[13:16:18] <inflatador>	 <o/
[13:57:02] <pfischer>	 o/
[14:01:51] <pfischer>	 inflatador: Are you around? Do you want to use the pairing slot for going over the tickets? I saw your question from yesterday.
[14:02:47] <inflatador>	 pfischer sure, apologies in advance if it's a repeat from last week 
[14:03:06] <inflatador>	 https://meet.google.com/aod-fbxz-joy
[15:01:05] <gehel>	 Office Hours starting, we have a few guests: https://meet.google.com/vgj-bbeb-uyi
[15:01:55] <inflatador>	 we had our first data-reload failure ;( Starting over, but luckily we have 2 more still running without error
[16:41:46] <inflatador>	 dr0ptp4kt still looking for that rdf repo for you. This one makes the docker img https://gitlab.wikimedia.org/repos/search-platform/flink-rdf-streaming-updater
[16:42:00] <inflatador>	 also, back in ~40
[17:23:57] <inflatador>	 back
[17:50:14] <ryankemper>	 inflatador: which rdf repo specifically?
[17:51:55] <inflatador>	 ryankemper was trying to find the one where the rdf-streaming-updater-producer java code lives. Still haven't found it, if you know hit us up
[17:58:12] <inflatador>	 it's probably traceable from the docker img repo, back to archiva/jenkins/gitlab CI or whatever builds the artifact
[17:58:44] <ryankemper>	 inflatador: Think it's just `wikidata/query/rdf`
[17:59:08] <ryankemper>	 https://gerrit.wikimedia.org/r/plugins/gitiles/wikidata/query/rdf/+/master/streaming-updater-producer
[17:59:45] <inflatador>	 ryankemper thanks, that looks right indeed
[17:59:53] <inflatador>	 dr0ptp4kt ^^ 
[18:01:43] <inflatador>	 lunch, back in ~40
[18:02:55] <ebernhardson>	 a very odd thing happens when i induce excessive slave lag,  i set my slave to have 10 minutes of lag so i could step through code and ensure the slave would still be lagged. In this state i can edit a page into a redirect, get a success, and then edit the page into a plain wiki page, get a success but the rev_id doesn't change and no events are produced
[18:03:07] <ebernhardson>	 editing pages via api
[18:03:27] <ebernhardson>	 essentially it looks like mediawiki is rejecting the edit, but the response says success
[18:48:51] <inflatador>	 back
[18:49:39] <inflatador>	 that's confusing
[18:57:49] <dr0ptp4kt>	 thanks inflatador for the repo path!
[18:58:43] <dr0ptp4kt>	 paths plural, that is :)
[18:59:16] <dr0ptp4kt>	 and ryankemper thx!
[19:46:32] <ebernhardson>	 The more i look at these events the more that is wrong :(  Page move events serialize the old page title
[19:46:42] <ebernhardson>	 but not the new one
[19:46:51] <ebernhardson>	 we can recover by loading by pageid, but awkward
[22:24:03] <ryankemper>	 i.nflatador and I just shipped https://gerrit.wikimedia.org/r/c/operations/cookbooks/+/968346/ so the `graph_split` data reload cookbook changes should be all squared away now