Real-time performance monitoring and troubleshooting
Post Reply
rmancinelli
Influencer
Posts: 10
Liked: never
Joined: Sep 15, 2010 9:02 pm
Full Name: Rick Mancinelli
Contact:

Delayed Alarm

Post by rmancinelli »

Is it possible to delay the triggering of an alarm, or to specify the amount of time that alarm condition must exist prior to triggering the alarm?

For example, lets take CPU usage on a guest VM. Can I set an alarm that says "warn me if CPU usage >90% for 2 minutes"?

I ask because CPU, Disk, and other resource "spikes" can occur, but they are typically only for a very short period of time. We need not concern ourselves with some of these and would like to screen them out, if at all possible. Many other monitoring platforms have this feature, but I cannot see to figure out how to do it in Veeam Monitor?!

Thoughts?

Vitaliy S.
Product Manager
Posts: 25781
Liked: 2393 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:

Re: Delayed Alarm

Post by Vitaliy S. »

Rick,

You can configure a time delay for triggering the alarm, if you choose to edit any alarm, which is based on the performance source data, and navigate to Rules tab. Also you may use tolerance and frequency options which are available in the Reporting Tab of alarms settings.

Besides, you may want to use suppress option in order not to fire an alarm on certain time or event in the virtual infrastructure which might cause those spikes.

desmith
Influencer
Posts: 13
Liked: never
Joined: Dec 15, 2021 11:41 pm
Full Name: Derek Smith
Contact:

[MERGED] Delay rules for transient failures

Post by desmith »

Looking for some guidance on suppressing alerts for temporary connection failures, e.g. I don't want to hear about "Veeam ONE agent client connection failure" unless it stays failed for more than 15 minutes.

I'm assuming the solution is to disable notification for that alarm, create a new alarm, add a rule based on existing alarm with chosen delay then set notification there? Seems like a complicated option, and would also delay the resolve notification?

I did wonder if increasing the Trigger after on the error event in the original rule would work, but couldn't find anywhere to review all events to tell if and how often that error event reoccurs when there's a problem.

RomanK
Veeam Software
Posts: 99
Liked: 55 times
Joined: Nov 01, 2016 11:26 am
Contact:

Re: Delay rules for transient failures

Post by RomanK »

Hello Derek,

Disabled origin alarm would never trigger the dependent one.

The use case is unclear a bit. If you have some issues with Veeam ONE agent connection, the best way to proceed is to resolve them. If you need to suppress the alarm for a specific time this should help.

BTW the delay would not affect the resolving.

Thanks

desmith
Influencer
Posts: 13
Liked: never
Joined: Dec 15, 2021 11:41 pm
Full Name: Derek Smith
Contact:

Re: Delay rules for transient failures

Post by desmith »

Hi,

The use case is common in all monitoring systems, some errors are not a concern unless they persist for a certain period of time (as I said in my example).

Also, my thought was to disable the notification on the existing alarm, not the alarm.

If there's a better way to achieve the outcome I'd be happy to hear it, documentation on "Delay" seems to be limited to one sub-point on https://helpcenter.veeam.com/docs/one/a ... ml?ver=110

RomanK
Veeam Software
Posts: 99
Liked: 55 times
Joined: Nov 01, 2016 11:26 am
Contact:

Re: Delay rules for transient failures

Post by RomanK »

Hello Derek,

I see your point now. Indeed, due to the nature of some alarms, there are could be some obstacles to achieving the delay. In case you disable notification the discussed workaround would work.

I think it is a bit complicated now and should be improved in the future. Thanks for the feedback. It's noted.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest