diff --git a/doc/sphinx/Pacemaker_Explained/nodes.rst b/doc/sphinx/Pacemaker_Explained/nodes.rst index 2f47da9f1c..cb001739d9 100644 --- a/doc/sphinx/Pacemaker_Explained/nodes.rst +++ b/doc/sphinx/Pacemaker_Explained/nodes.rst @@ -1,420 +1,437 @@ Cluster Nodes ------------- Defining a Cluster Node _______________________ Each cluster node will have an entry in the ``nodes`` section containing at least an ID and a name. A cluster node's ID is defined by the cluster layer (Corosync). .. topic:: **Example Corosync cluster node entry** .. code-block:: xml In normal circumstances, the admin should let the cluster populate this information automatically from the cluster layer. .. _node_name: Where Pacemaker Gets the Node Name ################################## The name that Pacemaker uses for a node in the configuration does not have to be the same as its local hostname. Pacemaker uses the following for a Corosync node's name, in order of most preferred first: * The value of ``name`` in the ``nodelist`` section of ``corosync.conf`` * The value of ``ring0_addr`` in the ``nodelist`` section of ``corosync.conf`` * The local hostname (value of ``uname -n``) If the cluster is running, the ``crm_node -n`` command will display the local node's name as used by the cluster. If a Corosync ``nodelist`` is used, ``crm_node --name-for-id`` with a Corosync node ID will display the name used by the node with the given Corosync ``nodeid``, for example: .. code-block:: none crm_node --name-for-id 2 .. index:: single: node; attribute single: node attribute .. _node_attributes: Node Attributes _______________ Pacemaker allows node-specific values to be specified using *node attributes*. A node attribute has a name, and may have a distinct value for each node. Node attributes come in two types, *permanent* and *transient*. Permanent node attributes are kept within the ``node`` entry, and keep their values even if the cluster restarts on a node. Transient node attributes are kept in the CIB's ``status`` section, and go away when the cluster stops on the node. While certain node attributes have specific meanings to the cluster, they are mainly intended to allow administrators and resource agents to track any information desired. For example, an administrator might choose to define node attributes for how much RAM and disk space each node has, which OS each uses, or which server room rack each node is in. Users can configure :ref:`rules` that use node attributes to affect where resources are placed. Setting and querying node attributes #################################### Node attributes can be set and queried using the ``crm_attribute`` and ``attrd_updater`` commands, so that the user does not have to deal with XML configuration directly. Here is an example command to set a permanent node attribute, and the XML configuration that would be generated: .. topic:: **Result of using crm_attribute to specify which kernel pcmk-1 is running** .. code-block:: none # crm_attribute --type nodes --node pcmk-1 --name kernel --update $(uname -r) .. code-block:: xml To read back the value that was just set: .. code-block:: none # crm_attribute --type nodes --node pcmk-1 --name kernel --query scope=nodes name=kernel value=3.10.0-862.14.4.el7.x86_64 The ``--type nodes`` indicates that this is a permanent node attribute; ``--type status`` would indicate a transient node attribute. Special node attributes ####################### Certain node attributes have special meaning to the cluster. Node attribute names beginning with ``#`` are considered reserved for these special attributes. Some special attributes do not start with ``#``, for historical reasons. Certain special attributes are set automatically by the cluster, should never be modified directly, and can be used only within :ref:`rules`; these are listed under :ref:`built-in node attributes `. For true/false values, the cluster considers a value of "1", "y", "yes", "on", or "true" (case-insensitively) to be true, "0", "n", "no", "off", "false", or unset to be false, and anything else to be an error. .. table:: **Node attributes with special significance** +----------------------------+-----------------------------------------------------+ | Name | Description | +============================+=====================================================+ | fail-count-* | .. index:: | | | pair: node attribute; fail-count | | | | | | Attributes whose names start with | | | ``fail-count-`` are managed by the cluster | | | to track how many times particular resource | | | operations have failed on this node. These | | | should be queried and cleared via the | | | ``crm_failcount`` or | | | ``crm_resource --cleanup`` commands rather | | | than directly. | +----------------------------+-----------------------------------------------------+ | last-failure-* | .. index:: | | | pair: node attribute; last-failure | | | | | | Attributes whose names start with | | | ``last-failure-`` are managed by the cluster | | | to track when particular resource operations | | | have most recently failed on this node. | | | These should be cleared via the | | | ``crm_failcount`` or | | | ``crm_resource --cleanup`` commands rather | | | than directly. | +----------------------------+-----------------------------------------------------+ | maintenance | .. index:: | | | pair: node attribute; maintenance | | | | | | Similar to the ``maintenance-mode`` | | | :ref:`cluster option `, but | | | for a single node. If true, resources will | | | not be started or stopped on the node, | | | resources and individual clone instances | | | running on the node will become unmanaged, | | | and any recurring operations for those will | | | be cancelled. | | | | | | **Warning:** Restarting pacemaker on a node that is | | | in single-node maintenance mode will likely | | | lead to undesirable effects. If | | | ``maintenance`` is set as a transient | | | attribute, it will be erased when | | | Pacemaker is stopped, which will | | | immediately take the node out of | | | maintenance mode and likely get it | | | fenced. Even if permanent, if Pacemaker | | | is restarted, any resources active on the | | | node will have their local history erased | | | when the node rejoins, so the cluster | | | will no longer consider them running on | | | the node and thus will consider them | | | managed again, leading them to be started | | | elsewhere. This behavior might be | | | improved in a future release. | +----------------------------+-----------------------------------------------------+ | probe_complete | .. index:: | | | pair: node attribute; probe_complete | | | | | | This is managed by the cluster to detect | | | when nodes need to be reprobed, and should | | | never be used directly. | +----------------------------+-----------------------------------------------------+ | resource-discovery-enabled | .. index:: | | | pair: node attribute; resource-discovery-enabled | | | | | | If the node is a remote node, fencing is enabled, | | | and this attribute is explicitly set to false | | | (unset means true in this case), resource discovery | | | (probes) will not be done on this node. This is | | | highly discouraged; the ``resource-discovery`` | | | location constraint property is preferred for this | | | purpose. | +----------------------------+-----------------------------------------------------+ | shutdown | .. index:: | | | pair: node attribute; shutdown | | | | | | This is managed by the cluster to orchestrate the | | | shutdown of a node, and should never be used | | | directly. | +----------------------------+-----------------------------------------------------+ | site-name | .. index:: | | | pair: node attribute; site-name | | | | | | If set, this will be used as the value of the | | | ``#site-name`` node attribute used in rules. (If | | | not set, the value of the ``cluster-name`` cluster | | | option will be used as ``#site-name`` instead.) | +----------------------------+-----------------------------------------------------+ | standby | .. index:: | | | pair: node attribute; standby | | | | | | If true, the node is in standby mode. This is | | | typically set and queried via the ``crm_standby`` | | | command rather than directly. | +----------------------------+-----------------------------------------------------+ | terminate | .. index:: | | | pair: node attribute; terminate | | | | | | If the value is true or begins with any nonzero | | | number, the node will be fenced. This is typically | | | set by tools rather than directly. | +----------------------------+-----------------------------------------------------+ | #digests-* | .. index:: | | | pair: node attribute; #digests | | | | | | Attributes whose names start with ``#digests-`` are | | | managed by the cluster to detect when | | | :ref:`unfencing` needs to be redone, and should | | | never be used directly. | +----------------------------+-----------------------------------------------------+ | #node-unfenced | .. index:: | | | pair: node attribute; #node-unfenced | | | | | | When the node was last unfenced (as seconds since | | | the epoch). This is managed by the cluster and | | | should never be used directly. | +----------------------------+-----------------------------------------------------+ .. index:: single: node; health .. _node-health: Tracking Node Health ____________________ A node may be functioning adequately as far as cluster membership is concerned, and yet be "unhealthy" in some respect that makes it an undesirable location for resources. For example, a disk drive may be reporting SMART errors, or the CPU may be highly loaded. Pacemaker offers a way to automatically move resources off unhealthy nodes. .. index:: single: node attribute; health Node Health Attributes ###################### Pacemaker will treat any node attribute whose name starts with ``#health`` as an indicator of node health. Node health attributes may have one of the following values: .. table:: **Allowed Values for Node Health Attributes** +------------+--------------------------------------------------------------+ | Value | Intended significance | +============+==============================================================+ | ``red`` | .. index:: | | | single: red; node health attribute value | | | single: node attribute; health (red) | | | | | | This indicator is unhealthy | +------------+--------------------------------------------------------------+ | ``yellow`` | .. index:: | | | single: yellow; node health attribute value | | | single: node attribute; health (yellow) | | | | | | This indicator is becoming unhealthy | +------------+--------------------------------------------------------------+ | ``green`` | .. index:: | | | single: green; node health attribute value | | | single: node attribute; health (green) | | | | | | This indicator is healthy | +------------+--------------------------------------------------------------+ | *integer* | .. index:: | | | single: score; node health attribute value | | | single: node attribute; health (score) | | | | | | A numeric score to apply to all resources on this node (0 or | | | positive is healthy, negative is unhealthy) | +------------+--------------------------------------------------------------+ .. index:: pair: cluster option; node-health-strategy Node Health Strategy #################### Pacemaker assigns a node health score to each node, as the sum of the values of all its node health attributes. This score will be used as a location constraint applied to this node for all resources. The ``node-health-strategy`` cluster option controls how Pacemaker responds to changes in node health attributes, and how it translates ``red``, ``yellow``, and ``green`` to scores. Allowed values are: .. table:: **Node Health Strategies** +----------------+----------------------------------------------------------+ | Value | Effect | +================+==========================================================+ | none | .. index:: | | | single: node-health-strategy; none | | | single: none; node-health-strategy value | | | | | | Do not track node health attributes at all. | +----------------+----------------------------------------------------------+ | migrate-on-red | .. index:: | | | single: node-health-strategy; migrate-on-red | | | single: migrate-on-red; node-health-strategy value | | | | | | Assign the value of ``-INFINITY`` to ``red``, and 0 to | | | ``yellow`` and ``green``. This will cause all resources | | | to move off the node if any attribute is ``red``. | +----------------+----------------------------------------------------------+ | only-green | .. index:: | | | single: node-health-strategy; only-green | | | single: only-green; node-health-strategy value | | | | | | Assign the value of ``-INFINITY`` to ``red`` and | | | ``yellow``, and 0 to ``green``. This will cause all | | | resources to move off the node if any attribute is | | | ``red`` or ``yellow``. | +----------------+----------------------------------------------------------+ | progressive | .. index:: | | | single: node-health-strategy; progressive | | | single: progressive; node-health-strategy value | | | | | | Assign the value of the ``node-health-red`` cluster | | | option to ``red``, the value of ``node-health-yellow`` | | | to ``yellow``, and the value of ``node-health-green`` to | | | ``green``. Each