I have just recently upgraded from 9.1.1117 to 9.2.7.53 and I am noticing some different behavior with 9.2.7.53 than with previous client versions. Everything is still working, but as a long time BigFix customer going back to BES 2.0, this is the first time I noticed this client behavior.
When sending an action, clients first report the action as not relevant even if they are relevant for that action. Eventually as they report in the relevancy eventually changes from non-relevant to relevant and the action executes on the client
Example:
I am upgrading Symantec AV on clients and doing a pre reboot on workstations to apply any pending patches so I can then install the upgraded AV client. The action is sent with relevancy of true…no client should report non-relevant.
Eventually they all switch to waiting or evaluating and complete the reboot if no user is present, but since I have never seen this before, I wanted to see if anyone else is experiencing this as well.
This is actually not true. If a client has NEVER been relevant for any fixlet it will NEVER report on it. If it has EVER been relevant for it, it will report on any changes.
A Mailboxed or specifically targeted action that is sent to a computer will be reported on as Not Relevant, but only if it is Not Relevant. Of course some computers can change relevance over time. From the image there it appears that you are targeting endpoints and not using dynamic targeting (where you wouldn’t see this)
That has always been the logic. If you report on all fixlets you would overwhelm the database in many cases. For example my deployment is rapidly approaching 400K fixlets and my server wouldn’t be able to handle the resulting influx of report data if every client reported on any fixlet or action it was not relevant for.
In your screenshots, I don’t see any reporting as “Not Relevant”. “Not Reported” is different - it means the server has not yet received any response from the client at all, and that the server is still expecting a response. Once the server receives a response from the client, the status would update to either “Evaluating”, “Waiting”, “Completed”, “Pending Restart”, “Not Relevant”, “Constrained”, “Locked”, etc.
That may all be true…but my point is I have never seen the non-relevant numbers popup under the status window before when deploying an action. This notification is new with 9.2.7.53 when the machines are not selected dynamically, that’s all I am saying.
That is possible that we started reporting this in 9.2 where we did not previously. The only think I’ll comment here is its hard generally to target computers that were not relevant to the original fixlet so that alone is a bit suspicious.
Thanks for highlighting, I totally missed that in your previous screenshot. I agree this does look odd, that they would report non-relevant for “true”.
So does the “X targeted computers reported this action non-relevant”, start off with the total count of the computers you targetted and then keep decreasing as the clients eventually report back a status?
I jumped from 9.2.3 to 9.5.0 / 9.5.1 over the last few days, I never saw 9.2.7.
Yes, any time you deploy a non dynamic job, all the machines report back immediately as non-relevant and then slowly shift back to reporting a status. Even if the relevance is “True” or come from a Fixlet where they were relevant for that Fixlet.
It was just odd behavior that I have never seen before, so I wanted to post it on the forum.
