VMware Cloud Community
AndyBH
Contributor
Contributor

How can I acknowledge/reset triggered alarms ?

Hi,

After a clean install of VC2.5 I upgraded out three hosts to ESX3.5 and then "grabbed" them into VC acknowledging the "this host is already managed by a.b.c.d are sure" messages.

All seems (reasonably) well - I hadn't realised a VMTools upgrade would require a reboot - but now I have three VMs which are showing as being alarmed.

One has the Red Triangle the other two have yellow.

The alarms tab for each VM shows several triggered alarms but the VMs are fine now.

In the main tree view of VC however, the VMs still have the alarm icons rather than nice green "running" icons.

How can I force a reset/refresh of these ?

Thanks,

Andy

Reply
0 Kudos
13 Replies
ianmellor
Enthusiast
Enthusiast

Hi,

Those alarms are for "Host Memory Usage" and "Virtual Machine CPU Usage" and "Host connection State", unless you have custom alarms.. They will go back to the green icon once the memory and cpu have gone down. You can't reset the alarm. You have to solve them problem that is triggering the alarm.

Hope this helps.

Reply
0 Kudos
AndyBH
Contributor
Contributor

Ian,

Sorry - perhaps I was not clear.

I know the purpose of the alarms and dare say that when they were triggered (last Thursday!) there may have been a brief performance issue.

Both VMs are now running fine and well below the alarm thresholds and have been for days now.

Andy

Reply
0 Kudos
ianmellor
Enthusiast
Enthusiast

Hi,

Sorry about that. You can try re-starting service vmware-vpxa stop,

service mgmt-vmware restart,

service vmware vpxa start on your ESX Server. See if that clears the alarm.

Hope this helps.

Reply
0 Kudos
akmolloy
Enthusiast
Enthusiast

This is the third thread I've seen where people's alarms are not clearing. Mine aren't as well. I have to disable/renable the alrm to make it clear.

-Tony

Reply
0 Kudos
Teddo
Contributor
Contributor

This seems to be a common bug across all versions of VC. It's unfortunate that the only way to "fix" it is to manually go and disable and then re-enable the alarm trigger. This is only a temporary band-aid solution. The problem comes back, and it's very annoying to see these stale alarms and to repeatedly have to do the disable/enable thing to clear them. What's the point of having alerts that don't clear themselves after they are resolved? Vmware should take this seriously and either fix it or remove the alarm functionality altogether. The way it works now is pretty useless.

Reply
0 Kudos
jc-rush
Contributor
Contributor

I'm having the same problem in a fresh install of VC2.5 and ESX3.5. The only way to reset the alarms are to edit them or disable/enable them.

Anyone at VMware know if there's a patch coming for this?

Reply
0 Kudos
craigc
Contributor
Contributor

We are having the same issues with VC2.5 and ESX versions 3.02 & 3.5.

Has anyone heard of an update or patch around to resolve this?

Reply
0 Kudos
Teddo
Contributor
Contributor

There is no update or patch yet that I am aware of. Vmware acknowledges that it's a bug, but who knows when they will fix it.... The workaround as mentioned before is to disable and then re-enable the alarm to reset it.

Reply
0 Kudos
bjmoore
Enthusiast
Enthusiast

I'm having the same problems with VC2.5 and ESX3.5 (fresh installs of both, new database). I just think I'm fortunate to get the alarms at all considering when I did my 2.0.2 upgrade, they just disappeared totally!

Reply
0 Kudos
Teddo
Contributor
Contributor

That's good to know. We will be testing that upgrade shortly as well. Kind of sucks though. Not only is there no fix for clearing stale alarms in 2.5, but now I find out the upgrade also wipes out any customized settings I've added to my alarms. Great!! Smiley Sad

Reply
0 Kudos
ServerChief
Contributor
Contributor

Hi All,

Just came accross the same annoying problem as everyone else.

Solution (in my case):

Restart Virtual Center Services on your VC server (not ESX), that should clear the alarms.

I suspect, the alarms are kept in the VC database, may be there is a SQL query that can clear the alarm?...

Hope it helps!

Serverchief

www.serverchief.com

Reply
0 Kudos
jpjeffery
Contributor
Contributor

The earliest report of this problem I've found is from Oct 2007 and referring to ESX 3.0.2, VI 2.0.2. We're on ESX 3.5 with VI 2.5 and we're also seeing the exact same problem in July 2008...TEN months later.

Have VMWare at least recognised this as an issue? It seems ridiculous that such a bug that detracts from us noticing real issues has been left for so long.

Jeff

London

Reply
0 Kudos
harkamal
Expert
Expert

How can I reset "Alarm to Green" proramatically. Doing so manually from VIClient works, but i need to code it,

Thanks

HK

Reply
0 Kudos