[09:50:28] <volans>	 test log
[09:50:29] <apergos>	 nice!
[09:50:43] <apergos>	 guess that should go in the topic
[09:50:56] <volans>	 gehel: if you want to can add https://wm-bot.wmflabs.org/libera_logs/%23wikimedia-search/ to the topic
[09:51:07] <volans>	 for now is just plain text AFAIK, the web UI needs to be updated
[09:51:21] <volans>	 to get the new network, the one at https://wm-bot.wmflabs.org/browser/
[09:55:21] * gehel has never changed an IRC topic ;)
[09:55:34] <gehel>	 I'm sure I can figure it out after lunch
[09:55:43] <gehel>	 Good exercise !
[09:55:52] <apergos>	 heh
[09:56:07] <dcausse>	 lunch
[09:56:19] <apergos>	  you will type  /topic         and then copy the result to clipboard... then   type it again with the new topic after it in the same line
[09:56:21] <apergos>	 easy peasy!
[09:56:28] <apergos>	 enjoy your lunch!
[10:13:21] <volans>	 gehel: as for wikibugs, which tags?
[10:23:45] <zpapierski>	 dcausse: I'm ready when you are - at least until 2:30PM, also 40min after that
[10:23:51] <gehel>	 volans: I might have mixed up the bots. Isn't wikibugs the one that pastes the link to phab when you post a task number?
[10:24:18] <gehel>	 pasting the phab activity to this channel is mostly noise in our case
[10:26:25] <volans>	 I think it does both things
[10:28:27] <gehel>	 ok, then an empty list of tag is probably what we want
[10:28:32] <gehel>	 where is this configured?
[10:29:05] <zpapierski>	 freenode reopened our channel there - "This channel has been reopened with respect to the communities and new users. The topic is in violation of freenode policy: https://freenode.net/policies"
[10:29:42] <gehel>	 looks like I'm not an operator, so I can't set the topic :/
[10:30:47] <volans>	 gehel: ack for the empty tags, I still need to figure out something there
[10:31:12] * gehel still not entirely back from lunch...
[10:32:45] <volans>	 did you had also !log in here?
[10:34:00] <apergos>	 gehel you are a founder, you should be able to get chanops here
[10:34:33] <zpapierski>	 ebernhardson is the founder
[10:34:35] <apergos>	 zpapierski: the channels over on freenode have been redirected to the same channel name with ##  
[10:34:57] <zpapierski>	 yeah, I saw that - what's up with that?
[10:35:00] <apergos>	 (12:37:03 μμ) ChanServ: (notice) Information on #wikimedia-search:
[10:35:00] <apergos>	 (12:37:03 μμ) ChanServ: (notice) Founder    : bd808, ebernhardson, gehel
[10:35:07] <zpapierski>	 a, my bad
[10:35:18] <zpapierski>	 didn't know that there can be more founders
[10:35:33] <apergos>	 what' up is that the freenode ops/owner decided to co-opt all channels with 'libera' in the topic
[10:35:39] <apergos>	 their way of combatting the move I guess
[10:35:55] <apergos>	 happened last night.
[10:37:22] <volans>	 hello stashbot 
[10:37:24] <volans>	 T2001
[10:37:24] <zpapierski>	 heh, rather desperate attempt
[10:37:25] <stashbot>	 T2001: [DO NOT USE] Documentation is out of date, incomplete (tracking) [superseded by #Documentation] - https://phabricator.wikimedia.org/T2001
[10:37:28] <volans>	 gehel: ^^^
[10:37:50] <volans>	 and no wikibugs for now if I understand it correctly
[10:57:36] <gehel>	 volans: thanks !
[11:14:58] <apergos>	 awesome!
[11:18:22] <zpapierski>	 meal 3 break
[11:56:15] <gehel>	 Errand, back in 20'
[11:57:21] <dcausse>	 zpapierski: back, I'm available when you want
[12:53:10] <zpapierski>	 dcausse: have a meeting now - 4pm? 
