Page MenuHomeClusterLabs Projects
Files F4639205

No OneTemporary
Actions

View Options

diff --git a/heartbeat/Raid1.in b/heartbeat/Raid1.in
index db478e797..83d6fe83c 100644
--- a/heartbeat/Raid1.in
+++ b/heartbeat/Raid1.in
@@ -1,331 +1,420 @@
#!/bin/sh
#
-# $Id: Raid1.in,v 1.2 2005/07/29 07:36:15 sunjd Exp $
+# $Id: Raid1.in,v 1.3 2005/08/03 05:53:38 sunjd Exp $
#
# License: GNU General Public License (GPL)
# Support: linux-ha@lists.linux-ha.org
#
# Raid1
# Description: Manages a software Raid1 device on a shared storage medium.
# Original Author: Eric Z. Ayers (eric.ayers@compgen.com)
# Original Release: 25 Oct 2000
-# Support: linux-ha-dev@lists.tummy.com
# RAID patches: http://people.redhat.com/mingo/raid-patches/
# Word to the Wise: http://lwn.net/2000/0810/a/raid-faq.php3
# Sympathetic Ear: mailto:linux-raid@vger.kernel.org
#
-# usage: $0 {start|stop|status|monitor|meta-data}
+# usage: $0 {start|stop|status|monitor|validate-all|usage|meta-data}
#
# OCF parameters are as below:
# OCF_RESKEY_raidconf
# (name of MD configuration file. e.g. /etc/raidtab)
# OCF_RESKEY_raiddev
# (of the form /dev/md?? the block device to use)
#
# in /etc/ha.d/haresources, use a line such as:
# nodea 10.0.0.170 Raid1::/etc/raidtab.md0::/dev/md0 Filesystem::/dev/md0::/data1::ext2
#
-# This script assumes you are running the so-called RAID v.90 patches vs.
-# the Linux 2.2 kernel (distributed with RedHat 6.2). I have not used
-# kernel version 2.4.
-#
# The "start" arg starts up the raid device
# The "stop" arg stops it. NOTE: all filesystems must be unmounted
# and no processes should be accessing the device.
# The "status" arg just prints out whether the device is running or not
#
#
# DISCLAIMER: Use at your own risk!
#
# Besides all of the usual legalese that accompanies free software,
# I will warn you that I do not yet use this kind of setup (software RAID
# over shared storage) in production, and I have reservations about doing so.
#
# The linux md driver/scsi drivers under Raid 0.90 and kernel version 2.2
# do not behave well when a drive is in the process of going bad.
# The kernel slows down, but doesn't completely crash. This is about the
# worst possible thing that could happen in an un-attended HA type
# environment. (Once the system is rebooted, the sofware raid stuff works
# like a champ.)
# My other reservation has to do with the interation of RAID recovery with
# journaling filesystems and other parts of the kernel. Subscribe to
# linux-raid@vger.kernel.org for other opinions and possible solutions.
#
# -EZA 25 Oct 2000
#
# SETUP:
#
# You might need to pass the command line parameter: raid=noautodetect
# in an HA environment so that the kernel doesn't automatically start
# up your raid partitions when you boot the node. This means that it isn't
# going to work to use RAID for the system disks and the shared disks.
#
# 0) partition the disks to use for RAID. Use normal Linux partition
# types, not the RAID autodetect type for your partitions.
# 1) Create /etc/raidtab.md? on both systems (see example file below)
# 2) Initialize your raid partition with
# /sbin/mkraid --configfile /etc/raidtab.md? /dev/md?
# 3) Format your filesystem
# mke2fs /dev/md0 # for ext2fs... a journaling filesystem would be nice
# 3) Create the mount point on both systems.
# DO NOT add your raid filesystem to /etc/fstab
# 4) copy this script (to /etc/rc.d/init.d if you wish) and edit it to
# reflect your desired settings.
# 5) Modify the heartbeat 'haresources' setup file
# 6) unmount the filesystem and stop the raid device with 'raidstop'
# 7) fire up heartbeat!
#
#
# EXAMPLE config file /etc/raidtab.md0
# This file must exist on both machines!
#
# raiddev /dev/md0
# raid-level 1
# nr-raid-disks 2
# chunk-size 64k
# persistent-superblock 1
# #nr-spare-disks 0
# device /dev/sda1
# raid-disk 0
# device /dev/sdb1
# raid-disk 1
#
#######################################################################
# Initialization:
. @hb_libdir@/ocf-shellfuncs
#######################################################################
prefix=@prefix@
exec_prefix=@exec_prefix@
# Utilities used by this script
MODPROBE=@MODPROBE@
FSCK=@FSCK@
FUSER=@FUSER@
RAIDSTART=@RAIDSTART@
MOUNT=@MOUNT@
UMOUNT=@UMOUNT@
RAIDSTOP=@RAIDSTOP@
+MDADM=@MDADM@
check_util () {
if [ ! -x "$1" ] ; then
- ocf_log "err" "setup problem: Couldn't find utility $1"
- exit 1
+ ocf_log err "setup problem: Couldn't find utility $1"
+ exit $OCF_ERR_GENERIC
fi
}
usage() {
cat <<-EOT;
- usage: $0 {start|stop|status|monitor|meta-data}
- $Id: Raid1.in,v 1.2 2005/07/29 07:36:15 sunjd Exp $
+ usage: $0 {start|stop|status|monitor|validate-all|usage|meta-data}
+ $Id: Raid1.in,v 1.3 2005/08/03 05:53:38 sunjd Exp $
EOT
}
meta_data() {
cat <<END
<?xml version="1.0"?>
<!DOCTYPE resource-agent SYSTEM "ra-api-1.dtd">
-<resource-agent name="RAID1" version="0.9">
+<resource-agent name="Raid1">
<version>1.0</version>
<longdesc lang="en">
Resource script for RAID1. It manages a software Raid1 device on a shared
storage medium.
</longdesc>
<shortdesc lang="en">RAID1 resource agent</shortdesc>
<parameters>
<parameter name="raidconf" unique="0">
<longdesc lang="en">
The name of RAID configuration file. e.g. /etc/raidtab.
</longdesc>
<shortdesc lang="en">RAID tab config file</shortdesc>
<content type="string" default="" />
</parameter>
<parameter name="raiddev" unique="0">
<longdesc lang="en">
The block device to use.
</longdesc>
<shortdesc lang="en">block device</shortdesc>
<content type="string" default="" />
</parameter>
</parameters>
<actions>
<action name="start" timeout="10" />
<action name="stop" timeout="10" />
<action name="status" depth="0" timeout="10" interval="10" start-delay="10" />
<action name="monitor" depth="0" timeout="10" interval="10" start-delay="10" />
+<action name="validate-all" timeout="5" />
<action name="meta-data" timeout="5" />
</actions>
</resource-agent>
END
}
#
# START: Start up the RAID device
#
raid1_start() {
# See if the md device is already mounted.
- # NOTE: this will not work right if you have more than 10 md devices!
- $MOUNT | grep -e "^$MDDEV" >/dev/null
+ $MOUNT | grep -e "^$MDDEV\>" >/dev/null
if [ $? -ne 1 ] ; then
- ocf_log "err" "Device $MDDEV is already mounted!"
- return 1;
+ ocf_log err "Device $MDDEV is already mounted!"
+ return $OCF_ERR_GENERIC
fi
+ if [ "running" = `raid1_status` ]; then
+ # We are already online, do not bother
+ return $OCF_SUCCESS
+ fi
+
# Insert SCSI module
$MODPROBE scsi_hostadapter
if [ $? -ne 0 ] ; then
- ocf_log "warn" "Couldn't insert SCSI module."
+ ocf_log warn "Couldn't insert SCSI module."
fi
# Insert raid personality module
$MODPROBE raid1
if [ $? -ne 0 ] ; then
- ocf_log "err" "Couldn't insert RAID1 module"
- return 1
+ # It is not fatal, chance is that we have raid1 builtin...
