19:28:54 #startmeeting 19:28:54 Meeting started Wed Jan 16 19:28:54 2019 UTC. The chair is sumpfralle. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:28:54 Useful Commands: #action #agreed #help #info #idea #link #topic. 19:28:59 hi 19:29:11 #chair TheSnide h01ger chteuchteu be0rn kenyon bcg 19:29:11 Current chairs: TheSnide bcg be0rn chteuchteu h01ger kenyon sumpfralle 19:29:26 #topic New things from the last week? 19:29:51 I do not have anything specific to share. Just minor things. 19:29:54 How about you? 19:30:33 i spent some time on dmm0, and I dont understand the reasons for those white bands 19:30:51 #topic Gaps on demo.mm0 19:30:53 that's the most worrysome thing to me currently 19:31:01 agrred 19:31:06 do you have a theory? 19:31:22 i had several, which all proved wrong. 19:31:29 progress! :) 19:31:41 edison style. 19:32:17 now, i am looking at the db update warnings. 19:32:59 as my last theory is that if a db stmt fails, it just breaks and nothing gets updated 19:34:29 i'm currently at one affecting the table url, which i dont understand currently. but i didnt spend much time on it yet. 19:35:20 Can we somehow log/track database statements? (maybe on the database side - not in munin) 19:35:36 Or is that too low level? 19:35:39 hint: it is a unique key violation, but i do a DELETE right before the INSERT. and it worka in the cli client 19:36:04 delete + insert sounds like "update" :) 19:36:05 we can. but i'm not there yet. 19:36:23 it is more an UPSERT 19:36:27 but yea. 19:36:54 ha - I have never seen that - nice! 19:36:58 so, that is my last theory. 19:37:26 latest, and last. 19:37:34 Would it help you to explain all your research to me? (like: talking to someone may enlighten you) 19:38:17 well, if that one doeant work out, ill revert to that. indeed 19:39:14 So I will leave you hunt it down - and if it fails I will be happy to listen to your complaints about the absurdity of this mystery. 19:40:22 It sounds like we can finish our meeting? :) 19:55:45 #endmeeting