[00:04:04] <georgebarnick>	 Is the source code for the new Wikipedia app for Android public?
[00:06:03] <Reedy>	 Yup
[00:06:05] <MatmaRex>	 georgebarnick: probably, if you manage to find it
[00:06:25] <Reedy>	 https://github.com/wikimedia/apps-android-wikipedia
[00:08:42] <georgebarnick>	 aha thanks Reedy
[00:11:43] <georgebarnick>	 lol why does the readme give instructions for mac but not linux or windows
[01:23:11] <sDrewth>	 can someone please explain to me how someone's recenteditcount can be higher than a person's edit count (api pull)
[01:23:29] <sDrewth>	 and I am talking signicantly higher
[01:24:27] <Krenair>	 recenteditcount = just by looking through recent changes and looking at all entries?
[01:26:00] <Krenair>	 sDrewth, pretty sure recent changes includes log entries.
[01:31:20] <sDrewth>	 okay, thanks
[01:33:46] <sDrewth>	 Krenair: not so sure, and something like    editcount="1" recenteditcount="121"  is weird
[01:35:15] <Krenair>	 wiki + user?
[01:35:57] <sDrewth>	 numbers of them, https://tt.wikipedia.org/w/api.php?action=query&list=allusers&auactiveusers&auwitheditsonly&aulimit=500&auprop=editcount
[06:44:44] <Nemo_bis>	 An edit to the browser compatibility page is OOM'ing my browser it seems
[06:45:41] <Nemo_bis>	 (Or perhaps it was an edit to GettingStarted.) How on earth is this possible, grrrrr
[06:52:29] <Nemo_bis>	 LOL 3 GB virtual memory, time to kill
[13:06:39] <Betacommand>	 Nemo_bis: leak in your browser?
[13:08:02] <Nemo_bis>	 Betacommand: unless you can think of something outside the browser that can raise the browser's memory consumption, I assume yes. :)
[13:23:08] <Betacommand>	 Nemo_bis: sometimes broken web pages can cause that
[16:39:21] <sDrewth>	 thx anomie, I even understood the summary ;-)
[16:39:56] <anomie>	 sDrewth: You're welcome, although I'm lacking the context to know for what ;)
[16:40:15] <sDrewth>	 api fix
[16:41:37] <Nemo_bis>	 yes, thanks for https://gerrit.wikimedia.org/r/#/c/130093/2 :)
[16:45:07] <Nemo_bis>	 lovely spam http://code.google.com/p/gerrit/issues/detail?id=2458
[17:20:19] <chrismcmahon>	 hi csteipp do you have a minute? I want to ask about a theory I have about why we might be losing sessions a la https://bugzilla.wikimedia.org/show_bug.cgi?id=65709
[17:20:37] <csteipp>	 chrismcmahon: Yeah-ish
[17:21:44] <chrismcmahon>	 csteipp: OK. Is it possible that a session could time out or get lost somewhere in a caching layer? I was remembering this problem: https://gerrit.wikimedia.org/r/#/c/103376/
[17:23:46] <chrismcmahon>	 csteipp: in the case of that patch, we were seeing 503s for a page, but I wonder if some cache timeout could actually log out a user. 
[17:24:32] <csteipp>	 chrismcmahon: That particular bug, probably not. If I understand faidon's comment, that was the timeout between varnish<->apache, and those should always be synchronized.
[17:24:54] <csteipp>	 chrismcmahon: But caching in general seems like a likely place for the issue
[17:26:05] <csteipp>	 I wouldn't doubt if we were caching a reply for a user, and when the next request comes at the right time, we get the cached version of a token instead of the current edit token. But I haven't been able to find any  places where it looks like that is happening.
[17:27:11] <chrismcmahon>	 csteipp: yeah, not that particular bug, but looking at the other nearby values 'connect_timeout' => "5s" and 'between_bytes_timeout' => "4s" make me wonder. We sometimes see the unexpected logout in places like Flow, but we see in VE a LOT
[17:28:13] <csteipp>	 If vanish timed out, you would either see the connection dropped or get a 500 error
[17:28:58] <csteipp>	 At least, that's what I would expect-- someone from Ops would know better than me
[18:51:08] <peteforsyth>	 Anybody know if there's a problem with WMFlabs right now?
[18:51:25] <peteforsyth>	 I tried to use tools.wmflabs.org/croptool but got the following error message:
[18:51:29] <peteforsyth>	 Warning: Unknown: write failed: No space left on device (28) in Unknown  on line 0  Warning: Unknown: Failed to write session data (files). Please verify  that the current setting of session.save_path is correct (/var/lib/php5)  in Unknown on line 0 
[18:51:54] <hoo>	 peteforsyth: Better ask in #wikimedia-labs this channel is about the production sites
[18:51:55] <Krenair>	 labs is fine as far as I can tell...
[18:52:05] <Krenair>	 I heard the tools project was having issues though
[18:53:45] <peteforsyth>	 thanks hoo and Krenair
[20:55:06] <abartov>	 hi.  Is there a tool that answers the following question? -- of files f1...fN in category X, which file has been used (i.e. incorporated) in the greatest number of wiki pages across the different projects?
[20:56:01] <abartov>	 (in other words, a MAX() function on the files' "Special:GlobalUsage" count.
[20:56:35] <twkozlowski>	 Yes.
[20:56:51] <twkozlowski>	 Though that's an interesting way to describe the function you're after.
[20:57:25] <abartov>	 twkozlowski, oh?
[20:57:31] <twkozlowski>	 http://tools.wmflabs.org/glamtools/glamorous.php ?
[20:58:13] <abartov>	 Isn't glamorous only giving me the total per category?
