Overriding storage latency alerts for specific LUNs

[wiredworm]

Hello,

We're getting a lot of storage latency alerts from the VEEAM Management Pack.

We actually run multiple storage networks and at least one of these is known to have issues. Unfortunately we can't migrate away from the 'problem' storage any quicker than we are doing, so for the time being i'd like to override the alert but only on datastores which contain specific characters in their name.

Is such an override possible and if so could you walk us through the process please?

Thanks

Pete

[-=MACROS=-]

Hi Pete,

Actually if you means the messages from SCOM you can allways select one, with right maus click select override or set override. Then choose for "this object". In additional grid you will see all parameters of this monitor. Select once you want to edit and ajust value of your choice.

This schould be enoght.

Regards

Rodi


Gesendet von iPhone mit Tapatalk

[wiredworm]

Thanks Rodi,

You might have guessed but i'm kinda new tom SCOM. I did the education a year or so ago but sitting in a class and learning this stuff isn't nearly enough; now i'm getting my hands into the system i'm realising just how much I have to learn.

Pete

[-=MACROS=-]

Hi Pete,

In Case if you will set overrides but you dont yet know who to set override, here a little Step-by-Step manual.

• 1. Start System Center Operations Manager Console
  Image: Open Console

• 2. Select Module Monitoring and then click on Active Alerts
  Image: Select Active Alerts

• 3. Select Object to adjust value that ganarate Warning or Error Message,
  then select Overrides -> Override the Monitor, and finally For the object: in my case HV_89
  Image: Select Override

• 4. Select your Override Items (Column: Override), then adjust value that should be active (Column: Override Value).
  Finally select one of Management Pack there you will apply you overrid values.
  Tip: Create for each new adjustment its own Management Pack (it's also recommended by Microsoft). In this case you've to click on button New, enter Name of new Management Pack it can calls what you want.
  Then click on create, after MP or Management Pack is created, you can select it and at least click on OK button
  Image: Set Override Values

I hope this little step-by-step manual can help you to create you own overrides and get you SCOM a little bit cleaner from unwished error or warning messages, especially for trusted alerts.

Good Luck

Regards

Rodi

[wiredworm]

Thanks for the guide Rodi

Unfortunately at the moment I can't see any way in the override configuration to set it so that it only alerts on specific LUNs whilst ignoring others.

Any pointers on how you might achieve this would be gratefully received.

Best regards

Pete

[Alec King]

Hi Pete,

I think you need to leverage SCOM capability for Grouping objects. So create a Group of Datastores, you can use pattern-matching so that only Datastores with certain name are included in the Group.

And also note, that overrides are applied in a hierarchy, like this -
1. MP default as shipped by Veeam
2. Override for the whole Class (e.g. override for ALL datastores) will win over the default configuration
3. Override for a Group e.g. Group of Datastores I mentioned above - this will win over the 'class' override
4. Override for a specific named instance e.g. a specific Datastore - this will win over the 'group' override

So, you could for example, use a Class override to globally disable the Veeam Monitor for all datastore class, with an override.
Then you could use a Group override to re-enable the Veeam Monitor, for a group of datastores, also maybe with a new threshold.
Then further, you could use a Specific Instance override to re-enable the Monitor for a specific named datastore, with yet another different threshold.

Above method gives you flexibility to run the same Monitor with different options on different objects.
Note that the overrides are all set in the same override dialog (as in Rodi's screenshots) - the hierarchy is imposed by the target you choose (Class, Group, or Instance)

We have a great Whitepaper from MVP Cameron Fuller on Veeam on this topic - https://www.veeam.com/wp-manage-scom-using-groups.html. The WP is not new, but these methods have not changed in SCOM for several versions now, so it's still valid.

If you do need more help in setting up the groups and configuring overrides - just let us know!

Cheers
Alec

[wiredworm]

Thanks Alec.

With your suggestions i've managed to achieve exactly my goal, which was to dispense with many of the alerts we're seeing for a system which we already know has issues.

Following your advice I managed to create a new Dynamic Group named 'All 3PAR DataStore Volumes' which automatically grouped together data stores which had the characters '3Pr' in their name.

It was then very easy to just override the 'Virtual Machine Storage Latency Analysis' alert for all members of this group. The net result is that we now have far fewer alerts for today and we can focus on those instead of spending time digging through alerts when we already know our 3PAR SAN has some issues.

Many thanks for your help,

Pete

[Alec King]

Thanks for the update Pete - glad it's all working for you now!