node is additionally assigned a score of | | | ``node-health-base`` (this allows resources to start | | | even if some attributes are ``yellow``). This strategy | | | gives the administrator finer control over how important | | | each value is. | +----------------+----------------------------------------------------------+ | custom | .. index:: | | | single: node-health-strategy; custom | | | single: custom; node-health-strategy value | | | | | | Track node health attributes using the same values as | | | ``progressive`` for ``red``, ``yellow``, and ``green``, | | | but do not take them into account. The administrator is | | | expected to implement a policy by defining :ref:`rules` | | | referencing node health attributes. | +----------------+----------------------------------------------------------+ +Exempting a Resource from Health Restrictions +############################################# + +If you want a resource to be able to run on a node even if its health score +would otherwise prevent it, set the resource's ``allow-unhealthy-nodes`` +meta-attribute to ``true`` *(available since 2.1.3)*. + +This is particularly useful for node health agents, to allow them to detect +when the node becomes healthy again. If you configure a health agent without +this setting, then the health agent will be banned from an unhealthy node, +and you will have to investigate and clear the health attribute manually once +it is healthy to allow resources on the node again. + +If you want the meta-attribute to apply to a clone, it must be set on the clone +itself, not on the resource being cloned. + + Configuring Node Health Agents ############################## Since Pacemaker calculates node health based on node attributes, any method that sets node attributes may be used to measure node health. The most common are resource agents and custom daemons. Pacemaker provides examples that can be used directly or as a basis for custom code. The ``ocf:pacemaker:HealthCPU``, ``ocf:pacemaker:HealthIOWait``, and ``ocf:pacemaker:HealthSMART`` resource agents set node health attributes based on CPU and disk status. To take advantage of this feature, add the resource to your cluster (generally as a cloned resource with a recurring monitor action, to continually check the health of all nodes). For example: .. topic:: Example HealthIOWait resource configuration .. code-block:: xml The resource agents use ``attrd_updater`` to set proper status for each node running this resource, as a node attribute whose name starts with ``#health`` (for ``HealthIOWait``, the node attribute is named ``#health-iowait``). When a node is no longer faulty, you can force the cluster to make it available to take resources without waiting for the next monitor, by setting the node health attribute to green. For example: .. topic:: **Force node1 to be marked as healthy** .. code-block:: none # attrd_updater --name "#health-iowait" --update "green" --node "node1" diff --git a/doc/sphinx/Pacemaker_Explained/resources.rst b/doc/sphinx/Pacemaker_Explained/resources.rst index c86559d655..be4d57d987 100644 --- a/doc/sphinx/Pacemaker_Explained/resources.rst +++ b/doc/sphinx/Pacemaker_Explained/resources.rst @@ -1,1063 +1,1067 @@ .. _resource: Cluster Resources ----------------- .. _s-resource-primitive: What is a Cluster Resource? ########################### .. index:: single: resource A resource is a service made highly available by a cluster. The simplest type of resource, a *primitive* resource, is described in this chapter. More complex forms, such as groups and clones, are described in later chapters. Every primitive resource has a *resource agent*. A resource agent is an external program that abstracts the service it provides and present a consistent view to the cluster. This allows the cluster to be agnostic about the resources it manages. The cluster doesn't need to understand how the resource works because it relies on the resource agent to do the right thing when given a **start**, **stop** or **monitor** command. For this reason, it is crucial that resource agents are well-tested. Typically, resource agents come in the form of shell scripts. However, they can be written using any technology (such as C, Python or Perl) that the author is comfortable with. .. _s-resource-supported: .. index:: single: resource; class Resource Classes ################ Pacemaker supports several classes of agents: * OCF * LSB * Systemd * Upstart (deprecated) * Service * Fencing * Nagios Plugins .. index:: single: resource; OCF single: OCF; resources single: Open Cluster Framework; resources Open Cluster Framework ______________________ The OCF standard [#]_ is basically an extension of the Linux Standard Base conventions for init scripts to: * support parameters, * make them self-describing, and * make them extensible OCF specs have strict definitions of the exit codes that actions must return [#]_. The cluster follows these specifications exactly, and giving the wrong exit code will cause the cluster to behave in ways you will likely find puzzling and annoying. In particular, the cluster needs to distinguish a completely stopped resource from one which is in some erroneous and indeterminate state. Parameters are passed to the resource agent as environment variables, with the special prefix ``OCF_RESKEY_``. So, a parameter which the user thinks of as ``ip`` will be passed to the resource agent as ``OCF_RESKEY_ip``. The number and purpose of the parameters is left to the resource agent; however, the resource agent should use the **meta-data** command to advertise any that it supports. The OCF class is the most preferred as it is an industry standard, highly flexible (allowing parameters to be passed to agents in a non-positional manner) and self-describing. For more information, see the `reference `_ and the *Resource Agents* chapter of *Pacemaker Administration*. .. index:: single: resource; LSB single: LSB; resources single: Linux Standard Base; resources Linux Standard Base ___________________ *LSB* resource agents are more commonly known as *init scripts*. If a full path is not given, they are assumed to be located in ``/etc/init.d``. Commonly, they are provided by the OS distribution. In order to be used with a Pacemaker cluster, they must conform to the LSB specification [#]_. .. warning:: Many distributions or particular software packages claim LSB compliance but ship with broken init scripts. For details on how to check whether your init script is LSB-compatible, see the `Resource Agents` chapter of `Pacemaker Administration`. Common problematic violations of the LSB standard include: * Not implementing the ``status`` operation at all * Not observing the correct exit status codes for ``start``/``stop``/``status`` actions * Starting a started resource returns an error * Stopping a stopped resource returns an error .. important:: Remember to make sure the computer is `not` configured to start any services at boot time -- that should be controlled by the cluster. .. _s-resource-supported-systemd: .. index:: single: Resource; Systemd single: Systemd; resources Systemd _______ Most Linux distributions have replaced the old `SysV `_ style of initialization daemons and scripts with `Systemd `_. Pacemaker is able to manage these services `if they are present`. Instead of init scripts, systemd has `unit files`. Generally, the services (unit files) are provided by the OS distribution, but there are online guides for converting from init scripts [#]_. .. important:: Remember to make sure the computer is `not` configured to start any services at boot time -- that should be controlled by the cluster. .. index:: single: Resource; Upstart single: Upstart; resources Upstart _______ Some distributions replaced the old `SysV `_ style of initialization daemons (and scripts) with `Upstart `_. Pacemaker is able to manage these services `if they are present`. Instead of init scripts, Upstart has `jobs`. Generally, the services (jobs) are provided by the OS distribution. .. important:: Remember to make sure the computer is `not` configured to start any services at boot time -- that should be controlled by the cluster. .. warning:: Upstart support is deprecated in Pacemaker. Upstart is no longer an actively maintained project, and test platforms for it are no longer readily usable. Support will likely be dropped entirely at the next major release of Pacemaker. .. index:: single: Resource; System Services single: System Service; resources System Services _______________ Since there are various types of system services (``systemd``, ``upstart``, and ``lsb``), Pacemaker supports a special ``service`` alias which intelligently figures out which one applies to a given cluster node. This is particularly useful when the cluster contains a mix of ``systemd``, ``upstart``, and ``lsb``. In order, Pacemaker will try to find the named service as: * an LSB init script * a Systemd unit file * an Upstart job .. index:: single: Resource; STONITH single: STONITH; resources STONITH _______ The STONITH class is used exclusively for fencing-related resources. This is discussed later in :ref:`fencing`. .. index:: single: Resource; Nagios Plugins single: Nagios Plugins; resources Nagios Plugins ______________ Nagios Plugins [#]_ are a way to monitor services. Pacemaker can use these as resources, to react to a change in the service's status. To use plugins as resources, Pacemaker must have been built with support, and OCF-style meta-data for the plugins must be installed on nodes that can run them. Meta-data for several common plugins is provided by the `nagios-agents-metadata `_ project. The supported parameters for such a resource are same as the long options of the plugin. Start and monitor actions for plugin resources are implemented as invoking the plugin. A plugin result of "OK" (0) is treated as success, a result of "WARN" (1) is treated as a successful but degraded service, and any other result is considered a failure. A plugin resource is not going to change its status after recovery by restarting the plugin, so using them alone does not make sense with ``on-fail`` set (or left to default) to ``restart``. Another value could make sense, for example, if you want to fence or standby nodes that cannot reach some external service. A more common use case for plugin resources is to configure them with a ``container`` meta-attribute set to the name of another resource that actually makes the service available, such as a virtual machine or container. With ``container`` set, the plugin resource will automatically be colocated with the containing resource and ordered after it, and the containing resource will be considered failed if the plugin resource fails. This allows monitoring of a service inside a virtual machine or container, with recovery of the virtual machine or container if the service fails. Configuring a virtual machine as a guest node, or a container as a :ref:`bundle `, is the preferred way of monitoring a service inside, but plugin resources can be useful when it is not practical to modify the virtual machine or container image for this purpose. .. _primitive-resource: Resource Properties ################### These values tell the cluster which resource agent to use for the resource, where to find that resource agent and what standards it conforms to. .. table:: **Properties of a Primitive Resource** +----------+------------------------------------------------------------------+ | Field | Description | +==========+==================================================================+ | id | .. index:: | | | single: id; resource | | | single: resource; property, id | | | | | | Your name for the resource | +----------+------------------------------------------------------------------+ | class | .. index:: | | | single: class; resource | | | single: resource; property, class | | | | | | The standard the resource agent conforms to. Allowed values: | | | ``lsb``, ``nagios``, ``ocf``, ``service``, ``stonith``, | | | ``systemd``, ``upstart`` | +----------+------------------------------------------------------------------+ | type | .. index:: | | | single: type; resource | | | single: resource; property, type | | | | | | The name of the Resource Agent you wish to use. E.g. | | | ``IPaddr`` or ``Filesystem`` | +----------+------------------------------------------------------------------+ | provider | .. index:: | | | single: provider; resource | | | single: resource; property, provider | | | | | | The OCF spec allows multiple vendors to supply the same resource | | | agent. To use the OCF resource agents supplied by the Heartbeat | | | project, you would specify ``heartbeat`` here. | +----------+------------------------------------------------------------------+ The XML definition of a resource can be queried with the **crm_resource** tool. For example: .. code-block:: none # crm_resource --resource Email --query-xml might produce: .. topic:: A system resource definition .. code-block:: xml .. note:: One of the main drawbacks to system services (LSB, systemd or Upstart) resources is that they do not allow any parameters! .. topic:: An OCF resource definition .. code-block:: xml .. _resource_options: Resource Options ################ Resources have two types of options: *meta-attributes* and *instance attributes*. Meta-attributes