Share via

SCOM distributed application alert

Niksa Jadric 96 Reputation points
2020-12-04T14:03:09.667+00:00

Hi community,

I am trying to accomplish an alert when Distributed Application changes health state.

I have overridden top level Entity Health monitor for the DA and that works fine, but I am getting alerts not just from this top level monitor but also every DA component and group Entity Health monitor that contains unhealthy object within this DA.
In other words, when an object within DA changes state, this state rolls up trough many Entity Health monitors until it reaches top level DA Entity Health monitor. All these monitors in a path generate alert!

My overrides are for this particular DA class (and instance), but for some reason, it overrides other Entity Health monitors.
I am still getting needed alert and i have created subscription using criteria, so that's not a problem, but I would like to stop other alerts and have only top level alert for the DA.

Any thought on how to achieve this is welcomed.

Niksa

Operations Manager
Operations Manager
A family of System Center products that provide infrastructure monitoring, help ensure the predictable performance and availability of vital applications, and offer comprehensive monitoring for datacenters and cloud, both private and public.
1,446 questions
0 comments
Accepted answer
  1. Niksa Jadric 96 Reputation points
    2020-12-09T15:13:26.567+00:00

    I have found an answer for this by following this advice:

    https://code4ward.net/2011/10/10/service-owner-notification-for-distributed-apps

    In short, we shouldn't override top level Health Entity monitor, but rather "All Contained Objects monitor".
    I did this for "All objects of another class: User Created Distributed Application" and now all my DAs alert on health state change.
    One of the caveats is that the rollup is "Worst Of" and there is no way to configure it further, but it works for us at this stage.

    We have further scoped this alerting in subscriptions so proper owners only get the alert.

    Hope this will help someone in the future, although it would be good to see how to have better control of health rollup.

    0 comments

2 additional answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Niksa Jadric 96 Reputation points
    2020-12-07T18:57:17.447+00:00

    Hi Stoyan,

    Thank you for exhaustive answer, but I couldn't find solution to my problem.
    My health rolls up as expected and my custom groups also roll up health (I have used Tao Yang's self-maintenance mp for that).

    The problem is multiple alerts, each for every component and group within the DA. I will try to explain better on my DA example below.

    One of the disk drives goes critical and health rolls up as desired all the way to the top.
    I have overridden top level Entity Health monitor (see bottom screenshot) for "DA Niksa NEW NA..." to generate alerts, and have enabled it only for this class (and instance since it is DA).
    This generates alert correctly, but also all others (Child DA, DA Niksa Storage N..., Disk compo, _child1 group) are generating alerts? I haven't enabled those alerts, but somehow the override applied to them as well.

    45874-untitled.png

    This is how I have overridden the monitor:

    45816-capture.png

    Because of this, I end up with 5 alerts, instead of 1. I also get 2 alerts for C: disk space:

    45690-22.png

    Only the first one is wanted, don't know why all these others (except the one for C:)

    0 comments
  2. SChalakov 10,371 Reputation points MVP
    2020-12-04T15:04:15.783+00:00

    Hi Niksa,

    I had the same case a few years ago and I can tell you that the greater the DA complexity, the more adjustments you need to make in order for this to work. Let me give you some helpful advices.

    • This is maybe a bit off topic, but if you have used custom groups in your DA, then you need to ensure that the health is being rolled up, otherwise the top level won't reflect the correct state. So you need to configure a health rollup monitor, so that the health si being rolled up to the group itself. This is described here:

    Monitoring the Availability and Health of a Group in SCOM
    https://sites.google.com/site/scomblogs/journal-blog/monitoringtheavailabilityandhealthofagroupinscom

    • Now to the point - depending on you DA complexity there are different approaches. Please read this following article carefully. Although it is some years old, the same principles apply when configuring proper DA alerting. This might be of great help to you:

    Using Distributed Applications to generate actionable alerting
    https://www.catapultsystems.com/blogs/using-distributed-applications-to-generate-actionable-alerting/

    So you will have to play a bit with Health Rollup Monitoring.

    • The next advices could be also a solution, depending on the DA structure. Please go through the following thread:

    Distributed Application alert on the whole application, wondering about 'extra' monitors
    https://social.technet.microsoft.com/Forums/systemcenter/en-US/25364886-6a49-496e-8fac-b5aded96828c/distributed-application-alert-on-the-whole-application-wondering-about-extra-monitors?forum=operationsmanagerauthoring

    That one could give you some ideas too:

    Health roll up issues in Distributed Applications in SCOM 2012 SP1
    https://social.technet.microsoft.com/Forums/en-US/4e34913a-c7a8-49cc-83ef-4501ec7661d8/health-roll-up-issues-in-distributed-applications-in-scom-2012-sp1?forum=operationsmanagerauthoring

    I hope I could help you with this!

    Best Regards,
    Stoyan

    If the Answer is helpful, please click "Accept Answer" and upvote it.

    0 comments