The set of scripts that determine how a resource is started, monitored, and stopped. There must be a set of action scripts specified for each resource type. The possible set of action scripts is: exclusive, start, stop, monitor, and restart.

A configuration in which all resource groups have the same primary node. The backup node does not run any highly available resource groups until a failover occurs.

A cluster is the set of systems (nodes) configured to work together as a single computing resource. A cluster is identified by a simple name and a cluster ID.

There is only one cluster that may be formed from a given pool of nodes.

Disks or logical units (LUNs) are assigned to clusters by recording the name of the cluster on the disk (or LUN). Thus, if any disk is accessible (via a Fibre Channel connection) from machines in multiple clusters, then those clusters must have unique names. When members of a cluster send messages to each other, they identify their cluster via the cluster ID. Thus, if two clusters will be sharing the same network for communications, then they must have unique cluster IDs.

Because of the above restrictions on cluster names and cluster IDs, and because cluster names and cluster IDs cannot be changed once the cluster is created (without deleting the cluster and recreating it), SGI advises that you choose unique names and cluster IDs for each of the clusters within your organization. Clusters that share a network and use XVM must have unique names.

A node in a coexecution cluster that is installed with the cluster_admin software product, allowing the node to perform cluster administration tasks and contain a copy of the cluster database. Also known as a CXFS administration node.

The person responsible for managing and maintaining a cluster.

Contains configuration information about all resources, resource types, resource groups, failover policies, nodes, and the cluster.

The group of nodes in the pool that are accessible to fs2d and therefore can receive cluster database updates; this may be a subset of the nodes defined in the pool. Also known as user-space membership and fs2d membership.

A unique number within your network in the range 1 through 128. The cluster ID is used by the IRIX kernel to make sure that it does not accept cluster information from any other cluster that may be on the network. The kernel does not use the database for communication, so it requires the cluster ID in order to verify cluster communications. This information in the kernel cannot be changed after it has been initialized; therefore, you must not change a cluster ID after the cluster has been defined.

A node that is defined as part of the cluster. See also node.

A group of application instances in a distributed application that cooperate to provide a service.

For example, distributed lock manager instances in each node would form a process group. By forming a process group, they can obtain membership and reliable, ordered, atomic communication services. There is no relationship between a UNIX process group and a cluster process group.

The nodes in the FailSafe cluster itself from which you want to gather statistics, on which PCP for FailSafe has installed the collector agents.

Messages that cluster software sends between the nodes to request operations on or distribute information about nodes and resource groups. FailSafe sends control messages for the purpose of ensuring that nodes and groups remain highly available. Control messages and heartbeat messages are sent through a node's network interfaces that have been attached to a control network. A node can be attached to multiple control networks.

The network that connects nodes through their network interfaces (typically Ethernet) such that FailSafe can maintain a cluster's high availability by sending heartbeat messages and control messages through the network to the attached nodes. FailSafe uses the highest priority network interface on the control network; it uses a network interface with lower priority when all higher-priority network interfaces on the control network fail.

A node must have at least one control network interface for heartbeat messages and one for control messages (both heartbeat and control messages can be configured to use the same interface). A node can have no more than eight control network interfaces.

A node that is installed with the cluster_admin software product, allowing the node to perform cluster administration tasks and contain a copy of the cluster database, but is not capable of coordinating CXFS metadata. FailSafe can run on a CXFS client-administration node.

A node that is installed with the cxfs_client.sw.base software product; it does not run cluster administration daemons and is not capable of coordinating cluster activity and metadata. FailSafe cannot run on a client-only node.

A node in a coexecution cluster that is installed with the cluster_admin product and is also capable of coordinating CXFS metadata. FailSafe can run on a CXFS server-capable administration node.

See cluster database.

See resource dependency or resource type dependency.

The process of allocating a resource group to another node according to a failover policy A failover may be triggered by the failure of a resource, a change in the FailSafe membership (such as when a node fails or starts), or a manual request by the administrator.

A string that affects the allocation of a resource group in a cluster. The administrator must specify system-defined attributes (such as Auto_Failback or Controlled_Failback ), and can optionally supply site-specific attributes.

The ordered list of nodes on which a particular resource group can be allocated. The nodes listed in the failover domain must be within the same cluster; however, the failover domain does not have to include every node in the cluster. The administrator defines the initial failover domain when creating a failover policy. This list is transformed into the run-time failover domain by the failover script the run-time failover domain is what is actually used to select the failover node. FailSafe stores the run-time failover domain and uses it as input to the next failover script invocation. The initial and run-time failover domains may be identical, depending upon the contents of the failover script. In general, FailSafe allocates a given resource group to the first node listed in the run-time failover domain that is also in the FailSafe membership; the point at which this allocation takes place is affected by the failover attributes.

The method used by FailSafe to determine the destination node of a failover. A failover policy consists of a failover domain, failover attributes, and a failover script. A failover policy name must be unique within the pool.

A failover policy component that generates a run-time failover domain and returns it to the FailSafe process. The process applies the failover attributes and then selects the first node in the returned failover domain that is also in the current FailSafe membership.

The list of FailSafe nodes in a cluster on which FailSafe can make resource groups online. It differs from the CXFS membership. For more information about CXFS, see the CXFS Administration Guide for SGI Infinite Storage.

See cluster database.

Messages that cluster software sends between the nodes that indicate a node is up and running. Heartbeat messages and control messages are sent through a node's network interfaces that have been attached to a control network. A node can be attached to multiple control networks.

Interval between heartbeat messages. The node timeout value must be at least 10 times the heartbeat interval for proper FailSafe operation (otherwise false failovers may be triggered). The higher the number of heartbeats (smaller heartbeat interval), the greater the potential for slowing down the network. Conversely, the fewer the number of heartbeats (larger heartbeat interval), the greater the potential for reducing availability of resources.

The ordered list of nodes, defined by the administrator when a failover policy is first created, that is used the first time a cluster is booted. The ordered list specified by the initial failover domain is transformed into a run-time failover domain by the failover script; the run-time failover domain is used along with failover attributes to determine the node on which a resource group should reside. With each failure, the failover script takes the current run-time failover domain and potentially modifies it; the initial failover domain is never used again. Depending on the run-time conditions and contents of the failover script, the initial and run-time failover domains may be identical. See also run-time failover domain.

A set of information that must be defined for a particular resource type. For example, for the resource type filesystem one key/value pair might be mount_point=/fs1 where mount_point is the key and fs1 is the value specific to the particular resource being defined. Depending on the value, you specify either a string or integer data type. In the previous example, you would specify string as the data type for the value fs1.

A log configuration has two parts: a log level and a log file, both associated with a log group. The cluster administrator can customize the location and amount of log output, and can specify a log configuration for all nodes or for only one node. For example, the crsd log group can be configured to log detailed level-10 messages to the /var/cluster/ha/log/crsd-foo log only on the node foo and to write only minimal level-1 messages to the crsd log on all other nodes.

A file containing notifications for a particular log group. A log file is part of the log configuration for a log group. By default, log files reside in the /var/cluster/ha/log directory, but the cluster administrator can customize this. Note: FailSafe logs both normal operations and critical errors to /var/adm/SYSLOG, as well as to individual logs for specific log groups.

A set of one or more FailSafe processes that use the same log configuration. A log group usually corresponds to one daemon, such as gcd.

A number controlling the number of log messages that FailSafe will write into an associated log group's log file. A log level is part of the log configuration for a log group.

Logical unit number

A workstation that has a display and is running the IRIS Desktop, on which PCP for FailSafe has installed the monitor client.

A node is an operating system (OS) image, usually an individual computer. (This use of the term node does not have the same meaning as a node in an SGI Origin 3000 or SGI 2000 system.)

A given node can be a member of only one pool (and therefore) only one cluster.

A 16-bit positive integer that uniquely defines a node. During node definition, FailSafe will assign a node ID if one has not been assigned by the cluster administrator. Once assigned, the node ID cannot be modified.

If no heartbeat is received from a node in this period of time, the node is considered to be dead. The node timeout value must be at least 10 times the heartbeat interval for proper FailSafe operation (otherwise false failovers may be triggered).

The command used to notify the cluster administrator of changes or failures in the cluster, nodes, and resource groups. The command must exist on every node in the cluster.

A resource group that is not highly available in the cluster. To put a resource group in offline state, FailSafe stops the group (if needed) and stops monitoring the group. An offline resource group can be running on a node, yet not under FailSafe control. If the cluster administrator specifies the detach only option while taking the group offline, then FailSafe will not stop the group but will stop monitoring the group.

A resource group that is highly available in the cluster. When FailSafe detects a failure that degrades the resource group availability, it moves the resource group to another node in the cluster. To put a resource group in online state, FailSafe starts the group (if needed) and begins monitoring the group. If the cluster administrator specifies the attach only option while bringing the group online, then FailSafe will not start the group but will begin monitoring the group.

A system that can control a node remotely, such as power-cycling the node. At run time, the owner host must be defined as a node in the pool.

The device file name of the terminal port (TTY) on the owner host to which the system controller serial cable is connected. The other end of the cable connects to the node with the system controller port, so the node can be controlled remotely by the owner host.

