My MOM databases must feel like a yo-yo by now, having moved three times in the past two months, and I am busy preparing for one last move before I finally kiss this old girl goodbye. The result of the moves on my groom jobs, particularly the job that grooms the OnePoint database (MOMX Partitioning and Grooming), has been rather interesting.
All the MOM jobs would run, as per the default schedule, and would report success. But the OnePoint database has been growing oddly, grooming completely over the week-end, but growing rapidly, as if it wasn’t being groomed, during the week.
I checked everything. Or I thought I had checked everything. I double checked the date in the GroomingSettings table. And the global settings in MOM. I double checked every single job that ran (other than the MOMX Partitioning and Grooming one, of course), read every article and generally drove myself crazy for a couple of days.
Then I checked the actual job itself, read the stored proc it invokes, and realised it relies more on the ReportingSettings table than the GroomingSettings table. I opened up the RS table, and found two entries – one referencing the old SQL server and one the new. I removed the entry to the old server and re-ran the groom job, and, voila! things look a lot more normal now 🙂
Stuff I was reading while troubleshooting: