Opened 9 years ago
Closed 9 years ago
#2455 closed defect (obsolete)
Timeouts when deleting larger experiments from Hive
| Reported by: | abeham | Owned by: | abeham |
|---|---|---|---|
| Priority: | medium | Milestone: | |
| Component: | Hive.Server | Version: | 3.3.12 |
| Keywords: | Cc: |
Description
After 10 minutes of waiting for the removal of a job a timeout occurs.
The job is then not deleted, there is no way to delete those jobs than by e-mailing an administrator. I understand that this may take time when the job is still being calculated as it needs to be stopped first, but when a job is finished removal should be much quicker.
One solution would be to just mark an experiment as deleted instead of actually deleting it and use the Janitor or another means to delete jobs in a batch later.
Change History (4)
comment:1 Changed 9 years ago by ascheibe
- Owner set to ascheibe
- Status changed from new to accepted
comment:2 Changed 9 years ago by ascheibe
comment:3 Changed 9 years ago by ascheibe
- Owner changed from ascheibe to abeham
- Status changed from accepted to reviewing
Now that we have deployed the new Hive service, can you test it again? Is it still an issue or does it work now for you?
comment:4 Changed 9 years ago by abeham
- Milestone HeuristicLab 3.3.13 deleted
- Resolution set to obsolete
- Status changed from reviewing to closed
I tried it and it worked. As there are no commits in this ticket, I'll close it as obsolete. Thanks!
This may get better with #2388 because on deleting jobs there is also statistics that is calculated. This has been removed so deleting jobs should be faster now. So before I will implement this I want to wait for the trunk integration of #2388 and see if this is enough improvement.