jaeger changed the topic of #crux-devel to: CRUX (https://crux.nu/) development channel | Logs: https://libera.irclog.whitequark.org/crux-devel/
SiFuh has quit [Remote host closed the connection]
SiFuh has joined #crux-devel
_whitelogger has joined #crux-devel
braewoods_ has quit [Remote host closed the connection]
braewoods has joined #crux-devel
jue has joined #crux-devel
jue has joined #crux-devel
jue has quit [Ping timeout: 250 seconds]
<jaeger> Just realized we're still running 3.6 on crux.nu... we should really update it sometime :)
<beerman> aye
<beerman> whenever we do it, we might want to work together on that to make the change to nginx/gitea maybe
<beerman> would be nice to have that going productive
<jaeger> I'd suggest maybe kernel first, then crux update and nginx... probably still more to do with gitea, right? Like it only has contrib hooks right now if I remember right
groovy2shoes has quit [Ping timeout: 246 seconds]
<jaeger> hrmm... and the hooks are broken due to logfile permissions, I wonder when that happened
<jaeger> Ah, not permissions, the logfiles are missing now
<beerman> definately, sounds like a good roadmap. switching to gitea shouldn't be much of a problem, i'd worry more about the mailing list and stuff
<beerman> pmwiki and stuff are easily switched, i might want to touch these configs once more for final deployment and move a few files but that should be it
<jaeger> I'll get an updated kernel built soon and reboot into that first
<beerman> then again, maybe a complete redo would be fine?
groovy2shoes has joined #crux-devel
<jaeger> ok
<jaeger> We still have no remote console access but if it goes wrong for some reason I can ask Stx for help
<beerman> i am thinking about the partition sizes
<farkuhar> speaking of mailing list, did stenur's message about /root/.config/git/attributes "Permission denied" ever get resolved? It sounds like the result of a git daemon running with incorrect environment variables.
<jaeger> I looked at it briefly but hadn't had time to figure out a solution yet. It seems to be caused by start-stop-daemon switching to the git user but preserving some of root's environment
<jaeger> start-stop-daemon doesn't seem to have an option to pass environment or change to the proper user's env but like I said, only looked briefly
<jaeger> If you guys have a suggestion for a proper way to fix it feel free to suggest it... I was thinking for the short term something like a script that uses su or sudo with whatever option does the right thing regarding environment
<jaeger> wrapper script, I mean
<farkuhar> I find it amusing that stenur is complaining about the side effects of a tool (start-stop-daemon) to which he's made the majority of the commits.
<farkuhar> But anyway, the recommended flag for sudo is -H (or --set-home). Meanwhile the manpage for doas implies that the target user's environment is always imported, no extra flag needed.
<beerman> I haven't looked into that, sorry
<beerman> well, just an idea, might be as good of a chance as any, really
<beerman> no idea if exchanging the server is a viable option, really
<beerman> so what other time will we do it? but its not critical, it should be a lot better than it was anyway
<jaeger> exchanging the server? other time? I feel like I missed part of a conversation
<beerman> sorry, I mean when else would we try to rebuild it?
<jaeger> I had no specific plan in mind
<beerman> at best, I feel, if we were to exchange the hardware. which seems unlikely
<jaeger> related to partition sizes?
<beerman> thats my motivation, yeah
<jaeger> Ah... doing a full rebuild could definitely be tough without console access
<beerman> we certainly would need ask stx for local help on that one without it
<jaeger> Yeah
<jaeger> Could do it in some kind of live env, I'm sure
<beerman> basics steps can all be prepared so it should be a quick one to set us up
<beerman> just needs a bit of preperation but after that we could take over with it
<beerman> i feel like this the local part is the kind of work that i'd be open to spend a crate of beer or equivalent on to have done :D
<jaeger> heh
<jaeger> I wonder if Stx would be willing to hook up the IPMI device finally
<beerman> I write gibberish but I'm just tired. but you get the point :D
<beerman> heh :D well, that would be nice
<jaeger> I just noticed your "complete redo" comment, now the rest makes more sense
<jaeger> I'm a bit tired as well
<beerman> yeah, starting from scratch-ish always has good things to it - besides being more work than it should be