Server-side log of Management Agent alerts

[Steve Kadish]

Is a log written anywhere on a server of the alerts/traps generated by the HP Management Agents?

I know that there are entries made in the system event log, but one of my servers has a corrupted system event log.

A log would be extremely useful/important for investigation and remediation.

Thanks,
- Steve

[David Claypool · Accepted solution]

If the OS is down, major environmental and POST errors are written to the Integrated Management Log that is in NVRAM in every ProLiant. With the OS down, it is accessible from iLO.

[Steve Kadish]

Hi David,

Thanks for the info. How about less major alerts - things like disk threshold violations, application exception traps, etc.?

Thanks,
- Steve

[David Claypool]

The amount written to the IML is limited by size, so only critical life-threatening events are written to it. The assumption is that if the problem is not going to take the box down, the agent redundantly writes to the Windows event log and sends an SNMP trap to all defined destinations (up to 12).

[Steve Kadish]

Thanks, David. I guess that is sufficient in most cases. I was looking for another level of redundancy, mainly to double-check the SNMP trap process (i.e., if the trap never arrives at its destination). It's just bad luck that I have a server with a corrupted Windows Event Log and therefore lost all of those notification entries.

Thanks,
- Steve