[ISS Support Case] Shelved alarm count mismatch

Description

The number of shelved alarms as identified by the platform engines do not match the visible shelved alarms. This happened after upgrading from 2014 to 2020 R2 SP 1 and has been continuious ever since. The only time the correct alarms are shown is after an undeploy and redeploy. Here are some of the questions I have asked: Isolated incident, or has this happened previously? If it has happened before, when and how frequently has it occurred? I dont believe this has happened before. There was a network disruption which appeared to precipitate this. Does refreshing the query help? The thing that fixed the issue was restarting the platforms that the intouch instances run on. Does changing the query and then switching back to the original query help? I did not try this. This looks like its one of the Situational Awareness controls (which is fine) just curious if they happen to have a normal EAC that they use, and if it shows the same results? I dont know what EAC means. This issue appears to have started when there was a network disruption (one of our core switches went offline), and tags stopped recording historical data (i.e. the trends for tags associated with a few of our app servers that lost comms were flatlined). I had to fail over process engines associated with the app servers that lost network connectivity during the switch outage. Once I failed over the process engines, the historical data repopulated; however, the shelved alarms associated with the process engines disappeared (which is why I opened the case in the first place). And ultimately, restarting the platforms that the InTouch instances were running on fixed that issue, and the shelved alarms reappeared on the SA control. This is not the first time that a network disruption caused a slew of bizarre behavior that was fixed by restarting platforms or redeploying objects until things went back to normal, and it probably wont be the last. I wish System Platform was more fault tolerant when network disruptions occurred, but seeing as this happened several times when we ran SP2014 R2 SP1 and is continuing to happen in SP2020 R2 SP1, Ill just accept this behavior as a limitation of the product suite. You can go ahead and close out this ticket for now, and Ill do more restarting/redeploying platforms and objects before opening cases in the future Screenshots and Hf's https://insourcess.box.com/s/a9wlg693sxdmy3lqoffl8hyw1fal4od3

• Author: Kevin Modlin
• Published: March 6th, 2025

Details: