-
Joined on
2025-08-13 -
-
Yeah our infra deployments should be fine, I'm more worried about external users such as CentOS, CoreOS, ELN, etc.
the proper fix would be to have Ipsilon use thread-local variables
I've looked at https://www.rabbitmq.com/docs/upgrade and here's the plan I've come up with. It's the longest but it should avoid downtime or message loss.
I think we should keep ca-combined.crt, it didn't work because of a bug in fedora-messaging but not because the principle of having multiple trusted CAs doesn't work at all. I'm delaying the…
If the problem happens again, please reopen this ticket.
OK I think that fixed the issue. Of course the proper fix would be to have Ipsilon use thread-local variables instead of class attributes, using for example cherrypy.thread_data, but that…
I've set mod_wgi to 1 thread only (and added 2 processes to compensate). Let's see if it helps.
Hey folks. I have spent a few hours on this, I think I may know where it's coming from. I think it's a bug in Ipsilon.
52f39daff4
This is related to #13323 that I fixed upstream with the latest commit. It's not urgent but we should…
I suppose we could start by making ipsilon02 a backup server and monitor the load on ipsilon01. Hopefully the current IPA status will not mess up the mesurements.
It looks like the user was being sent randomly to ipsilon01 and ipsilon02 while in the middle of a transaction, which apparently caused issues. They disabled "tracking protection" on id.fp.o in…
