#9175 random "500 - Internal Server Error" failures @ FAS login(s)
Closed: Fixed 5 months ago by kevin. Opened 6 months ago by pgfed.

on an infrequent, seemingly random, basis, attempts to login via FAS login to infrastructure resources, FAIL with:

500 - Internal Server Error
Ipsilon encountered an unexpected internal error while trying to fulfill your request.
Please retry again.
If the error persists, contact the server administrator to resolve the problem.

it's inconsistent -- sometimes I can login @ RH Bugizlla, but not Pagure. or vice versa. or neither.

waiting 30 minutes or so, &/or simply retrying 10+ times, seems to 'fix' the issue.


this morning, I was unable to FAS login @ RH Bugzilla, getting that^ error.
FAS login to @ Pagure, was OK.

~30 mins later, all's sorted ... login @ both Pagure & RH Bugs is OK.

I understand the 'infrequent'/'random' isn't particularly helpful; @ #irc requested I file here anyway.

Metadata Update from @mobrien:
- Issue priority set to: Waiting on Assignee (was: Needs Review)
- Issue tagged with: groomed, high-trouble, medium-gain

6 months ago

Can you perhaps record here the exact time(s) you see the 500s? It might help track down the cosponsoring log entries...

Also, when it fails with bugzilla.redhat.com, can you try https://partner-bugzilla.redhat.com/ might give more datapoints to see where the issue is...

I'm getting these 500s right now from RH bugzilla. trying the partner-bugzilla returns the same ISE page from id.fedoraproject.org

to paste a time slice in the ticket, that's Aug 3 2020 20:34 ET

I just tried 10 min ago (so 15:05 CT) or so to log into BZ via FAS and i got a 500 error. I went to go post this info on this issue per requested in #fedora-admin. I had to log into FAS ( was totally logged out) first. I logged into Pagure via FAS just fine and then I was able to use that same login session to get into BZ without issue.

Hope this helps!

Have any of you seen this recently?

My theory is that it was varnish being overloaded because our container registry was using it instead of our cdn... if thats true, then it will have gone away.

Have any of you seen this recently?

not for quite awhile, now that you mention it ... of late, site logins have been reliable.

ok. I am going to close this now and if anyone sees it again, please re-open or file a new ticket. :)


Metadata Update from @kevin:
- Issue close_status updated to: Fixed
- Issue status updated to: Closed (was: Open)

5 months ago

Login to comment on this ticket.