How can we improve Change Tracking Feature in Azure Automaton?

New Management Pack releases and change tracking

When the OpInsights team releases new IPs/MPs they're automatically downloaded and executed on agents. While I'm sure you're performing all the tests needed, there can be disruptions on the agent side, just like it happens with OpsMgr. The main difference is that with OpsMgr the admin is in full control, with OpsInsights it is not. You should give the user a way to track MP changes, a few ideas:
- use the change tracking IP and add MP tracking
- use a newsfeed / tile on the portal with the latest release for every IP and core management packs. It should state the release date and a short abstract of what has changed.

It would be great to have both :-).

39 votes
Sign in
(thinking…)
Sign in with: Microsoft
Signed in as (Sign out)

We’ll send you updates on this idea

Daniele shared this idea  ·   ·  Flag idea as inappropriate…  ·  Admin →

4 comments

Sign in
(thinking…)
Sign in with: Microsoft
Signed in as (Sign out)
Submitting...
  • Ian Pickering commented  ·   ·  Flag as inappropriate

    IMHO, just like allowing OpsMgr to download & install MPs from the MS web service, it is poor admin practice to deploy new MPs without any change control. It would be great to be able to see the MP guide before any new MPs are released into OMS.
    Cheers,
    Ian

  • B Carlson commented  ·   ·  Flag as inappropriate

    We may not be able to revert changes, but at least we'd be able to use OMS for what it's good at, Analyzing our environments to determine "What recently changed, and who did it?"

  • Daniele commented  ·   ·  Flag as inappropriate

    The goal is to able to understand if an issue root cause can be related to a newly released MP. Take for example the issues we're facing with the latest wire data MP, it took some time to correlate the CPU spikes with the MP release. Let's say I have my CPU perf in OpInsights and I can see when the spike started, then I can time window into change tracking and see what's happened. 2' and I have a possible cause across al my systems.
    On the other hand, you're right I cannot do anything except open a support call, and this is anothjer topic I'm going to post: give us the ability to revert to a last known good configuration, or stage updates. Today I had to remove the WireData MP that was killing the DCs, this has been possibile since we're just evaluating the module, but if it happens once in full production? Not a good situation to be in, for the customer and for OpInsights credibility. OpInsights is susceptible to bad MPs as SCOM is, the difference is with SCOM I can always revert, I can have a staged process, multiple MGs, in a way or another I'm able to limit the damage.

Feedback and Knowledge Base