This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Can we configure Warning and critical threshold in a single alert?

Hi,

Maybe a very basic question but I need to know this.

I want to trigger a warning alert when CPU reaches above 80% and Critical alert when it goes above 90%. Now the question is can I configure this in a single alert, or I have to do it in two separate conditions?

  • really should be two separate alerts...in my env. and with our paradigm

    basically, when the alert is triggered as warning...it wouldn't re-trigger as critical...it's already triggered.

    What you could do, if it makes most sense in your env. to have them both in same alert....

       "escalate" when critical

  • Thanks njoylif

    Yea escalate option could be handy, will check that..

  • As a follow up to the separate alerts, if you had a separate alert for warning and critical thresholds, how do you handle not retriggering if one is already active?

    For example, let's say you have the following alerts:

    - CPU warning at 80%

    - CPU critical at 90%

    There are a few options:

    1. Leave both alerts with their basic trigger condition (CPU has breached X threshold). If CPU is at 85%, it will trigger the warning alert. If CPU climbs to 99%, it will then also trigger a critical alert and you now have 2 active alerts for that node.
    2. Modify the warning alert so it resets at 90%. I'm not a fan of this as it causes reset messages to be sent when the CPU drops below the threshold AND when it breaches the critical threshold. The bonus is you only have 1 active alert at a time.
    3. A single alert that alerts if you have breached either warning or critical threshold. However if you trigger on the warning threshold and the critical threshold is breached shortly after, you won't be notified as the alert is already "active". I think this is the alert you were alluding to but I'm not sure where you'd configure it to escalate if the critical condition is breached?
  • Hi Steven,

    Yes infact I was doing some testing with what I thought could be possible. But its better to have it alerted separately. Else it might create unnecessary probs...

    And for using escalate option, it would be just like notifying additionally with some text pre-written but again that doesn't actually serve the purpose.

    Thanks...

  • I meant njoylif​​'s reference to "escalation". Not the alert escalation. I was hoping he could expand on that.

  • I was referring to alert escalation, but gave caveat that I was not familiar with its capabilities, not using them currently emoticons_confused.png

  • well, meant to caveat anyways....oops.  sorry all

  • It would be nice if there were a way to trigger the alert based on normal/warning/critical status change so that this could all be done within one alert, as well as a "previous" status and value variables so that the email alert could specify the previous and current status/value.