apply to any type of resource, while instance attributes are specific to each resource agent. Resource Meta-Attributes ________________________ Meta-attributes are used by the cluster to decide how a resource should behave and can be easily set using the ``--meta`` option of the **crm_resource** command. .. table:: **Meta-attributes of a Primitive Resource** +----------------------------+----------------------------------+------------------------------------------------------+ | Field | Default | Description | +============================+==================================+======================================================+ | priority | 0 | .. index:: | | | | single: priority; resource option | | | | single: resource; option, priority | | | | | | | | If not all resources can be active, the cluster | | | | will stop lower priority resources in order to | | | | keep higher priority ones active. | +----------------------------+----------------------------------+------------------------------------------------------+ | critical | true | .. index:: | | | | single: critical; resource option | | | | single: resource; option, critical | | | | | | | | Use this value as the default for ``influence`` in | | | | all :ref:`colocation constraints | | | | ` involving this resource, | | | | as well as the implicit colocation constraints | | | | created if this resource is in a :ref:`group | | | | `. For details, see | | | | :ref:`s-coloc-influence`. *(since 2.1.0)* | +----------------------------+----------------------------------+------------------------------------------------------+ | target-role | Started | .. index:: | | | | single: target-role; resource option | | | | single: resource; option, target-role | | | | | | | | What state should the cluster attempt to keep this | | | | resource in? Allowed values: | | | | | | | | * ``Stopped:`` Force the resource to be stopped | | | | * ``Started:`` Allow the resource to be started | | | | (and in the case of :ref:`promotable clone | | | | resources `, promoted | | | | if appropriate) | | | | * ``Unpromoted:`` Allow the resource to be started, | | | | but only in the unpromoted role if the resource is | | | | :ref:`promotable ` | | | | * ``Promoted:`` Equivalent to ``Started`` | +----------------------------+----------------------------------+------------------------------------------------------+ | is-managed | TRUE | .. index:: | | | | single: is-managed; resource option | | | | single: resource; option, is-managed | | | | | | | | Is the cluster allowed to start and stop | | | | the resource? Allowed values: ``true``, ``false`` | +----------------------------+----------------------------------+------------------------------------------------------+ | maintenance | FALSE | .. index:: | | | | single: maintenance; resource option | | | | single: resource; option, maintenance | | | | | | | | Similar to the ``maintenance-mode`` | | | | :ref:`cluster option `, but for | | | | a single resource. If true, the resource will not | | | | be started, stopped, or monitored on any node. This | | | | differs from ``is-managed`` in that monitors will | | | | not be run. Allowed values: ``true``, ``false`` | +----------------------------+----------------------------------+------------------------------------------------------+ | resource-stickiness | 1 for individual clone | .. _resource-stickiness: | | | instances, 0 for all | | | | other resources | .. index:: | | | | single: resource-stickiness; resource option | | | | single: resource; option, resource-stickiness | | | | | | | | A score that will be added to the current node when | | | | a resource is already active. This allows running | | | | resources to stay where they are, even if they | | | | would be placed elsewhere if they were being | | | | started from a stopped state. | +----------------------------+----------------------------------+------------------------------------------------------+ | requires | ``quorum`` for resources | .. _requires: | | | with a ``class`` of ``stonith``, | | | | otherwise ``unfencing`` if | .. index:: | | | unfencing is active in the | single: requires; resource option | | | cluster, otherwise ``fencing`` | single: resource; option, requires | | | if ``stonith-enabled`` is true, | | | | otherwise ``quorum`` | Conditions under which the resource can be | | | | started. Allowed values: | | | | | | | | * ``nothing:`` can always be started | | | | * ``quorum:`` The cluster can only start this | | | | resource if a majority of the configured nodes | | | | are active | | | | * ``fencing:`` The cluster can only start this | | | | resource if a majority of the configured nodes | | | | are active *and* any failed or unknown nodes | | | | have been :ref:`fenced ` | | | | * ``unfencing:`` The cluster can only start this | | | | resource if a majority of the configured nodes | | | | are active *and* any failed or unknown nodes have | | | | been fenced *and* only on nodes that have been | | | | :ref:`unfenced ` | +----------------------------+----------------------------------+------------------------------------------------------+ | migration-threshold | INFINITY | .. index:: | | | | single: migration-threshold; resource option | | | | single: resource; option, migration-threshold | | | | | | | | How many failures may occur for this resource on | | | | a node, before this node is marked ineligible to | | | | host this resource. A value of 0 indicates that this | | | | feature is disabled (the node will never be marked | | | | ineligible); by constrast, the cluster treats | | | | INFINITY (the default) as a very large but finite | | | | number. This option has an effect only if the | | | | failed operation specifies ``on-fail`` as | | | | ``restart`` (the default), and additionally for | | | | failed ``start`` operations, if the cluster | | | | property ``start-failure-is-fatal`` is ``false``. | +----------------------------+----------------------------------+------------------------------------------------------+ | failure-timeout | 0 | .. index:: | | | | single: failure-timeout; resource option | | | | single: resource; option, failure-timeout | | | | | | | | How many seconds to wait before acting as if the | | | | failure had not occurred, and potentially allowing | | | | the resource back to the node on which it failed. | | | | A value of 0 indicates that this feature is | | | | disabled. | +----------------------------+----------------------------------+------------------------------------------------------+ | multiple-active | stop_start | .. index:: | | | | single: multiple-active; resource option | | | | single: resource; option, multiple-active | | | | | | | | What