[11:03] <bluca> #2056461 this one is a bit of a pain, as it of course started to appear at the same time as one of the qemu-only tests started to fail, only in noble -.-
[11:03] <bluca> are there any known workarounds aside from downgrading kernel?
[11:04] <bluca> looks like there's no bot around here https://bugs.launchpad.net/ubuntu/+source/autopkgtest/+bug/2056461
[11:25] <Skia> not yet, sadly
[11:26] <Skia> only thing I can say is that I have this on my radar, and plan on investigating that shortly
[11:29] <andersson1234> bluca: another obvious workaround which you're probably aware of is using lxd instead of qemu, which I've been doing since this issue has arisen
[11:29] <andersson1234> ah, wait, I see you said qemu-only, nevermind :D
[11:32] <bluca> yeah need to run on the noble kernel for this
[12:20] <bluca> are there proxy issues from the autopkgtest datacenter to salsa or something? been getting this for the past 2 hours, despite retries:
[12:20] <bluca> fatal: unable to access 'https://salsa.debian.org/bluca/systemd.git/': Received HTTP code 503 from proxy after CONNECT
[12:20] <bluca> can clone/fetch fine from here
[12:21] <bluca> doesn't seem to be the usual intermittent hiccups, happens on all the jobs
[12:22] <andersson1234> This issue is familiar to me - I'll look into it shortly :) 
[12:22] <bluca> thanks!
[12:48] <andersson1234> bluca: do you have an autopkgtest logfile you can point me to please?
[12:48] <andersson1234> this seems datacentre dependent 
[12:48] <andersson1234> or *feels*
[12:50] <bluca> andersson1234: https://autopkgtest.ubuntu.com/results/autopkgtest-noble-upstream-systemd-ci-systemd-ci/noble/amd64/s/systemd-upstream/20240502_110511_6a2a7@/log.gz
[13:00] <andersson1234> ah, a pesky issue has reared its head once again
[13:01] <andersson1234> working on a fix now 
[13:06] <andersson1234> bluca: I'd expect within the hour for the salsa.debian.org issue to be fixed.
[13:06] <andersson1234> Please ping me if this isn't the case!
[13:08] <bluca> thank you, will let you know
[16:54] <andersson1234> bluca: FYI I have a fix I'm testing for the issue you saw yesterday (jobs being lost with a restart of the workers)
[16:55] <bluca> thanks
[16:56] <bluca> still seeing the salsa problem as of one hour ago
[16:57] <bluca> triggered again just now to see if it persists
[16:58] <andersson1234> ah that's annoying, sorry, I'll look into that again now
[17:12] <bluca> actually looks like it recovered in the past half hour or so
[17:12] <bluca> I see jobs moving past the cloning and onto the compilation step
[17:12] <bluca> so the fix seems to have worked, thanks