CONTAP-608818: Incorrect ASUP sent after WDR: Call home for CONTROLLER TAKEOVER COMPLETE MANUAL
Issue
- BMC reset would trigger ONTAP panic.
- However, the node that got reset by BMC was not in a state to talk to partner via IC about its HA state transition from SF_UP to SF_DUMPCORE.
[?] Tue Oct 07 00:08:49 +0000 [xxx: idle: cpu23: cf.fm.localFwTransition:debug]: params: {'prevstate': 'SF_UP', 'newstate': 'SF_DUMPCORE', 'progresscounter': '28'}
[?] Tue Oct 07 00:08:49 +0000 [xxx: idle: cpu23: *ha.panicInfoNotSent:notice]: Local node cannot send a panic information message to the HA partner over the HA interconnect due to error: Not HA, or not ESX/MVRD layer 3. Partner name: xxx. Partner system ID: 538233958.*
[?] Tue Oct 07 00:08:49 +0000 [xxx: idle: cpu23: cf.fm.localFwTransition:debug]: params: {'prevstate': 'SF_DUMPCORE', 'newstate': 'SF_SPARECORE', 'progresscounter': '29'}
- Before the partner completed the countdown to takeover, ONTAP initiated an SP power cycle to clear the error condition.
- It was unable to communicate with the HA partner, which would have caused callhome automatic.
- This triggered an expected hwassist trap to HA, which considered the power cycle as a manually triggered takeover.
