19:29:50 <sumpfralle> #startmeeting
19:29:50 <MeetBot> Meeting started Wed Dec  5 19:29:50 2018 UTC.  The chair is sumpfralle. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:29:50 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:29:57 <sumpfralle> Welcome to the weekly IRC meeting
19:30:03 <sumpfralle> who is around?
19:31:58 * sumpfralle could imagine that this will be a short meeting ...
19:32:20 * h01ger waves
19:32:27 <h01ger> i'm not here ;)
19:32:40 <h01ger> glad that 2.0.43 can migrate to buster now
19:32:49 <h01ger> looking forward to do more uploads
19:33:14 <h01ger> now i'm really not here anymore ;) am a bit sick and need to do other stuff :/
19:34:14 * sumpfralle waves back and is also happy about the buster migration
19:35:36 * sumpfralle is away from keyboard for five minute and is curious if someone joins meanwhile ...
19:42:52 <TheSnide> ok
20:01:24 <sumpfralle> sorry - being back late
20:01:33 <sumpfralle> TheSnide: are you still in the mood for an exchange?
20:01:41 <TheSnide> sure
20:02:12 <sumpfralle> So: did anything happen during your last week in munin-land?
20:03:16 <TheSnide> not really. I discovered the sorry state of dmm0
20:03:43 <TheSnide> ... And the package 2.999.9-x doesn't really work with PG
20:03:44 <sumpfralle> Do you have an idea, why there were gaps in the recorded data?
20:03:52 <sumpfralle> it does not?
20:04:31 <sumpfralle> (I added some (upstreamed) patches to the Debian pacakge, that helped a bit with postgresql, I think)
20:04:31 <TheSnide> nope, it does not
20:04:41 <sumpfralle> ok
20:04:49 <sumpfralle> Anyway: time for 2.999.10 :)
20:04:57 <TheSnide> i had basically to cp -R the *.pm from master to /usr...
20:05:06 <TheSnide> which magically made it work ;
20:05:08 <TheSnide> which magically made it work ;)
20:05:15 <TheSnide> it seems so.
20:05:29 <TheSnide> Which will be released hopefully _today_
20:05:37 <h01ger> hear hear
20:05:39 <h01ger> ;p
20:05:40 <sumpfralle> Are you sure? I am quite confident, that I created autopkgtests, that tested, whether munin-httpd delivers graphs via http
20:05:54 <sumpfralle> today/soon - great!
20:05:55 <TheSnide> yup, ootb it works with sqlite
20:06:01 <sumpfralle> ah - ok
20:06:14 <sumpfralle> so I should add autopkgtests for postgresql :)
20:06:26 <TheSnide> but for some reason the sqlite has tremendous overhead
20:07:11 <sumpfralle> Do you have an idea whether it is reading or writing?
20:09:31 <TheSnide> i suppose it's done when it reopen the DB
20:09:40 <TheSnide> as it does for each request.
20:09:51 <TheSnide> as, SQLite has no state. By design.
20:11:05 <sumpfralle> Did you already dive into profiling?
20:13:42 <TheSnide> i didn't really care, as with PG is flies
20:13:52 <sumpfralle> ok :)
20:13:52 <TheSnide> it flies*
20:14:55 <TheSnide> sqlite is only designed for small sites. So, we might improve things there, but it's not high prio
20:15:04 <sumpfralle> good
20:15:05 <TheSnide> I prefer to focus on features
20:15:24 <sumpfralle> an excellent choice!
20:22:50 <sumpfralle> I guess we should close the meeting and just continue talking?
20:23:11 <TheSnide> yup
20:23:18 <TheSnide> #endmeeting
20:23:58 <TheSnide> hmm... meeting wasn't started ;)
20:28:38 <sumpfralle> #endmeeting