing event * * \param[in] event Event to describe * * \return Newly allocated string with description of \p event * \note The caller is responsible for freeing the return value. * This function asserts on memory errors and never returns NULL. */ char * stonith__event_description(const stonith_event_t *event) { // Use somewhat readable defaults const char *origin = pcmk__s(event->client_origin, "a client"); const char *origin_node = pcmk__s(event->origin, "a node"); const char *executioner = pcmk__s(event->executioner, "the cluster"); const char *device = pcmk__s(event->device, "unknown"); const char *action = pcmk__s(event->action, event->operation); const char *target = pcmk__s(event->target, "no node"); const char *reason = stonith__event_exit_reason(event); const char *status; if (action == NULL) { action = "(unknown)"; } if (stonith__event_execution_status(event) != PCMK_EXEC_DONE) { status = pcmk_exec_status_str(stonith__event_execution_status(event)); } else if (stonith__event_exit_status(event) != CRM_EX_OK) { status = pcmk_exec_status_str(PCMK_EXEC_ERROR); } else { status = crm_exit_str(CRM_EX_OK); } if (pcmk__str_eq(event->operation, PCMK__VALUE_ST_NOTIFY_HISTORY, pcmk__str_none)) { return crm_strdup_printf("Fencing history may have changed"); } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_ADD, pcmk__str_none)) { return crm_strdup_printf("A fencing device (%s) was added", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_DEL, pcmk__str_none)) { return crm_strdup_printf("A fencing device (%s) was removed", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_ADD, pcmk__str_none)) { return crm_strdup_printf("A fencing topology level (%s) was added", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_DEL, pcmk__str_none)) { return crm_strdup_printf("A fencing topology level (%s) was removed", device); } // event->operation should be PCMK__VALUE_ST_NOTIFY_FENCE at this point return crm_strdup_printf("Operation %s of %s by %s for %s@%s: %s%s%s%s (ref=%s)", action, target, executioner, origin, origin_node, status, ((reason == NULL)? "" : " ("), pcmk__s(reason, ""), ((reason == NULL)? "" : ")"), pcmk__s(event->id, "(none)")); } diff --git a/lib/fencing/st_rhcs.c b/lib/fencing/st_rhcs.c index 6d0c6f9447..d091ea1153 100644 --- a/lib/fencing/st_rhcs.c +++ b/lib/fencing/st_rhcs.c @@ -1,330 +1,330 @@ /* * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include // xmlXPathObject, etc. #include #include #include #include #include "fencing_private.h" /*! * \internal * \brief \c scandir() filter for RHCS fence agents * * \param[in] entry Directory entry * * \retval 1 if \p entry is a regular file whose name begins with \c "fence_" * \retval 0 otherwise */ static int rhcs_agent_filter(const struct dirent *entry) { char *buf = NULL; struct stat sb; int rc = 0; if (!pcmk__starts_with(entry->d_name, "fence_")) { goto done; } // glibc doesn't enforce PATH_MAX, so don't limit buf size buf = crm_strdup_printf(PCMK__FENCE_BINDIR "/%s", entry->d_name); if ((stat(buf, &sb) != 0) || !S_ISREG(sb.st_mode)) { goto done; } rc = 1; done: free(buf); return rc; } /*! * \internal * \brief Add available RHCS-compatible agents to a list * * \param[in,out] List to add to * * \return Number of agents added */ int stonith__list_rhcs_agents(stonith_key_value_t **devices) { struct dirent **namelist = NULL; const int file_num = scandir(PCMK__FENCE_BINDIR, &namelist, rhcs_agent_filter, alphasort); if (file_num < 0) { int rc = errno; crm_err("Could not list " PCMK__FENCE_BINDIR ": %s", pcmk_rc_str(rc)); free(namelist); return 0; } for (int i = 0; i < file_num; i++) { *devices = stonith_key_value_add(*devices, NULL, namelist[i]->d_name); free(namelist[i]); } free(namelist); return file_num; } static void stonith_rhcs_parameter_not_required(xmlNode *metadata, const char *parameter) { char *xpath = NULL; xmlXPathObject *xpathObj = NULL; CRM_CHECK(metadata != NULL, return); CRM_CHECK(parameter != NULL, return); xpath = crm_strdup_printf("//" PCMK_XE_PARAMETER "[@" PCMK_XA_NAME "='%s']", parameter); /* Fudge metadata so that the parameter isn't required in config * Pacemaker handles and adds it */ xpathObj = pcmk__xpath_search(metadata->doc, xpath); if (pcmk__xpath_num_results(xpathObj) > 0) { xmlNode *tmp = pcmk__xpath_result(xpathObj, 0); if (tmp != NULL) { crm_xml_add(tmp, "required", "0"); } } xmlXPathFreeObject(xpathObj); free(xpath); } /*! * \brief Execute RHCS-compatible agent's metadata action * * \param[in] agent Agent to execute * \param[in] timeout_sec Action timeout * \param[out] metadata Where to store output xmlNode (or NULL to ignore) */ static int stonith__rhcs_get_metadata(const char *agent, int timeout_sec, xmlNode **metadata) { xmlNode *xml = NULL; xmlNode *actions = NULL; xmlXPathObject *xpathObj = NULL; stonith_action_t *action = stonith__action_create(agent, PCMK_ACTION_METADATA, - NULL, 0, timeout_sec, - NULL, NULL, NULL); + NULL, timeout_sec, NULL, + NULL, NULL); int