should the cluster do if it ever finds the | | | | resource active on more than one node? Allowed | | | | values: | | | | | | | | * ``block``: mark the resource as unmanaged | | | | * ``stop_only``: stop all active instances and | | | | leave them that way | | | | * ``stop_start``: stop all active instances and | | | | start the resource in one location only | +----------------------------+----------------------------------+------------------------------------------------------+ | allow-migrate | TRUE for ocf:pacemaker:remote | Whether the cluster should try to "live migrate" | | | resources, FALSE otherwise | this resource when it needs to be moved (see | | | | :ref:`live-migration`) | +----------------------------+----------------------------------+------------------------------------------------------+ + | allow-unhealthy-nodes | FALSE | Whether the resource should be able to run on a node | + | | | even if the node's health score would otherwise | + | | | prevent it (see :ref:`node-health`) *(since 2.1.3)* | + +----------------------------+----------------------------------+------------------------------------------------------+ | container-attribute-target | | Specific to bundle resources; see | | | | :ref:`s-bundle-attributes` | +----------------------------+----------------------------------+------------------------------------------------------+ | remote-node | | The name of the Pacemaker Remote guest node this | | | | resource is associated with, if any. If | | | | specified, this both enables the resource as a | | | | guest node and defines the unique name used to | | | | identify the guest node. The guest must be | | | | configured to run the Pacemaker Remote daemon | | | | when it is started. **WARNING:** This value | | | | cannot overlap with any resource or node IDs. | +----------------------------+----------------------------------+------------------------------------------------------+ | remote-port | 3121 | If ``remote-node`` is specified, the port on the | | | | guest used for its Pacemaker Remote connection. | | | | The Pacemaker Remote daemon on the guest must | | | | be configured to listen on this port. | +----------------------------+----------------------------------+------------------------------------------------------+ | remote-addr | value of ``remote-node`` | If ``remote-node`` is specified, the IP | | | | address or hostname used to connect to the | | | | guest via Pacemaker Remote. The Pacemaker Remote | | | | daemon on the guest must be configured to accept | | | | connections on this address. | +----------------------------+----------------------------------+------------------------------------------------------+ | remote-connect-timeout | 60s | If ``remote-node`` is specified, how long before | | | | a pending guest connection will time out. | +----------------------------+----------------------------------+------------------------------------------------------+ As an example of setting resource options, if you performed the following commands on an LSB Email resource: .. code-block:: none # crm_resource --meta --resource Email --set-parameter priority --parameter-value 100 # crm_resource -m -r Email -p multiple-active -v block the resulting resource definition might be: .. topic:: An LSB resource with cluster options .. code-block:: xml In addition to the cluster-defined meta-attributes described above, you may also configure arbitrary meta-attributes of your own choosing. Most commonly, this would be done for use in :ref:`rules `. For example, an IT department might define a custom meta-attribute to indicate which company department each resource is intended for. To reduce the chance of name collisions with cluster-defined meta-attributes added in the future, it is recommended to use a unique, organization-specific prefix for such attributes. .. _s-resource-defaults: Setting Global Defaults for Resource Meta-Attributes ____________________________________________________ To set a default value for a resource option, add it to the ``rsc_defaults`` section with ``crm_attribute``. For example, .. code-block:: none # crm_attribute --type rsc_defaults --name is-managed --update false would prevent the cluster from starting or stopping any of the resources in the configuration (unless of course the individual resources were specifically enabled by having their ``is-managed`` set to ``true``). Resource Instance Attributes ____________________________ The resource agents of some resource classes (lsb, systemd and upstart *not* among them) can be given parameters which determine how they behave and which instance of a service they control. If your resource agent supports parameters, you can add them with the ``crm_resource`` command. For example, .. code-block:: none # crm_resource --resource Public-IP --set-parameter ip --parameter-value 192.0.2.2 would create an entry in the resource like this: .. topic:: An example OCF resource with instance attributes .. code-block:: xml For an OCF resource, the result would be an environment variable called ``OCF_RESKEY_ip`` with a value of ``192.0.2.2``. The list of instance attributes supported by an OCF resource agent can be found by calling the resource agent with the ``meta-data`` command. The output contains an XML description of all the supported attributes, their purpose and default values. .. topic:: Displaying the metadata for the Dummy resource agent template .. code-block:: none # export OCF_ROOT=/usr/lib/ocf # $OCF_ROOT/resource.d/pacemaker/Dummy meta-data .. code-block:: xml 1.1 This is a dummy OCF resource agent. It does absolutely nothing except keep track of whether it is running or not, and can be configured so that actions fail or take a long time. Its purpose is primarily for testing, and to serve as a template for resource agent writers. Example stateless resource agent Location to store the resource state in. State file Fake password field Password Fake attribute that can be changed to cause a reload Fake attribute that can be changed to cause a reload Number of seconds to sleep during operations. This can be used to test how the cluster reacts to operation timeouts. Operation sleep duration in seconds. Start, migrate_from, and reload-agent actions will return failure if running on the host specified here, but the resource will run successfully anyway (future monitor calls will find it running). This can be used to test on-fail=ignore. Report bogus start failure on specified host If this is set, the environment will be dumped to this file for every call. Environment dump file .. index:: single: resource; action single: resource; operation .. _operation: Resource