Issue Number | 4130 |
---|---|
Summary | Failure verifying Push Job (on lower tiers) |
Created | 2016-07-07 18:42:54 |
Issue Type | Bug |
Submitted By | Englisch, Volker (NIH/NCI) [C] |
Assigned To | Englisch, Volker (NIH/NCI) [C] |
Status | Closed |
Resolved | 2018-07-06 11:41:12 |
Resolution | Won't Fix |
Path | /home/bkline/backups/jira/ocecdr/issue.187729 |
I'm seeing frequent publishing failures on the lower tiers with a
message that the previous job is still in process.
This is a result of the previous job not getting updated during the
verification process, so the job is still reported with a status of
'Verifying' in the pub_proc database table instead of getting updated to
a status of 'Success'.
This problem has started about 2 months ago and it appears that the
process is loosing network connectivity - likely at the GK end - and the
verification process continues until the status for that particular job
gets manually set to 'Failure'.
Tracking Failures:
07/11 - STAGE
The problem has been identified and a ticket created on the GK side.
I probably forgot but this problem had been identified as GK issue that turned out to be a result of OCECDR-3910. The reason this problem is now showing up more frequently is that the lower tiers of the GK server have been refreshed.
A temporary solution - at least for DEV and QA - is to reset the Job-ID in the pub_proc table using the following command:
DBCC CHECKIDENT (pub_proc, RESEED, NNNN)
where NNNN should be larger than the latest Job-ID on gatekeeper with the Source entry of CDR-PROD.
Story points include:
nagging WCMS team to pick up ticket WCMSGK-47
manual cleanup of the data while waiting for that to happen
This ticket is related to OCECDR-4215. If we're able to implement
that ticket (creating a web interface to reset the Job-ID counter) we
will be able to close this ticket.
I will have to find out if we're having the permissions needed to update
the Job-ID counter on STAGE in order to make this problem disappear.
~volker: see comment on OCECDR-4215.
~volker: please close this ticket, unless there's a reason to keep it open (in which case please explain).
The status Resolve as Won't Fix Issue isn't exactly right
because we've implemented a work-around.
Closing ticket.
Elapsed: 0:00:00.001238