Datastore Usage on Disk Alarm?

[defrogger]

I setup the defualt Datastore Usage on Disk alarm to email when it triggers.

Now my understanding of the alarm is, if Datastore uses more than 75% of its space it will trigger then email me.  Actually 75% to warn, 85% to alert.

The alarm triggered. as one of my datastores actually is using above 90% of the space.

But the email confused me

Target: Sata_Datastore
Previous Status: Gray
New Status: Red

Alarm Definition:
([Yellow Metric Is above 7,500KB; Red Metric Is above 8,500KB])

Current values for metric/state:
Metric Space actually used = 9,617KB

Description:
Alarm 'Datastore usage on disk' on Sata_Datastore changed from Gray to Red

It says above 7500KB, actual used is 9617KB.

Is it just giving me wrong info, where the 7500kb should be percent?  or is the Datastore usage on disk mean something other than what I thought?

Thanks

Mike

[AnthonyF2011101]

Ok, from looking at it im getting the same thing and am seeing the percentage being shown in kb.

Is there a fix for this?

[Walfordr]

AnthonyF wrote:

Ok, from looking at it im getting the same thing and am seeing the percentage being shown in kb.

Is there a fix for this?

Try disabling the email notification and create a brand new alarm definition at the datacenter level with the exact settings.

For some reason after my upgrade the default "Datastore Usage on disk" alarm was not being triggered for me and creating a brand one worked, with the correct information.

[defrogger]

I never got it working, it was such a small thing that we never put the time into looking at it.

I know there is an update to esxi 4.1, so at some point we might update and see if it goes away.

When I have time ill try recreating the alarm at the datacenter level and see if that works.

Mike

[tmancini]

While I have not installed it yet this was supposedly fixed in vCenter 4.1 Update 1 according to the Release Notes;

Messages for Datastore usage on disk or Datastore Disk Overallocation alarms display metrics in kilobytes instead of percent
The text in SNMP traps or email messages for Datastore usage on disk or Datastore Disk Overallocation alarms is inaccurate. The message displays the disk usage value in kilobytes when the value correlates to disk usage in percent.

This issue is resolved in this release.

Unfortunately, this still remains an issue;

The Datastore usage on disk alarm might not be triggered after vCenter Server is restarted *
The Datastore usage on disk alarm or an alarm that uses the Datastore Disk Usage (%) trigger type might not be triggered if vCenter Server is restarted.

Workaround: Edit the alarms making trivial changes such as changing the alarm description, everytime vCenter Server is restarted.