rc = stonith__execute(action); pcmk__action_result_t *result = stonith__action_result(action); if (result == NULL) { if (rc < 0) { crm_warn("Could not execute metadata action for %s: %s " QB_XS " rc=%d", agent, pcmk_strerror(rc), rc); } stonith__destroy_action(action); return rc; } if (result->execution_status != PCMK_EXEC_DONE) { crm_warn("Could not execute metadata action for %s: %s", agent, pcmk_exec_status_str(result->execution_status)); rc = pcmk_rc2legacy(stonith__result2rc(result)); stonith__destroy_action(action); return rc; } if (!pcmk__result_ok(result)) { crm_warn("Metadata action for %s returned error code %d", agent, result->exit_status); rc = pcmk_rc2legacy(stonith__result2rc(result)); stonith__destroy_action(action); return rc; } if (result->action_stdout == NULL) { crm_warn("Metadata action for %s returned no data", agent); stonith__destroy_action(action); return -ENODATA; } xml = pcmk__xml_parse(result->action_stdout); stonith__destroy_action(action); if (xml == NULL) { crm_warn("Metadata for %s is invalid", agent); return -pcmk_err_schema_validation; } xpathObj = pcmk__xpath_search(xml->doc, "//" PCMK_XE_ACTIONS); if (pcmk__xpath_num_results(xpathObj) > 0) { actions = pcmk__xpath_result(xpathObj, 0); } xmlXPathFreeObject(xpathObj); // Add start and stop (implemented by pacemaker, not agent) to meta-data xpathObj = pcmk__xpath_search(xml->doc, "//" PCMK_XE_ACTION "[@" PCMK_XA_NAME "='" PCMK_ACTION_STOP "']"); if (pcmk__xpath_num_results(xpathObj) == 0) { xmlNode *tmp = NULL; const char *timeout_str = NULL; timeout_str = pcmk__readable_interval(PCMK_DEFAULT_ACTION_TIMEOUT_MS); tmp = pcmk__xe_create(actions, PCMK_XE_ACTION); crm_xml_add(tmp, PCMK_XA_NAME, PCMK_ACTION_STOP); crm_xml_add(tmp, PCMK_META_TIMEOUT, timeout_str); tmp = pcmk__xe_create(actions, PCMK_XE_ACTION); crm_xml_add(tmp, PCMK_XA_NAME, PCMK_ACTION_START); crm_xml_add(tmp, PCMK_META_TIMEOUT, timeout_str); } xmlXPathFreeObject(xpathObj); // Fudge metadata so parameters are not required in config (pacemaker adds them) stonith_rhcs_parameter_not_required(xml, STONITH_ATTR_ACTION_OP); stonith_rhcs_parameter_not_required(xml, "plug"); stonith_rhcs_parameter_not_required(xml, "port"); if (metadata) { *metadata = xml; } else { pcmk__xml_free(xml); } return pcmk_ok; } /*! * \brief Retrieve metadata for RHCS-compatible fence agent * * \param[in] agent Agent to execute * \param[in] timeout_sec Action timeout * \param[out] output Where to store action output (or NULL to ignore) */ int stonith__rhcs_metadata(const char *agent, int timeout_sec, char **output) { GString *buffer = NULL; xmlNode *xml = NULL; int rc = stonith__rhcs_get_metadata(agent, timeout_sec, &xml); if (rc != pcmk_ok) { goto done; } buffer = g_string_sized_new(1024); pcmk__xml_string(xml, pcmk__xml_fmt_pretty|pcmk__xml_fmt_text, buffer, 0); if (pcmk__str_empty(buffer->str)) { rc = -pcmk_err_schema_validation; goto done; } if (output != NULL) { pcmk__str_update(output, buffer->str); } done: if (buffer != NULL) { g_string_free(buffer, TRUE); } pcmk__xml_free(xml); return rc; } bool stonith__agent_is_rhcs(const char *agent) { struct stat prop; char *buffer = crm_strdup_printf(PCMK__FENCE_BINDIR "/%s", agent); int rc = stat(buffer, &prop); free(buffer); return (rc >= 0) && S_ISREG(prop.st_mode); } int stonith__rhcs_validate(stonith_t *st, int call_options, const char *target, const char *agent, GHashTable *params, const char * host_arg, int timeout, char **output, char **error_output) { int rc = pcmk_ok; int remaining_timeout = timeout; xmlNode *metadata = NULL; stonith_action_t *action = NULL; pcmk__action_result_t *result = NULL; if (host_arg == NULL) { time_t start_time = time(NULL); rc = stonith__rhcs_get_metadata(agent, remaining_timeout, &metadata); if (rc == pcmk_ok) { uint32_t device_flags = 0; stonith__device_parameter_flags(&device_flags, agent, metadata); if (pcmk_is_set(device_flags, st_device_supports_parameter_port)) { host_arg = "port"; } else if (pcmk_is_set(device_flags, st_device_supports_parameter_plug)) { host_arg = "plug"; } } pcmk__xml_free(metadata); remaining_timeout -= time(NULL) - start_time; if (rc == -ETIME || remaining_timeout <= 0 ) { return -ETIME; } } else if (pcmk__str_eq(host_arg, PCMK_VALUE_NONE, pcmk__str_casei)) { host_arg = NULL; } - action = stonith__action_create(agent, PCMK_ACTION_VALIDATE_ALL, target, 0, + action = stonith__action_create(agent, PCMK_ACTION_VALIDATE_ALL, target, remaining_timeout, params, NULL, host_arg); rc = stonith__execute(action); result = stonith__action_result(action); if (result != NULL) { rc = pcmk_rc2legacy(stonith__result2rc(result)); // Take ownership of output so stonith__destroy_action() doesn't free it if (output != NULL) { *output = result->action_stdout; result->action_stdout = NULL; } if (error_output != NULL) { *error_output = result->action_stderr; result->action_stderr = NULL; } } stonith__destroy_action(action); return rc; }