IRC Logs for #crux-devel Tuesday, 2017-07-25

*** heroux has quit IRC00:22
*** chinarulezzz has quit IRC00:45
*** chinarulezzz has joined #crux-devel01:22
*** chinarulezzz has quit IRC01:37
*** heroux has joined #crux-devel01:39
*** heroux has quit IRC01:54
*** _________mavric6 has quit IRC02:30
*** heroux has joined #crux-devel02:31
*** _________mavric6 has joined #crux-devel02:31
*** heroux has quit IRC02:35
*** heroux has joined #crux-devel03:41
*** heroux has quit IRC03:46
*** heroux has joined #crux-devel03:57
*** heroux has quit IRC04:06
*** heroux has joined #crux-devel04:17
*** heroux has quit IRC04:36
*** heroux has joined #crux-devel05:12
*** heroux has quit IRC05:33
*** chinarulezzz has joined #crux-devel06:58
*** heroux has joined #crux-devel09:40
*** nthwyatt has joined #crux-devel09:48
*** darfo- has joined #crux-devel09:51
*** nthwyatt has quit IRC09:55
teK_got a new mirror request11:59
teK_jaeger and frinnst: I'm on vacation starting today (=tomorrow). Should we have a chat abouto open issues with our new ride^Wserver?12:00
pedjajaeger, you might want to push nvidia update soon-ish :)
teK_pkgmk does not create a signature if source is empty12:44
teK_this is a bug12:48
teK_fixed in git since February.   :-)12:58
frinnstfeel free to roll a new release. I can push it later today13:27
frinnstI think most stuff if fixed. jaeger fixed the hooks yesterday13:39
frinnsti think the biggest remaining thing is gitolite?13:39
*** chinarulezzz has quit IRC13:45
jaegerpedja: planning to test the new version today13:50
teK_frinnst: gitolite should be easily doable :)13:57
teK_pkgmk was fixed already. But not my version in ~/bin .. ;)13:57
teK_most important question will be: who still needs shell access13:58
teK_my guess: frinnst jue jaeger teK_ and maybe sepen13:58
pedjajaeger, since LTS and the 'regular' version are the same (384.x), will you bump both nvidia and nvidia-sl to 384.x, or ?13:59
jaeger still shows them at different versions14:06
pedjadamn it14:10
jaegerIs "official release" new terminology for "short lived branch"?14:13
jaegerWould be nice if all of that shit were consistent14:14
jaegerstill, that last edit reads to me that he meant to say it was long-lived, not official14:14
pedjait's both :)14:15
pedjait would make sense that 375.82 is LTS and 384.x short-lived a.k.a 'official'14:17
jaegerThey aren't consistent with higher version numbers being short lived14:18
pedjalooks like a last minute change, or smth14:19
pedjaI use the latest ine, whichever branch it is, anyway :)14:20
frinnstteK_: currently everyone has shell access i think14:56
jaegeryeah, but wouldn't be necessary with gitolite15:11
jaegerfrinnst: what about cruxbot, did you pick a new piece of software to use?15:29
jaegerLooks like another permissions issue was breaking the opt ML hook, I've fixed that up as well15:41
teK_frinnst: only some should have ;)17:54
teK_I would be up for checking if this can easily be done in python.17:54
teK_also: did it break because of the new install?17:55
frinnstyeah the old cruxbot depended on old gems that were deprecated a looong time ago18:13
frinnstjaeger: havent looked at it yet but there seems to be plenty of choice out there. the current bot is a bit too simple for my liking18:13
*** chinarulezzz has joined #crux-devel19:04
*** onodera has joined #crux-devel19:42
*** onodera has quit IRC19:46
*** onodera has joined #crux-devel19:50
*** frinnst has quit IRC21:13
*** heroux has quit IRC22:07
*** heroux has joined #crux-devel22:12
*** onodera has quit IRC22:13
*** heroux has quit IRC22:35
pedjateK_, git is at 2.13.3. quite a few bug fixes :)22:43
*** heroux has joined #crux-devel23:12
*** Workster has quit IRC23:20
*** twunmuxsstwikpsj has joined #crux-devel23:20
*** Workster has joined #crux-devel23:22
*** chinarulezzz has quit IRC23:29
*** Workster has quit IRC23:30
*** frinnst has joined #crux-devel23:54
*** frinnst has quit IRC23:54
*** frinnst has joined #crux-devel23:54
jaegerfrinnst: that fifo one seemed kinda cool, though we might want to implement some kind of queueing for it23:57
teK_I drafted something reading from a fifo with knowing about that approach, too :o23:58
jaegerMaybe rather than using a fifo, stick the output into a redis queue or similar with a bot that spews them into IRC with rate limiting23:58
teK_fifo can be bitchy wrt timing23:58
teK_or open for reading/writing semantics23:59
jaegerand doesn't have queueing in case of a failure, I imagine23:59
jaegerer, buffering, I mean23:59
teK_as long as you dont delete it?23:59
teK_why not, that's the point of a FIFO I guess23:59
jaegerNot sure, I haven't used them much23:59
teK_minus size limits, but that should not be a problem23:59

Generated by 2.14.0 by Marius Gedminas - find it at!