Re-running Identification Log jam

In which Wildbook did the issue occur?
Snail Codex

What operating system were you using? (eg. MacOS 10.15.3)
Windows 10

What web browser were you using? (eg. Chrome 79)
Microsoft Edge

What is your role on the site? (admin, researcher, etc)
User Manager

What happened?
Myself and another User-Manager account are working through separate bulk import sightings. All was running smoothly until I selected to re-run the identification of a sighting while the codex was actively running an identification for that particular (sighting Identification started on Jun 19, 2024, 3:44 PM.) I re-ran the identification because I mistakenly set the hotspotter parameters to the wrong region in the initial identification. All subsequent committed sightings from myself and the second User are being queued behind the re-run. The first is queued behind 26 jobs, the next is behind 38 jobs and the third committed sighting is queued behind 65 jobs.

What did you expect to happen?
I thought a re-run command would halt the active identification job. Instead I log jammed the snail codex for both users.

What are some steps we could take to reproduce the issue?
Commit a sighting and run the identification, then before the identification job is done re-run the identification for that sighting and select a different region. The job is still “fresh” so maybe it just needs a night to work through the system?

If this is a bulk import report, send the spreadsheet to services@wildme.org with the email subject line matching your bug report

Hi @Kupaa

Great question! I was also under the impression that sending a new ID job before the original one completed would pull it out of the queue, but I checked with a teammate and they confirmed that doing this would just run both ID jobs in sequence.

I checked the Snail Codex queue this morning and it looks like the ID jobs completed.

Hi @Anastasia,

the jam seems to still be there.
Sighting a1d59ed4-a8e9-4d60-adb9-5134381cbd87 • Snails.wildme.org

I can view the sighting that jammed the system, but the identification has been stuck at 33% since the original post last week.

I elected to rerun the identification and now the job is queued behind 24,000 jobs. I am the only user active on the Snail Codex in the past few days, so there shouldn’t be any jobs running.

Thanks for letting me know. I’ll look into this today.

Sorry about the wait, @Kupaa. I’ll need to get back to you next week while we continue to work on this.

Hi @Kupaa

Your snails should have match results now. I did find additional sightings from June 19th that were stuck in ID and I’ll let you know once those are ready to go, too.

Thanks for hanging in there!

1 Like

These all have match results now. All stuck sightings should have completed now. If you come across any sightings in your account where old ID jobs seem stuck, re-running them should get them going again.

thank you @Anastasia

1 Like