Monitoring physical drives with Veeam Monitor...

[MarcBesken]

I downloaded Veeam Monitor Free Edition v4.0.1.156 in the hope to monitor physical drives connected to a RAID-controller on an HP ML370 G3 server system.
I am using VMware ESXi 3.5.0 U4 and I am able to see the controller and the status of physical drives in the Configuration -> Health tab.

I was able to monitor the Datastores with Veeam Monitor Free Edition, but is it also possible to monitor the phyisical drives?

I don't want to be informed when the Datastore fails, as in this case the hardware RAID already failed. I would like to be informed about a failed physical hard disk from a physical RAID-drive in my VMware system and would like to be informed via mail or whatever to be able to react accordingly.

Is this possible with Veeam Monitor Free Edition and if yes, where can i monitor physical drives and not Datastores?

Best regards,
Marc

[Gostev]

Marc, check if you see your drives in the Hardware Monitoring tab. If you see them there, then you will be able to monitor them via alerts. If you do not see them there, it means that your ESX server does not implement physical drive monitoring via SMASH - in that case you will not be able to monitor them.

[MarcBesken]

Hello Gostev,

thanks very much for your prompt reply. What you mean? Hardware Monitoring tab in Veeam Monitor Free Edition or do you mean in the VMware Infrastructure Client under Configuration -> Health Status ?

I can see the status of the controller and physical drives under Storage in the VMware Infrastructure Client!

Best regards,
Marc

[Gostev]

Marc, I meant Hardware Monitoring tab in Veeam Monitor.

[MarcBesken]

Ok, found it and the controller and physical drives are visible under Storage. Now, I must find out how to set alarms and how to send mails in case of emergency!

Thanks for the moment

Best regards,
Marc

[MarcBesken]

Ok,

will try to comment my results. I used the Alarms tab to configure alarms. Under General I chose ESX server, under Rules I didn't really know what to select other than 'Hardware sensor state changes' ? There is no special choice for e.g. 'physical drive status changed'!

And there is already a standard or default alarm for 'Hardware problems' configured! Is this the alarm that reports hardware problems (like failed physical disk drives) via mail, when I configure the 'Mail settings' under the 'Mail' tab under 'Client settings' under 'Options' ?

And what kind of message is sent? Hardware problems in general or any specific message about the hardware problem?

Best regards,
Marc

[Gostev]

Marc, check out the default alarm settings and verify if this alarm is set to send emails.

If I recall correctly, the email will be like this:
Hardware-Sensor-Name has changed from 'GREEN' to 'RED'

[MarcBesken]

Ok, will check again!

Regards,
Marc

[MarcBesken]

Hmm strange,

on 2 different HP systems, namely ML370G3 with SA6400 controller and DL380G3 with SA532 controller, I just get an info, when a rebuild of a failed RAID-drive is occuring! I don't get any infos, when I pull the drive manually?

And suddenly I couldn't reproduce the alarm, when reattaching the drive on both machines?

When all RAID-drives are fully exisiting and synchronized, I can see the controller, the physical drives and the logical RAID-drives. When I pull a drive, the health status in Virtual Infrastructure Client and in Veeam Monitor didn't change at all. When I reattach the drive in both machines, I can only see the controller until the synchronization is finished and I get all details again.

Very strange behavior. I thought, I would see the corresponding drive going yellow or red and return to green but seeing nothing but the controller is disappointing.

Maybe I have some comprehension problems with the alarms and settings but at the moment I am a little bit unsure, if it's hardware related or Veeam Monitor related.

Thanks for any advice.

Regards,
Marc

[Gostev]

Marc, this most likely indicates SMASH/Server Management API implementation problem on your server, sounds like your HP servers do not implement hardware status for individual drives (or implement it incorrectly).