I have had a question from one of our customers that I have come across at most DC-RUM installations that I have been at, and I figured it was about time to try to get a proper answer. I believe the customer is on 12.2, but I have seen this in 11.x versions through 12.3. From the customer:
Quick question – “Hardware Health” isn't showing up for a few of our AMDs in the RUM Console, do you know why that would be the case? The AMDs it’s not showing up on are physical and virtual. It doesn't seem to be related to the version, either. Did a quick search on the Compuware community site but didn't see anything relevant. The AMDs seem to be functioning fine, otherwise.
I have been on a customer's site with 12 virtual AMDs (but I have seen the same behavior with physical AMDs), all built to the same spec, version, OS, and in the same subnet, and there are some that always seem to show Hardware/Process Health in the RUM Console, some that vary, and some that seem to always NOT show Hardware/Process Health in the RUM console. ALL AMDs at any given time are functioning as expected in all other aspects, and when you look at AMD Statistics in the CAS, none of the AMDs have breaks in their health monitoring.
Any insight into why this data is shown in the RUM Console for only some devices some of the time would be much appreciated.
I am facing this same issue now. I have 40 identical AMDs, all virtual, and the host health shows up for about 35 / 40 of them. The AMDs haven't been upgraded and are 12.3.0 and everything else is 12.3.4. Any insight would be appreciated!
Are there data for these AMDs in CAS reports?
Are the amdstatdata files generated on these AMDs?
We noticed some problems at one of our customers with directory rights for the amdstatmerger process on the AMD. After upgrading the AMD the rights are mixed up and amdstatdata files aren't generated anymore.
I have two different scenarios - in both, the AMD stats don't appear on the CAS.
Either the AMDs do have CPU, Mem, and Disk in the RUM Console but all have the unknown threshold issue, or the AMDs have no stats.
Again, they're identical instances running on the same box, so I'm not sure why this happens.
Thanks for your response!
Sorry, Friederike, I forgot to comment on that part.
Half of the amdstatdata files are processed - the 6 AMDs in the same data center, not the 6 from the other data center. But, some of the ones that were processed, in this data center, still show the unknown threshold issue in RUM and don't report on the CAS.
I checked the amdstatsmerger.log on an AMD with unknown threshold issues and there were no errors.