[07:15] <LocutusOfBorg> hello, timeout during upload and syncpackage...
[07:19] <jugmac00> LocutusOfBorg: Looks like there is a major infrastructure issue. IS is aware of it.
[07:59] <LocutusOfBorg> ack thanks
[08:12] <LocutusOfBorg> I'll check for topic changes then :)
[11:20] <mapreri> cjwatson: https://launchpadlibrarian.net/615051648/buildlog.txt.gz - is this a LP problem?  "Could not open file /home/buildd/work/apt/./Sources - open (13: Permission denied)" and then just fails to build the source in that ppa recipe
[11:27] <LocutusOfBorg> and s390x is taking a rest again?
[11:31] <cjwatson> mapreri: I guess it must be - file a launchpad-buildd bug if there isn't one already please?
[11:38] <cjwatson> LocutusOfBorg: One of the two compute nodes is broken again.  I've reported it internally.
[11:40] <LocutusOfBorg> I hope such reports are useful, otherwise let me know, and I'll stop reporting them, maybe you already have tools to understand in advance if something breaks, other than "waiting for me to report here" :)
[11:56] <cjwatson> In principle we have alerts for this but they happen to be broken right now, IIRC following a Grafana migration
[11:56] <cjwatson> Need to sort that out ...
[12:19] <mapreri> cjwatson: couldn't easily find one already filed, so https://bugs.launchpad.net/launchpad-buildd/+bug/1983021
[15:28] <mapreri> guruprasad: sev=low? :(
[15:59] <guruprasad> mapreri, I don't know much about the details of this issue and marked it as low mostly as a rough guess. But I have already asked cjwatson, if I should bump it up.
[22:04] <ted> Howdy, I think the Inkscape snaps are timing out at 2 hours. Is there a way to request a longer timeout?
[22:04] <ted> https://launchpad.net/~ted/inkscape/+snap/inkscape-1.2.x/+build/1825901
[22:07] <cjwatson> ted: That's not a timeout LP is imposing, it's ... something else.  Some kind of LXD failure?
[22:08] <cjwatson> Not something we're in control of, anyway.
[22:08] <cjwatson> It might not be deterministic?
[22:08] <ted> Hmm, not sure. I figured it was LP keeping me from mining crypto 😆
[22:09] <cjwatson> Heh, no, that timeout is longer.
[22:09] <cjwatson> And you're not being timed out by the proxy here anyway.
[22:09] <ted> Is there a way to get more entries? It started breaking a week or two ago, but I'm not sure when.
[22:09] <cjwatson> I dunno, this is a couple of levels down from us.
[22:10] <cjwatson> Maybe?  It's a bit hard to tell
[22:10] <cjwatson> I don't really know what lxd means when it says "Error: websocket: close 1006 (abnormal closure): unexpected EOF"
[22:10] <cjwatson> Though that "Killed" there looks like maybe the OOM killer got involved?
[22:11] <cjwatson> snapcraft really shouldn't be running into the OOM killer, so maybe something is spiralling out of control?
[22:11] <ted> Yeah, was just thinking about that.
[22:12] <ted> The snapcraft output there is usually when I think it is handling things over to snap to build the actual squashfs.
[22:12] <ted> Which really shouldn't run out of memory more than for instance gcc.
[22:13] <cjwatson> Quite
[22:14] <ted> But, then, not seeing the actual compile in the log. So, feels like the log isn't complete.
[22:15] <ted> Just grabbing the packages shouldn't take 1h15m.
[22:16] <cjwatson> The last timestamped thing in the log is about five minutes after the build started.
[22:16] <cjwatson> The log's as complete as LP had available to it.
[22:17] <cjwatson> Really looks like it got to "Initializing parts lifecycle" and then sat there churning for an hour and change before the OOM killer killed it.
[22:18] <cjwatson> Oh, this is core22, so you'll also be running into latest snapcraft being inappropriately quiet
[22:18] <cjwatson> https://bugs.launchpad.net/snapcraft/+bug/1981072
[22:19] <cjwatson> That's a snapcraft regression and I think they're working on it
[22:19] <ted> Oh, I bet this broken when snapcraft 7.x came out...
[22:19] <cjwatson> Yeah, very plausible
[22:19] <cjwatson> If you dig back you'll probably find the snapcraft version in successful/failing build logs
[22:20] <ted> Trying a build with 6.x snapcraft
[22:21] <cjwatson> https://launchpad.net/~ted/inkscape/+snap/inkscape-1.2.x/+build/1788763 was the last successful build, https://launchpad.net/~ted/inkscape/+snap/inkscape-1.2.x/+build/1796416 the first failing one
[22:21] <cjwatson> Success was with snapcraft 6.1, failure was with snapcraft 7.0.7
[22:21] <cjwatson> j'accuse
[22:30] <sarnold> :)
[22:31] <ted> cjwatson, thanks, definitely looks like the cause.
[22:31] <ted> Sorry for the delay conference Wifi :-)
[22:44] <nomadz> hi, i'm taking a look at the launchpad installation instructions on dev.launchpad.net, it is still true everything above 18.04 is not supported/working as the instructions states?
[22:44] <nomadz> or is just outdated?
[22:44] <nomadz> https://launchpad.readthedocs.io/en/latest/how-to/running.html
[23:04] <ted> K, just confirming. A build with the snapcraft channel set to "6.x/stable" works.