Hi all,
We've been experiencing issues with the History probe on a particular IBM-I Server.
Usually experienced after an unexpected reboot the History probe starts to read old QHST files and then takes days/weeks to catch up to the latest file.
This then causes false/late alerts. We use the History probe to mark an event taking place with informational alerts, which a script then looks back in the alert history for. If the alert is there, great, no action (job completed as expected), if the informational alert is not there - generate a visible alert.
The only fix I know for this is to back up the config, delete the History probe (note the probe doesn't remove cleanly, it shows in a Yellow state in IM and only a restart of the Nimbus service will clear it). Once the service is restarted the deletion is complete and there is no trace of the History probe in the file system. I then redeploy the History probe and then the backup template and everything continues fine. Until the next time.
Today, I saw the issue happening again, but this time the system has not had any unexpected reboots and hasn't been IPL'd. But the above process was needed again.
This probe has other issues too (listed as a known issue in the archive on support.nimsoft.com) and the probe hasn't been updated since 2015!
I've reported issues with this probe many many times with support, but keep getting told that it's in the pipeline to be looked at..
Does anyone else have issues with this probe? Or any ideas on what could be causing this strange, intermittent problem?
Regards,
Sam
------------------------------
CA - UIM administrator
------------------------------