Operations ################### *Operations* are actions the cluster can perform on a resource by calling the resource agent. Resource agents must support certain common operations such as start, stop, and monitor, and may implement any others. Operations may be explicitly configured for two purposes: to override defaults for options (such as timeout) that the cluster will use whenever it initiates the operation, and to run an operation on a recurring basis (for example, to monitor the resource for failure). .. topic:: An OCF resource with a non-default start timeout .. code-block:: xml Pacemaker identifies operations by a combination of name and interval, so this combination must be unique for each resource. That is, you should not configure two operations for the same resource with the same name and interval. .. _operation_properties: Operation Properties ____________________ Operation properties may be specified directly in the ``op`` element as XML attributes, or in a separate ``meta_attributes`` block as ``nvpair`` elements. XML attributes take precedence over ``nvpair`` elements if both are specified. .. table:: **Properties of an Operation** +----------------+-----------------------------------+-----------------------------------------------------+ | Field | Default | Description | +================+===================================+=====================================================+ | id | | .. index:: | | | | single: id; action property | | | | single: action; property, id | | | | | | | | A unique name for the operation. | +----------------+-----------------------------------+-----------------------------------------------------+ | name | | .. index:: | | | | single: name; action property | | | | single: action; property, name | | | | | | | | The action to perform. This can be any action | | | | supported by the agent; common values include | | | | ``monitor``, ``start``, and ``stop``. | +----------------+-----------------------------------+-----------------------------------------------------+ | interval | 0 | .. index:: | | | | single: interval; action property | | | | single: action; property, interval | | | | | | | | How frequently (in seconds) to perform the | | | | operation. A value of 0 means "when needed". | | | | A positive value defines a *recurring action*, | | | | which is typically used with | | | | :ref:`monitor `. | +----------------+-----------------------------------+-----------------------------------------------------+ | timeout | | .. index:: | | | | single: timeout; action property | | | | single: action; property, timeout | | | | | | | | How long to wait before declaring the action | | | | has failed | +----------------+-----------------------------------+-----------------------------------------------------+ | on-fail | Varies by action: | .. index:: | | | | single: on-fail; action property | | | * ``stop``: ``fence`` if | single: action; property, on-fail | | | ``stonith-enabled`` is true | | | | or ``block`` otherwise | The action to take if this action ever fails. | | | * ``demote``: ``on-fail`` of the | Allowed values: | | | ``monitor`` action with | | | | ``role`` set to ``Promoted``, | * ``ignore:`` Pretend the resource did not fail. | | | if present, enabled, and | * ``block:`` Don't perform any further operations | | | configured to a value other | on the resource. | | | than ``demote``, or ``restart`` | * ``stop:`` Stop the resource and do not start | | | otherwise | it elsewhere. | | | * all other actions: ``restart`` | * ``demote:`` Demote the resource, without a | | | | full restart. This is valid only for ``promote`` | | | | actions, and for ``monitor`` actions with both | | | | a nonzero ``interval`` and ``role`` set to | | | | ``Promoted``; for any other action, a | | | | configuration error will be logged, and the | | | | default behavior will be used. *(since 2.0.5)* | | | | * ``restart:`` Stop the resource and start it | | | | again (possibly on a different node). | | | | * ``fence:`` STONITH the node on which the | | | | resource failed. | | | | * ``standby:`` Move *all* resources away from the | | | | node on which the resource failed. | +----------------+-----------------------------------+-----------------------------------------------------+ | enabled | TRUE | .. index:: | | | | single: enabled; action property | | | | single: action; property, enabled | | | | | | | | If ``false``, ignore this operation definition. | | | | This is typically used to pause a particular | | | | recurring ``monitor`` operation; for instance, it | | | | can complement the respective resource being | | | | unmanaged (``is-managed=false``), as this alone | | | | will :ref:`not block any configured monitoring | | | | `. Disabling the operation | | | | does not suppress all actions of the given type. | | | | Allowed values: ``true``, ``false``. | +----------------+-----------------------------------+-----------------------------------------------------+ | record-pending | TRUE | .. index:: | | | | single: record-pending; action property | | | | single: action; property, record-pending | | | | | | | | If ``true``, the intention to perform the operation | | | | is recorded so that GUIs and CLI tools can indicate | | | | that an operation is in progress. This is best set | | | | as an *operation default* | | | | (see :ref:`s-operation-defaults`). Allowed values: | | | | ``true``, ``false``. | +----------------+-----------------------------------+-----------------------------------------------------+ | role | | .. index:: | | | | single: role; action property | | | | single: action; property, role | | | | | | | | Run the operation only on node(s) that the cluster | | | | thinks should be in the specified role. This only | | | | makes sense for recurring ``monitor`` operations. | | | | Allowed (case-sensitive) values: ``Stopped``, | | | | ``Started``, and in the case of :ref:`promotable | | | | clone resources `, | | | | ``Unpromoted`` and ``Promoted``. | +----------------+-----------------------------------+-----------------------------------------------------+ .. note:: When ``on-fail`` is set to ``demote``, recovery from failure by a successful demote causes the cluster to recalculate whether and where a new instance should be promoted. The node with the failure is eligible, so if promotion scores have not changed, it will be promoted again. There is no direct equivalent of ``migration-threshold`` for the promoted role, but the same effect can be achieved with a location constraint using a :ref:`rule ` with a node attribute expression for the resource's fail count. For example, to immediately ban the promoted role from a node with any