We have had an ongoing issue in our version 7.0.1 environment. We have had a couple of support tickets open and the issue has yet to be totally resolved. A couple of months after we deployed this version from a fresh install we started having issues with our post-sync dbqueue tasks. It's taken us a while, with the help of support, to narrow down which task was actually causing the hang up, but we have narrowed it down to the 'Roles and Organizations' task, sort order 720550. This has not been 'officially' confirmed, but that is what appears to be happening. Sometimes these tasks will take days to process and it holds up all of the other dbqueue tasks, as well as any adhoc projection updates to our target systems. We thought that we had the issue resolved after performing two actions:
1. Enabling merging on a number of the membership tables (which is default in 7.1 as I understand)
2. Enabling the Trace Flag 2453 on the DB, after seeing the warning in the journal and per support's recommendation
Upon enabling these two features, performance seemed to have greatly increased. The entire sync time, including post-sync dbqueue tasks, went from days to under 2 hours. Daily scheduled syncs ran great for over a week. In the last couple of days, however, the issue has come back. There has been no major changes to the target system, 1IM, or the companies organization during this time. Post-sync tasks have been queued up for over 24 hrs, the target sync ended 26+ hours ago, and the dbqueue appears to be slowly processing task 720550. It's been doing this all night and all of this morning.
I have two questions:
1. Are others seeing this, or similar, behavior? If so, were you able to resolved it by upgrading or some other solution?
2. Can anyone offer recommended methods to monitor the DBQueue? Finsing out that the DBQueue is 'stuck' 12+ hours later isn't going to fly.
