Server Management - Systems Insight Manager
1753819 Members
8148 Online
108805 Solutions
New Discussion юеВ

Re: HP SIMM ESX Agents (Heartbeat Loss)

 
Leon Warren
New Member

HP SIMM ESX Agents (Heartbeat Loss)

Hello, hoping someone can help me please or point me in the right direction... We have multiple ESX 3.0.2/3.5U2 hosts with the HP SIMM ESX Agents installed. My question is, is there someway of changing the timeout values for heartbeat loss traps? We're getting hundreds of messages a day with heartbeat loss for Virtual Machines when there isn't any problem with them. I've ruled out VMM as we're getting them on hosts without it installed and it isn't any local HP SIM servers we're using as I've disabled them. Does anyone have any ideas or suggestions please? I would be grateful for any feedback. Thank you.
3 REPLIES 3
Derek_56
Valued Contributor

Re: HP SIMM ESX Agents (Heartbeat Loss)

Leon,
Is Virtual Center involved in your virtual infrastructure? The VM heartbeat alarm settings can be "tweaked" in VC by altering both trigger settings for severity and the tolerance and frequency under the Reporting tab of the alarm. The last option is under VC 2.5. VMware has extended some flexibility to account for this type of scenario exactly.
If you are referring to losing connectivity to the hosts themselves, you can tweak that in SIM by altering the polling interval for the (http or icmp) ping.
Hope this helps,
Derek
Leon Warren
New Member

Re: HP SIMM ESX Agents (Heartbeat Loss)

Hi Derek! Thanks for the reply... No VC isn't being used to send any traps. The traps are being sent directly from the ESX hosts with the HP SIM Agents installed. The VMs themselves are fine and we haven't had any issues with them. However, we keep gettong so many heartbeat loss traps from them that it's driving us bonkers. Is there a file somwhere I can edit on the HP agent or within ESX where I can extend the timeout for a heartbeat loss? Thanks again for you help!
Derek_56
Valued Contributor

Re: HP SIMM ESX Agents (Heartbeat Loss)

Leon,
Could you post the full text from one of the traps? That may help track it back to a particular MIB. We don't use VMM since we've got VC, but if you're receiving them from hosts w/o VMM, I'm guessing it's the VMware SNMP agents. For us, that's configured through VC. I'm sure there's a file on the host that would configure that.

As long as it's not a bandwidth (IE, just resource) issue for your trap receiver, you could instruct the receiver (I'm assuming SIM) to ignore the hearbeat alerts. Kind of an ostrich approach, I agree.