failed promote or promoted instance monitor: .. code-block:: xml This example assumes that there is a promotable clone of the ``my_primitive`` resource (note that the primitive name, not the clone name, is used in the rule), and that there is a recurring 10-second-interval monitor configured for the promoted role (fail count attributes specify the interval in milliseconds). .. _s-resource-monitoring: Monitoring Resources for Failure ________________________________ When Pacemaker first starts a resource, it runs one-time ``monitor`` operations (referred to as *probes*) to ensure the resource is running where it's supposed to be, and not running where it's not supposed to be. (This behavior can be affected by the ``resource-discovery`` location constraint property.) Other than those initial probes, Pacemaker will *not* (by default) check that the resource continues to stay healthy [#]_. You must configure ``monitor`` operations explicitly to perform these checks. .. topic:: An OCF resource with a recurring health check .. code-block:: xml By default, a ``monitor`` operation will ensure that the resource is running where it is supposed to. The ``target-role`` property can be used for further checking. For example, if a resource has one ``monitor`` operation with ``interval=10 role=Started`` and a second ``monitor`` operation with ``interval=11 role=Stopped``, the cluster will run the first monitor on any nodes it thinks *should* be running the resource, and the second monitor on any nodes that it thinks *should not* be running the resource (for the truly paranoid, who want to know when an administrator manually starts a service by mistake). .. note:: Currently, monitors with ``role=Stopped`` are not implemented for :ref:`clone ` resources. .. _s-monitoring-unmanaged: Monitoring Resources When Administration is Disabled ____________________________________________________ Recurring ``monitor`` operations behave differently under various administrative settings: * When a resource is unmanaged (by setting ``is-managed=false``): No monitors will be stopped. If the unmanaged resource is stopped on a node where the cluster thinks it should be running, the cluster will detect and report that it is not, but it will not consider the monitor failed, and will not try to start the resource until it is managed again. Starting the unmanaged resource on a different node is strongly discouraged and will at least cause the cluster to consider the resource failed, and may require the resource's ``target-role`` to be set to ``Stopped`` then ``Started`` to be recovered. * When a node is put into standby: All resources will be moved away from the node, and all ``monitor`` operations will be stopped on the node, except those specifying ``role`` as ``Stopped`` (which will be newly initiated if appropriate). * When the cluster is put into maintenance mode: All resources will be marked as unmanaged. All monitor operations will be stopped, except those specifying ``role`` as ``Stopped`` (which will be newly initiated if appropriate). As with single unmanaged resources, starting a resource on a node other than where the cluster expects it to be will cause problems. .. _s-operation-defaults: Setting Global Defaults for Operations ______________________________________ You can change the global default values for operation properties in a given cluster. These are defined in an ``op_defaults`` section of the CIB's ``configuration`` section, and can be set with ``crm_attribute``. For example, .. code-block:: none # crm_attribute --type op_defaults --name timeout --update 20s would default each operation's ``timeout`` to 20 seconds. If an operation's definition also includes a value for ``timeout``, then that value would be used for that operation instead. When Implicit Operations Take a Long Time _________________________________________ The cluster will always perform a number of implicit operations: ``start``, ``stop`` and a non-recurring ``monitor`` operation used at startup to check whether the resource is already active. If one of these is taking too long, then you can create an entry for them and specify a longer timeout. .. topic:: An OCF resource with custom timeouts for its implicit actions .. code-block:: xml Multiple Monitor Operations ___________________________ Provided no two operations (for a single resource) have the same name and interval, you can have as many ``monitor`` operations as you like. In this way, you can do a superficial health check every minute and progressively more intense ones at higher intervals. To tell the resource agent what kind of check to perform, you need to provide each monitor with a different value for a common parameter. The OCF standard creates a special parameter called ``OCF_CHECK_LEVEL`` for this purpose and dictates that it is "made available to the resource agent without the normal ``OCF_RESKEY`` prefix". Whatever name you choose, you can specify it by adding an ``instance_attributes`` block to the ``op`` tag. It is up to each resource agent to look for the parameter and decide how to use it. .. topic:: An OCF resource with two recurring health checks, performing different levels of checks specified via ``OCF_CHECK_LEVEL``. .. code-block:: xml Disabling a Monitor Operation _____________________________ The easiest way to stop a recurring monitor is to just delete it. However, there can be times when you only want to disable it temporarily. In such cases, simply add ``enabled=false`` to the operation's definition. .. topic:: Example of an OCF resource with a disabled health check .. code-block:: xml This can be achieved from the command line by executing: .. code-block:: none # cibadmin --modify --xml-text '' Once you've done whatever you needed to do, you can then re-enable it with .. code-block:: none # cibadmin --modify --xml-text '' .. [#] See https://github.com/ClusterLabs/OCF-spec/tree/main/ra. The Pacemaker implementation has been somewhat extended from the OCF specs. .. [#] The resource-agents source code includes the **ocf-tester** script, which can be useful in this regard. .. [#] See http://refspecs.linux-foundation.org/LSB_3.0.0/LSB-Core-generic/LSB-Core-generic/iniscrptact.html for the LSB Spec as it relates to init scripts. .. [#] For example, http://0pointer.de/blog/projects/systemd-for-admins-3.html .. [#] The project has two independent forks, hosted at https://www.nagios-plugins.org/ and https://www.monitoring-plugins.org/. Output from both projects' plugins is similar, so plugins from either project can be used with pacemaker. .. [#] Currently, anyway. Automatic monitoring operations may be added in a future version of Pacemaker.