Traps work but no email when hardware status is Major or Critial

[Mike Swank]

Hello,

I recently set up Systems Insight Manager 5.1 with SP1 and have several clients configured. I have set the email settings and added an event handler for all major and critial events. I have tested sending traps from the clients and this sucessfully generates emails. However, when I simulate a harware failure, by pulling a power supply on a client that I have sucessfully tested send traps, no email is generated. The fault, however, is noted in the SIM GUI.

I must be missing some little set up step. Can anyone give me a clue?

Thanks for any help!

[Rob Buxton]

Check in the Events for the server to see if a Critical Event was generated.
You might also want to check the version of the agents.
Under the Agents different sub-agents can be activated/deactivated so you may also want to check that as well.

[H.Y. Yiu]

You can check the results of the event task. My own event tasks also clear the event which is another way to see whether the tasks were activated.

[Mike Swank]

Hi,

Thank you for your replies. I did check the results of the event task. While I do see where I manually send traps, I don't see anything that is a status change, like when I unplug a power supply on a dual power supply system. The GUI does show a "major" indicator so it sees that somewhere. Am I correct in assuming that all emails are only generated from traps?

Sorry for the newb question, but is there anywhere you can see all traps that get sent to the CMS, not just ones realized by an event handler?

BTW - I'm monitoring agents installed on VMWare ESX.

Agian, thank you for your replies.

[H.Y. Yiu]

ok, I see what you mean. I think in your case, status change was the result of status polling, which does not generate any events so no emails will be sent.
usually, when a power supply is defect (you've pulled it out), the local Insight Agent on that server will sense it and generate an alert, which will results in some event logging and an SNMP trap.
so I think that your problem is in your local ESX Insight agents which somehow does not send SNMP traps or sending them to the wrong IP address, in case of an hardware error.