Servicedesk Incident Resolution Time is Very Slow
Recently, a Servicedesk 7.1 Sp1 user encountered a very perplexing problem. When resolving incidents, the amount of time required for the process viewer to refresh and show that the incident is resolved or closed is normally a short time (2-3 mins). However, this users system was not showing the incident resolution until anywhere from 14 - 40 minutes after the Close Incident task was completed.
Thru SQL profiler trace, and Info level log review, it was discovered that this issue surrounded incident escalations. Further analysis showed that in the SD.IncidentManagement project, in the Initial Diagnosis Model, under the Event Configuration tab of the Initial Diagnosis dialog workflow component, in the Escalation Configuration section, there was an escalation named “Scheduled”. It has a decision model with 40 elements in it, and was running every 10 minutes. It was designed to be sent out for a scheduled re-opening of incidents that had been put on some form of hold. It was an older process and, for the most part does not get used, and was removed in the 7.1 Sp2 versions of Incident Management.
This is what was slowing things down. In situations where there are large numbers of open incident processes, this ended up running over the top of itself.
the way to bring speeds back up to expected performance is to remove the "Scheduled" escalation from the Escalation Configuration. Save the project and republish.
The Endpoint Management Community Blog is the perfect place to share short, timely insights including product tips, news and other information relevant to the Endpoint Management community. Any authenticated Connect member can contribute to this blog.
Comments 1 Comment • Jump to latest comment
To add a few comments to BRING's article on how you see this behavior.
We saw this behavior escalating over time as the use of ServiceDesk got more "heavy".
So here are some numbers on open incidents and resolution time with SD 7.1 SP1,
we observed together with our customer.
So if you have more than 300 open incidents and running SD 7.1 SP1, it might be a very good idea to implement this change BRING has described in the article above.
Thanks to BRING and his team to solve this issue, which hasn't been an easy task.
Would you like to reply?
Login or Register to post your comment.