Hi all – I’ve seen this before but can’t find any info on it or what it means. This is 2.37.8.
The job is scheduled and the status says “scheduled”, and the “next run” says “not scheduled”.
The behavior is strange, it was working, and now the job is skipped. Anyone seen this before and now a resolution or how to troubleshoot? Is there an endpoint for scheduled jobs I can look at?
We are currently experiencing this issue where scheduled jobs are not running automatically even though the configuration is in tact. Status says “scheduled” but next run says “Not scheduled”. A daily manual run is what has been implemented.
The client is using version 2.38.2.1 SNAPSHOT
Predictors and respective groups were created. Predictors push data into the next period
3 Schedules for 3 predictor groups were configured to run at 3, 4 and 5am respectively. Relative start is -1200 and relative end is 31
Analytics is set to run at 6am
It was observed that predictors were not pushing data automatically but when schedular is triggered manually, expected data becomes visible.
Thank for the response! To stay on the safe side, I think it’s important to mention that SNAPSHOT is not supposed to be used in production. Please try to use the latest stable release for that version instead of snapshot!
Thanks! I can’t guarantee that using stable release will resolve the issue (we’ll need to test there first) but as a start I’ve always seen issues posted when using SNAPSHOT instead of stable so you could even expect to face even other issues.
Interesting that Phil noted that it was overlooked–this was also true for us for 2.37.8.1-SNAPSHOT being the newest stable release, even with SNAPSHOT attached. Anyway. Clarity seems valuable. Sorry for the tangent!
We’ve discussed this a bit internally. The issue when it says Status: Scheduled, Next run: Not Scheduled occurs because the Next Run date on the server is populated at the time of the last successful run. If something goes wrong at some point, the next run date can fall out of sync, and then we display “Not scheduled” as a kind of fall back.
In terms of the UI, we can update this to display a Next run derived from the cron expression (this will be accurate in most cases, but the scheduler can be a little off, or it could be down and then the next run time that we estimate would be off).
We’re open to other suggestions though if you have ideas for improving the UI. If you’re experiencing this problem regularly, there might be a larger problem with the scheduler failing that we need to look at, so let us know if you have any thoughts. We also discussed briefly the idea of including information about the last successful run in this table, but we’re not sure if that would be helpful either.