[20:58:26] <twkozlowski>	 No, it can give you per-file numbers as well
[21:00:33] <twkozlowski>	 http://tools.wmflabs.org/glamtools/glamorous.php?doit=1&category=Barack+Obama+in+2014&use_globalusage=1&ns0=1&show_details=1&projects[wikipedia]=1
[21:03:11] <abartov>	 ah, the "details" checkbox!  Missed it.  Thanks, twkozlowski!
[21:03:37] <twkozlowski>	 Sure :-)
[21:27:19] <jackmcbarn>	 Nemo_bis: it did run successfully the last time that it was scheduled to (which was april)
[21:27:44] <Nemo_bis>	 jackmcbarn: great!
[21:27:49] <jackmcbarn>	 i'm asking for a rerun because the query that that special page used was changed, making the results completely wrong now
[21:28:21] <Nemo_bis>	 You can submit a patch to make the update monthly for all special pages even on en.wiki.
[21:29:18] <Nemo_bis>	 Just add springle to cc. He said it's not a problem nowadays with the slaves and the query killer, it should be easy to get the patch merged. (At worst it gets reverted after some days/weeks. ;) )
[21:29:57] <Nemo_bis>	 Asking one-time runs doesn't scale, I don't like it.
[21:30:31] <Nemo_bis>	 But it's not a big deal, anyone with shell access can do it if you poke around.
[21:32:13] <Nemo_bis>	 Speaking of which, I was wondering if we should run updateArticleCount.php regularly (monthly? yearly?) on all wikis.
[21:32:37] <Reedy>	 What's the special page name?
[21:33:21] <Nemo_bis>	 Reedy: MostTranscludedPages
[21:33:29] <Nemo_bis>	 Watch the case ;)
[21:34:13] <Nemo_bis>	 IAlex made updateArticleCount fast many months ago and the bugs/questions/doubts/FUD of those who see differences between special:statistics and WikiStats are so annoying...
[21:34:26] <Reedy>	 The public one is useless
[21:34:56] <Reedy>	 * Special:MostLinkedTemplates has been renamed to Special:MostTranscludedPages.
[21:35:27] <Nemo_bis>	 Does the script work still?
[21:35:43] <Nemo_bis>	 On de.wiki it was run two days ago successfully AFAICS
[21:35:46] <Reedy>	 mwscript updateSpecialPages.php --wiki=enwiki --only=Mostlinkedtemplates --override
[21:45:51] <greg-g>	 twkozlowski: https://wikitech.wikimedia.org/w/index.php?title=Deployments&diff=117999&oldid=117958
[21:46:11] <greg-g>	 twkozlowski: due to bugs found over the weekend
[21:48:55] <twkozlowski>	 greg-g: Monday evening UTC is way too late for me to do anything :-)
[21:49:52] <twkozlowski>	 I'm not even surprised. Got used to this. Meh.
[21:52:32] <twkozlowski>	 greg-g: You informed the stewards yet?
[21:53:04] <Nemo_bis>	 Perhaps we should announce more things in the past and less which are still in the future. :)
[21:53:57] <twkozlowski>	 Yes, but then the point of alerting people to things that might change their experience would be, well, quite lost.
[21:56:41] <greg-g>	 twkozlowski: I was just the messenger here, ask Deskana ;)
[21:57:01] <twkozlowski>	 I asked at #wikimedia-stewards
[21:57:43] <greg-g>	 Deskana's on point for that stuff
[21:57:45] <twkozlowski>	 Deskana... is he the guy who was supposed to convert Oversight log into suppression log a few months ago?
[21:57:52] <greg-g>	 they arne't clamoring, apparently, so it should be ok
[22:02:23] <twkozlowski>	 greg-g: Do I need to repeat that this is awful planning? :-)
[22:03:11] <greg-g>	 twkozlowski: again, see Deskana :)
[22:03:16] <Nemo_bis>	 lol
[22:03:38] <twkozlowski>	 <shrug>
[22:04:25] <greg-g>	 I'm not everyone's PM :)
[22:04:27] <Nemo_bis>	 twkozlowski: there is no need to announce global rename in advance I think.
[22:09:16] <twkozlowski>	 <shrug>
[22:09:42] <twkozlowski>	 Maybe I'll just ban announcing new features in Tech News from now on.
[22:10:41] <Nemo_bis>	 Nah. Just announce the features you need to find after they're live, and those which explode in your face beforehand. :)
[22:10:56] <Nemo_bis>	 It could be a nice way to keep the two sections past vs. future more balanced.
[22:11:20] <Nemo_bis>	 It may need having two issues in draft at the same time but that's hopefully not a tragedy.
[22:11:42] <twkozlowski>	 They are, generally, only in the past couple of issues there's been this imbalance
[22:12:19] <Nemo_bis>	 Yes, I mean it could be a way to keep a balance more easily/with less effort. Your call though.
[22:12:34] <Nemo_bis>	 twkozlowski: speaking of which, did you fix those bugzilla search links to have relative dates in the end?
[22:12:43] <twkozlowski>	 nope
[22:17:39] <Nemo_bis>	 a search like this is also useful-ish https://bugzilla.wikimedia.org/buglist.cgi?chfield=resolution&chfieldfrom=-1w&chfieldto=Now&chfieldvalue=FIXED&f1=votes&f2=longdescs.count&j_top=OR&list_id=325822&o1=greaterthan&o2=greaterthan&query_format=advanced&resolution=FIXED&v1=0&v2=30
[23:18:26] <Qcoder00>	 Hi people
[23:18:32] <Qcoder00>	 Getting a slower response?
[23:18:40] <Qcoder00>	 Noticing anything/
[23:18:42] <Qcoder00>	 ?