Having an issue with the PA daily data collection job.

[Joel Pomales1]

Hello all.

We recently upgraded our sub-production environments to London. We noticed that the PA incident daily data collection job is errorring out. The error happens at the beginning of the run, and the job completes. We can't find the SysID referenced in the log anywhere. 

There is an entry on the indicators related to the job that appears as (empty). However, the SysID of that entry does not correspond to that entry.

Weirdness!

[Uncle Rob · Accepted solution]

Have you validated that the Incidators table has no Indicator for that sys_id?  
Keep in mind that when you're looking at the "indicator" records at the bottom of a job, you're not actually looking at indicators.  You're looking at an m2m record linking indicators to jobs.  That might explain the "this sys_id isn't that" weirdness you describe.

Check your deleted records to see if you've got any deleted indicators.  Sounds like maybe one was deleted but the delete didn't cascade to the job-indicators m2m table.

[Adam Stout]

Look at the "Indicators" related list on the job. Is there an entry with a blank indicator?  You'll want to figure out if you need that or not.  If you do, figure out how to restore it from deleted records, if you don't, you can just remove the blank entry.

[Uncle Rob]

Echo! 😄

[Joel Pomales1]

Yup. There was an empty record in the related indicators for the job. So I removed it and it ran without issue. It didn't occur to me removing it the first time since the SysID in the log didn't match that record, but if it was an m2m issue then that would be the issue.