IBM SG24-5131-00 User Manual
Cluster Testing
141
• Monitor cluster logfiles on NodeT if HACMP has been customized to
monitor 7133 disk failures.
• Since the 7133 disk is hot pluggable, remove a disk from drawer 1
associated with NodeF's shared volume group.
• The failure of the 7133 disk will be detected in the error log (
errpt -a |
more
) on NodeF, and the logical volumes with copies on that disk will be
marked stale (
lsvg -l NodeFvg
).
• Verify that all NodeFvg file systems and paging spaces are accessible (
df
-k and lsps -a
).
• Plug the failed disk back in, then sync the volume group (
syncvg NodeFvg
).
• Verify that all NodeFvg file systems and paging spaces are accessible (
df
-k
and
lsps -a
) and that the partitions are not stale (
lsvg -l NodeFvg
).
6.2.5 Application Failure
By default, HACMP does not recognize application failures. With some
additional configuration, it is possible to “teach” HACMP application failures
and trigger events (for more information see 2.6.2.3, “Application Failure” on
page 47).
additional configuration, it is possible to “teach” HACMP application failures
and trigger events (for more information see 2.6.2.3, “Application Failure” on
page 47).
So, the way of testing application failures is strongly dependent on your
configuration.
configuration.
Before you start to do the configuration and the testing application failure
notification, analyse your application for possible failures. Then try to
reproduce them.
notification, analyse your application for possible failures. Then try to
reproduce them.