[12:53:24] <dcausse>	 zpapierski: sure
[14:09:32] <zpapierski>	 dcausse: sorry for being late -meet.google.com/mfu-yfwh-rgz
[14:12:04] <dcausse>	 np
[14:21:40] <apergos>	 back to my usual haunts, see you in the other channels!
[14:41:15] <dcausse>	 errand
[15:11:43] <gehel>	 ebernhardson / volans: about T283230, there is still an "ACL" and  "mode" columns. What should we check about those?
[15:11:44] <stashbot>	 T283230: Move SRE-related IRC channels to Libera - https://phabricator.wikimedia.org/T283230
[15:12:29] <volans>	 gehel: I can help with knowledge but as I'm not op here I can do it for you
[15:12:47] <gehel>	 volans: I am, but I have no idea what to do :)
[15:12:58] <volans>	 ok, pvt
[15:13:01] * gehel also has to pick the kids in 15'
[15:17:57] <gehel>	 ryankemper: I talked to Faidon about T280382. Let's see how much it woudl take to solve the problem at the hardware level (adding more SSD so that we stay on RAID10)
[15:17:58] <stashbot>	 T280382: WDQS hosts low on /srv disk space - https://phabricator.wikimedia.org/T280382
[15:18:34] <gehel>	 I wanted to start on it, but time flew to fast today.
[15:18:54] <ryankemper>	 gehel: how much $$$-wise, primarily? or just in general (time etc)
[15:19:41] <gehel>	 mostly the cost, or at this point : what hardware we need (we can estimate the cost next)
[15:20:52] <gehel>	 a few scenarios: 1) add disks to the problematic servers (with mixed size), 2) add disks to all the older servers so that we move them all back to RAID10 3) replace existing disk in older servers so we have all the same disks on RAID10
[15:21:31] <gehel>	 I've also checked with joanna to see if we can get some more time from John to fix it within our current H/W limitations
[15:21:52] <gehel>	 I'll work on the various scenarios tomorrow, unless you have time today to get started on it
[15:22:10] <ryankemper>	 great
[15:22:23] <ryankemper>	 yeah I'll start getting the scenarios mapped out
[15:22:58] <ryankemper>	 so I assume scenario (3) we'd be replacing with disks twice as big? because otherwise we'd still need raid0 we just wouldn't run into the layout issue
[15:23:25] <gehel>	 yep, for all those scenarios, we want to migrate to RAID10
[15:23:48] <gehel>	 I would start by an inventory spreadsheet, with the current number of disk / sizes for each servers
[15:24:32] <ryankemper>	 ack
[17:50:56] <dcausse>	 dinner
[17:57:17] <tgr>	 ebernhardson: hi! we have an Add Links bug (on our side) that caused probably 1-2K dangling weighted tag search index entries per wiki (we have 4 pilot wikis). We'd like to fix those with a script, ie. iterate through them and call CirrusSearch::resetWeightedTags(). Do you think that could be a problem?
[17:57:42] <ebernhardson>	 tgr: shouldn't be a problem, a few thousand indexing requests only takes a few minutes
[17:57:57] <tgr>	 ack, thanks!
[18:46:26] <gehel>	 ryankemper: did you get started on a spreadsheet yet?
[18:48:53] <ryankemper>	 gehel: no, went to run an errand post-unmeeting and about to srive back now so I’ll have the first iteration in about an hour if you’re still around
[18:49:12] <gehel>	 I'll start a spreadsheet, just to show what I had in mind
[18:52:23] <ryankemper>	 Sounds great
[18:53:24] <ebernhardson>	 hmm, the wikimedia-discovery room in leenode has been co-opted, asked on #freenode-services if i can get it re-closed but not expecting much :)
[19:28:41] * ebernhardson is going to assume after ~30 minutes with no reply, but the ops talking to someone else, that they aren't interested in even investigating possible solutions
[19:52:22] * ebernhardson loves finding a directory named `${env.XDG_CACHE_HOME}`.  Wonder where i put the wrong thing :)
[20:51:39] <ebernhardson>	 wow, maven has been downloading for more than 20 minutes now .... and i thought it was slow when CI ran it....
[21:02:02] <ebernhardson>	 maven claims it was only 15 minutes...feels longer :P