2022-08-31T03:28:51  *** Senhara[m] is now known as Senhara7762[m]
2022-08-31T08:52:31  *** teepee_ is now known as teepee
2022-08-31T12:25:44  <bmwiedemann> I'm planning to change the CPU emulation of most VMs tomorrow and that needs a short shutdown of VMs. shoud be <1m for each of them
2022-08-31T14:17:18  <lkocman[m]> Jacob Michalskie: would you please have a look at https://github.com/openSUSE/get-o-o/pull/117 ?
2022-08-31T14:17:26  <lkocman[m]> I could use a review
2022-08-31T14:17:42  <SasiOlin[m]> I was just going to comment
2022-08-31T14:18:15  <lkocman[m]> ah thank you!
2022-08-31T14:19:45  <lkocman[m]> I'm wondering if it wasn't related to some cache clearing again. But in general after adding few prints I received current => nil
2022-08-31T14:20:14  <lkocman[m]> not sure what was wrong, could have been some accidental string somewhere that was fixed or so
2022-08-31T14:20:15  <lkocman[m]> not sure
2022-08-31T15:30:34  <bmwiedemann[m]> so on download.o.o we are around 2500-3000 MBit/s and a 2 second capture showed most is rsync traffic.... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/73a8d71723b10a422d53917b22879126ae6b8aac)
2022-08-31T15:35:31  <bmwiedemann[m]> about 10 slower and 60 faster mirrors pull at the same time
2022-08-31T16:37:18  <pjessen> bmwiedemann[m]: yup, its repopush
2022-08-31T16:38:36  <pjessen> I'm trying to optimize repopush, it's way behind. once it has caught up, there ought to be less traffic, but it depends.
2022-08-31T16:57:30  <pjessen> rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrttttttyyyyy
2022-08-31T16:58:24  <bmwiedemann[m]> I think, we need to let more mirrors sync from rsync.o.o instead of stage
2022-08-31T17:00:31  <pjessen> sorry, cat on keyboard
2022-08-31T17:00:32  <pjessen> rsync.o.o is not up to date
2022-08-31T17:01:49  <bmwiedemann[m]> cat > /dev/keyboard :-)
2022-08-31T17:02:16  <pjessen> we have a new kitten, 15 weeks old
2022-08-31T17:02:44  <pjessen> bmwiedemann[m]: hahaha, sorry, I was a bit slow :-)
2022-08-31T17:03:33  <bmwiedemann[m]> Do these kitten keep you awake at night? That would explain the slowness...
2022-08-31T17:06:47  <pjessen> I think it's old age, I'm past 30
2022-08-31T17:11:20  <pjessen> wrt repopush, we need to push about 1Tb per day per mirror.  We have 6 active push mirrors.
2022-08-31T17:37:45  <ctlinux[m]> <pjessen> "sorry, cat on keyboard" <- You have the next-gen Keyboard Cat
2022-08-31T17:38:21  <cboltz> pjessen: FYI: mailman is terribly slow again (gateway timeout), I'm just restarting it
2022-08-31T17:38:25  <SFaulken[m]> I can't let my cat on the keyboard anymore, he's old, and fat, and I'm afraid he will just break it.
2022-08-31T17:38:57  <pedja> as excuses to get new mech keyboard go, that one is not that bad
2022-08-31T17:39:40  <SFaulken[m]> I don't want a new keyboard, this one is just getting broken in.
2022-08-31T17:41:34  <acidsys> broken in.. by the cat
2022-08-31T17:42:31  <SFaulken[m]> heh, nah, just by my comically heavy typing style.
2022-08-31T18:55:59  <pjessen> cboltz: I restarted it earlier too :-(
2022-08-31T18:57:38  <cboltz> great :-/
2022-08-31T18:58:18  <cboltz> just wondering - is there a log that would help to explain what causes the 100% cpu usage?
2022-08-31T18:58:41  <cboltz> I've seen the uwsgi log, but it's basically an access log
2022-08-31T18:59:21  <pjessen> I've tried stracing, but without understanding the code, it's little use
2022-08-31T19:00:26  <cboltz> that, and debugging with strace isn't fun in general (too much noise)
2022-08-31T19:01:47  <pjessen> I try to get a feel for whats going on, I use -etrace=, but without some idea of what is supposed to happen .....
2022-08-31T19:40:10  <SasiOlin[m]> <lkocman[m]> "I'm wondering if it wasn't..." <- did you do it while a server was running? you need to perform a full restart of jekyll serve for any changes in plugins to take on
2022-08-31T19:41:35  <SasiOlin[m]> I should just quickly replace uwsgi with gunicorn
2022-08-31T23:17:41  *** teepee_ is now known as teepee