Hello experts,
We are running OBM 2023.05 with an HA architecture, and the latest hotfix installed is OBM202305HF20240207T092715Z. Our Operations Team, who monitors the Event Browser 24/7 for Critical and Major events, has reported that some events appear in the console with a delay. The events always display the correct "time created" and "time received" timestamps, but they show up in the console with an age of already 15 or 20 minutes.
We found this very odd and had trouble finding any evidence of this issue. However, we discovered something interesting when looking into the integration with OO. We use Automatic Runbook Executions to integrate with OO (2023.05), which is triggered for many events in the production environment. In the cases where we observed the described delay, the execution was triggered much later than expected. This seems to corroborate the Operations Team's observations, as the OO integration should trigger as soon as the event is created.
For example, if an event has timestamps like this:
8/10/2024 11:13:25 PM
8/10/2024 11:13:26 PM
8/10/2024 11:13:26 PM
8/11/2024 01:03:26 AM
8/11/2024 01:03:26 AM
The first entry displayed in the history tab would be something like this:
System (System) - 8/10/24, 11:29 PM Added Annotation with valueFlow "OBM Notifications" launched in Operations Orchestration. See <https://$oo_server:$port/oo/drilldown-ex.html#$flow_id>
This issue occurs sporadically, but we have no idea how to troubleshoot it further. Could you advise on which logs to check, so we can increase the log file size and space allocation, or suggest any tests we could perform, bearing in mind that this is a production environment?
Thank you in advance.