+ ocf_log warn "Couldn't insert RAID1 module"
+# return 1
fi
-
- # Run raidstart to start up the RAID array
- $RAIDSTART --configfile $RAIDTAB_CONFIG $MDDEV
+ grep -q "^Personalities.*\[raid1\]" /proc/mdstat 2>/dev/null
if [ $? -ne 0 ] ; then
- ocf_log "err" "Couldn't start RAID for $MDDEV"
- return 1
+ ocf_log err "We don't have RAID1 support! Exiting"
+ return $OCF_ERR_GENERIC
fi
- return 0
+ if [ $HAVE_RAIDTOOLS = "true" ]; then
+ # Run raidstart to start up the RAID array
+ $RAIDSTART --configfile $RAIDTAB_CONFIG $MDDEV
+ else
+ # Run mdadm
+ $MDADM --run $MDDEV --config=$RAIDTAB_CONFIG
+ fi
+
+
+ # We can't rely the exit status above, $MDADM will return 0 on some failures, see
+ #
+ # # mdadm --run /dev/nosuchdevice
+ # mdadm: error opening /dev/nosuchdevice: No such file or directory
+ # # echo $?
+ # 0
+
+ if [ "running" = `raid1_status` ]; then
+ return $OCF_SUCCESS
+ else
+ ocf_log err "Couldn't start RAID for $MDDEV"
+ return $OCF_ERR_GENERIC
+ fi
}
#
# STOP: stop the RAID device
#
raid1_stop() {
# See if the MD device is online
- grep -e "^$MD" /proc/mdstat >/dev/null
+ grep -e "^$MD[ \t:]" /proc/mdstat >/dev/null
if [ $? -ne 0 ] ; then
- ocf_log "warn" "device $MD is not online according to kernel"
- return 0
+ ocf_log warn "device $MD is not online according to kernel"
+ return $OCF_SUCCESS
fi
# See if the MD device is mounted
# NOTE: this will not work right if you have more than 10 md devices!
- $MOUNT | grep -e "^$MDDEV" >/dev/null
+ $MOUNT | grep -e "^$MDDEV\>" >/dev/null
if [ $? -ne 1 ] ; then
# Kill all processes open on filesystem
$FUSER -mk $MOUNTPOINT
# the return from fuser doesn't tell us much
#if [ $? -ne 0 ] ; then
# ocf_log "err" "Couldn't kill processes on $MOUNTPOINT"
# return 1;
#fi
# Unmount the filesystem
$UMOUNT $MDDEV
if [ $? -ne 0 ] ; then
- ocf_log "err" "Couldn't unmount filesystem for $MDDEV"
- return 1
+ ocf_log err "Couldn't unmount filesystem for $MDDEV"
+ return $OCF_ERR_GENERIC
fi
- $MOUNT | grep -e "^$MDDEV" >/dev/null
+ $MOUNT | grep -e "^$MDDEV\>" >/dev/null
if [ $? -ne 1 ] ; then
- ocf_log "err" "filesystem for $MDDEV still mounted"
- return 1
+ ocf_log err "filesystem for $MDDEV still mounted"
+ return $OCF_ERR_GENERIC
fi
fi
# Turn off raid
- $RAIDSTOP --configfile /etc/raidtab.$MD $MDDEV
+ if [ $HAVE_RAIDTOOLS = "true" ]; then
+ $RAIDSTOP --configfile /etc/raidtab.$MD $MDDEV
+ else
+ $MDADM --stop $MDDEV --config=$RAIDTAB_CONFIG
+ fi
+
if [ $? -ne 0 ] ; then
- ocf_log "err" "Couldn't stop RAID for $MDDEV"
- return 1
+ ocf_log err "Couldn't stop RAID for $MDDEV"
+ return $OCF_ERR_GENERIC
fi
- return 0
+ return $OCF_SUCCESS
}
#
# STATUS: is the raid device online or offline?
#
raid1_status() {
# See if the MD device is online
- grep -e "^$MD" /proc/mdstat >/dev/null
+ grep -e "^$MD[ \t:]" /proc/mdstat >/dev/null
if [ $? -ne 0 ] ; then
echo "stopped"
+ return $OCF_NOT_RUNNING
else
echo "running"
+ return $OCF_SUCCESS
fi
- return 0
}
+raid1_validate_all() {
+ if [ $HAVE_RAIDTOOLS = "true" ]; then
+
+ # $MDDEV should be an md device
+ lsraid -a $MDDEV 2>&1 | grep -q -i "is not an md device"
+ if [ $? -eq 0 ]; then
+ ocf_log err "$MDDEV is not an md device!"
