IBM AIX HACMP SG24-5131-00, Hardware User Manual

Page: 137 / 240

Cluster Customization
119
event occurs only after a node_up_remote event
has successfully completed.
Sequence of node_down Events
node_down This event occurs when a node intentionally
leaves the cluster or fails. Depending on
whether the exiting node is local or remote, this
event initiates either the node_down_local or
node_down_remote event, which in turn initiates
a series of subevents.
node_down_local Processes the following events:
stop_server Stops application servers.
release_takeover_addr (If configured for IP address takeover.) Identifies
a takeover address to be released because a
standby adapter on the local node is
masquerading as the service address of the
remote node. Reconfigures the local standby
with its original IP address (and hardware
address, if necessary).
release_vg_fs Releases volume groups and file systems that
are part of a resource group the local node is
serving.
release_service_addr (If configured for IP address takeover.)
Detaches the service address and reconfigures
the service adapter to its boot address.
node_down_remote Processes the following events:
acquire_takeover_addr (If configured for IP address takeover.) Checks
for a configured standby address currently seen
as up by the Cluster Manager, and then does a
standby_address to takeover_address swap
(and hardware address, if necessary.
get_disk_vg_fs Acquires disk, volume group, and file system
resources as part of a takeover.
node_down_complete This event occurs only after a node_down event
has successfully completed. Depending on
whether the node is local or remote, this event
initiates either a node_down_local_complete or
node_down_remote_complete event.