ISO8859-1%OH]p"J r 2 ?H # ;  3 I <@ ) ' Q : kb = )GRqGc*8c5{ Z!' "44#>i$%VControls the timeout for coordinated configuration changes per phase per node. A positive value indicates the maximum number of seconds CAA will wait on the execution of client side callouts including scripts and CAA configuration code. A value of zero disables the timeout. Controls the behavior of the node monitor which periodically evaluates the set of activated heartbeating sources to determine whether a node is UP or DOWN. Specify the number of milliseconds that the node monitor must wait after it makes the determination that another node is DOWN before posting it as DOWN. Controls the frequency with which the node heartbeats across the various enabled heartbeating sources as defined by the hb_src_xxx tunables. Specify the number of milliseconds after which another node may consider node X to be DOWN if it receives no incoming heartbeats from node X. CAA uses this attribute to control the frequency of Gossip Packets, the aggregated heartbeat message sent by gateway servers in linked clusters as well as the interval- based heartbeat algorithms that tick over SAN and disk. Applied as a delta to the node_timeout tunable for nodes in remote sites. The node monitor will wait node_timeout + link_timeout milliseconds before determining that a remote node is DOWN. Then, it will wait an additional node_down_delay milliseconds before posting that the remote node is DOWN. Controls the behavior of the deadman timer. The interval associated with this timer is the node_timeout tunable. When the value is set to "a" (assert), the node will crash upon the deadman timer popping. When the value is set to "e" (event), an AHAFS event is generated. A node-specific setting trumps the clusterwide setting and may be used to override the behavior on a per-node basis. Controls node behavior when cluster repository access is lost. When the value is set to "a" (assert), the node will crash upon losing access to the cluster repository. When the value is set to "e" (event), an AHAFS event is generated. A node-specific setting trumps the clusterwide setting and may be used to override the behavior on a per-node basis. Sets the order in which network interfaces should be used. A value of 0 restricts the interface from use for CAA communications. Positive values are treated as priorities, with lower numbers indicating higher priority. Interfaces with the same priority will be used equally. Interfaces with different priorities will be ordered in a failover mode. Example: Setting all priorities to 1 will cause all interfaces to be used for communications. Setting en0 to 1, en1 to 2, and en2 to 3 will cause en0 to be used exclusively until it fails; then en1 will be used until it fails; then en2 will be used. How infrequently to send a message to remote site. A value of 10, means every 10th message will be sent to the remote site. Policy of used by orchestrator to merge sites when out of sync. p = Priority, h = Heuristic, m = Manual The time-to-live value of CAA packets. A value of 32, means the time-to-live value located in the IP header will be 32. Controls interval for the node monitor and deadman timer. The node monitor will wait node_timeout + link_timeout milliseconds before determining that a remote node is DOWN. Then, it will wait an additional node_down_delay milliseconds before posting that the remote node is DOWN. Also the deadman timer will pop for a node after node_timeout milliseconds of node isolation, reacting depending on deadman_mode. Policy used by orchestrator to merge sites when out of sync. p = Priority, h = Heuristic, m = Manual Communication mode used by CAA for heartbeat and other protocol message transfer. When the value is set to "m" (multicast), CAA uses multicast communication mode. When the value is set to "u" (unicast), CAA uses unicast communication mode. Monitors network traffic to determine the state of network interfaces. When the tunable is set to 0, CAA monitors the interface traffic. When the tunable is set to 1, CAA does not monitor the interface traffic. Policy used to merge nodes within a site. n= None, m = Majority Tuning the ethernet interface failure detection time Default is 0, which uses the existing behavior of marking an ethernet interface down immediately on hard failures and waiting at least 5 seconds prior to marking an ethernet interface down on soft failures. If set, CAA will wait for a minimum of the specified interval prior to marking an ethernet interface down for both hard and soft failures. n = None, h = Heuristic, m = Majority p = Priority, h = Heuristic, m = Manual , n = None Allows CAA Disaster Recovery(DR) to be handled automatically. When the tunable is set to 0, Manual steps are required to recover on the DR site. When the tunable is set to 1, CAA automatically corrects the repository in the DR site. A positive value indicates the maximum number of seconds CAA will wait on the execution of client side callouts including scripts and CAA configuration code.