+ exit $OCF_ERR_ARGS
+ fi
+
+ COMMENT="\(#.*\)"
+ grep -q "^[[:space:]]*raiddev[[:space:]]\+$MDDEV[[:space:]]*$COMMENT\?$" $RAIDTAB_CONFIG 2>/dev/null
+ if [ $? -ne 0 ]; then
+ ocf_log err "Raid device $MDDEV does not appear in $RAIDTAB_CONFIG"
+ exit $OCF_ERR_GENERIC
+ fi
+ else
+ error=`$MDADM --query $MDDEV 2>&1`
+ if [ $? -ne 0 ]; then
+ ocf_log err "$error"
+ exit $OCF_ERR_GENERIC
+ fi
+ echo $error | grep -q -i "^$MDDEV[ \t:].*is not an md array"
+ if [ $? -eq 0 ]; then
+ ocf_log err "$MDDEV is not an md array!"
+ exit $OCF_ERR_ARGS
+ fi
+ fi
+
+ return $OCF_SUCCESS
+}
+
if
- ( [ $# -eq 0 ] || [ $# -gt 1 ] )
+ ( [ $# -ne 1 ] )
then
usage
- exit 1
+ exit $OCF_ERR_ARGS
fi
+case "$1" in
+ meta-data)
+ meta_data
+ exit $OCF_SUCCESS
+ ;;
+ usage)
+ usage
+ exit $OCF_SUCCESS
+ ;;
+ *)
+ ;;
+esac
+
#
# Check the necessary enviroment virable's setting
#
+
RAIDTAB_CONFIG=$OCF_RESKEY_raidconf
MDDEV=$OCF_RESKEY_raiddev
-if [ ! -f "$RAIDTAB_CONFIG" ] ; then
- ocf_log "err" "Couldn't open file $RAIDTAB_CONFIG"
- usage
- exit 1
+if [ -z "$RAIDTAB_CONFIG" ] ; then
+ ocf_log err "Please set OCF_RESKEY_raidconf!"
+ exit $OCF_ERR_ARGS
+fi
+
+if [ ! -r "$RAIDTAB_CONFIG" ] ; then
+ ocf_log err "Configuration file [$RAIDTAB_CONFIG] does not exist, or can not be opend!"
+ exit $OCF_ERR_ARGS
+fi
+
+if [ -z "$MDDEV" ] ; then
+ ocf_log err "Please set OCF_RESKEY_raiddev to the Raid device you want to control!"
+ exit $OCF_ERR_ARGS
fi
if [ ! -b "$MDDEV" ] ; then
- ocf_log "err" "Couldn't find MD device $MDDEV. Expected /dev/md* to exist"
- usage
- exit 1
+ ocf_log err "Couldn't find MD device $MDDEV. Expected /dev/md* to exist"
+ exit $OCF_ERR_ARGS
fi
# strip off the /dev/ prefix to get the name of the MD device
MD=`echo $MDDEV | sed -e 's/\/dev\///'`
# Check to make sure the utilites are found
check_util $MODPROBE
check_util $FUSER
-check_util $RAIDSTART
+#check_util $RAIDSTART
check_util $MOUNT
check_util $UMOUNT
-check_util $RAIDSTOP
+HAVE_RAIDTOOLS=false
+
+if [ -z "$RAIDSTART" -o ! -x "$RAIDSTART" ]; then
+ ocf_log info "Raidstart not found, trying mdadm..."
+ check_util $MDADM
+else
+ check_util $RAIDSTOP
+
+ HAVE_RAIDTOOLS=true
+
+fi
+
+# At this stage,
+# [ $HAVE_RAIDTOOLS = false ] <=> we have $MDADM,
+# otherwise we have raidtools (raidstart and raidstop)
# Look for how we are called
case "$1" in
- meta-data)
- meta_data
- exit $OCF_SUCCESS
- ;;
start)
raid1_start
;;
stop)
raid1_stop
;;
- status)
+ status|monitor)
raid1_status
;;
-
- usage)
- usage
- exit $OCF_SUCCESS
+ validate-all)
+ raid1_validate_all
;;
*)
usage
exit $OCF_ERR_UNIMPLEMENTED
;;
esac
exit $?

File Metadata

Mime Type
text/x-diff
Expires
Thu, Jul 10, 2:13 AM (1 d, 12 h)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
2009711
Default Alt Text
(12 KB)

Event Timeline