The set of software required to make an application highly available, including a resource type and action scripts. There are plug-ins provided with the base FailSafe release, optional plug-ins available for purchase from SGI, and customized plug-ins you can write using the instructions in this guide.

The pool is the set of nodes from which a particular cluster may be formed. Only one cluster may be configured from a given pool, and it need not contain all of the available nodes. (Other pools may exist, but each is disjoint from the other. They share no node or cluster definitions.)

A pool is formed when you connect to a given node and define that node in the cluster database using the CXFS GUI or cmgr command. You can then add other nodes to the pool by defining them while still connected to the first node, or to any other node that is already in the pool. (If you were to connect to another node and then define it, you would be creating a second pool).

The password for the system controller port, usually set once in firmware or by setting jumper wires. (This is not the same as the node's root password.)

When powerfail mode is turned on, FailSafe tracks the response from a node's system controller as it makes reset requests to a node. When these requests fail to reset the node successfully, FailSafe uses heuristics to try to estimate whether the machine has been powered down. If the heuristic algorithm returns with success, FailSafe assumes the remote machine has been reset successfully. When powerfail mode is turned off, the heuristics are not used and FailSafe may not be able to detect node power failures.

A list of process instances in a cluster that form a process group. There can multiple process groups per node.

The process of moving the physical medium access control (MAC) address of a network interface to another interface. It is done by using the macconfig command.

A single physical or logical entity that provides a service to clients or other resources. For example, a resource can be a single disk volume, a particular network address, or an application such as a web server. A resource is generally available for use over time on two or more nodes in a cluster, although it can be allocated to only one node at any given time. Resources are identified by a resource name and a resource type. Dependent resources must be part of the same resource group and are identified in a resource dependency list.

The condition in which a resource requires the existence of other resources.

A list of resources upon which a resource depends. Each resource instance must have resource dependencies that satisfy its resource type dependencies before it can be added to a resource group.

A collection of resources. A resource group is identified by a simple name; this name must be unique within a cluster. Resource groups cannot overlap; that is, two resource groups cannot contain the same resource. All interdependent resources must be part of the same resource group. If any individual resource in a resource group becomes unavailable for its intended use, then the entire resource group is considered unavailable. Therefore, a resource group is the unit of failover.

Variables that define a resource of a given resource type. The action scripts use this information to start, stop, and monitor a resource of this resource type.

The simple name that identifies a specific instance of a resource type. A resource name must be unique within a given resource type.

A particular class of resource. All of the resources in a particular resource type can be handled in the same way for the purposes of failover. Every resource is an instance of exactly one resource type. A resource type is identified by a simple name; this name must be unique within a cluster. A resource type can be defined for a specific node or for an entire cluster. A resource type that is defined for a node overrides a cluster-wide resource type definition with the same name; this allows an individual node to override global settings from a cluster-wide resource type definition.

A set of resource types upon which a resource type depends. For example, the filesystem resource type depends upon the volume resource type, and the Netscape_web resource type depends upon the filesystem and IP_address resource types.

A list of resource types upon which a resource type depends.

The ordered set of nodes on which the resource group can execute upon failures, as modified by the failover script. The run-time failover domain is used along with failover attributes to determine the node on which a resource group should reside. See also initial failover domain.

See CXFS server-capable administration node

Each resource type has a start/stop order, which is a nonnegative integer. In a resource group, the start/stop orders of the resource types determine the order in which the resources will be started when FailSafe brings the group online and will be stopped when FailSafe takes the group offline. The group's resources are started in increasing order, and stopped in decreasing order; resources of the same type are started and stopped in indeterminate order. For example, if resource type volume has order 10 and resource type filesystem has order 20, then when FailSafe brings a resource group online, all volume resources in the group will be started before all file system resources in the group.

A port located on a node that provides a way to power-cycle the node remotely. Enabling or disabling a system controller port in the cluster database (CDB) tells FailSafe whether it can perform operations on the system controller port. (When the port is enabled, serial cables must attach the port to another node, the owner host.) System controller port information is optional for a node in the pool, but is required if the node will be added to a cluster; otherwise resources running on that node never will be highly available.

A node identified as a tie-breaker for FailSafe to use in the process of computing the FailSafe membership for the cluster, when exactly half the nodes in the cluster are up and can communicate with each other. If a tie-breaker node is not specified, FailSafe will use the node with the lowest node ID in the cluster as the tie-breaker node.

Required information used to define a resource of a particular resource type. For example, for a resource of type filesystem you must enter attributes for the resource's volume name (where the file system is located) and specify options for how to mount the file system (for example, as readable and writable).