[08:04:49] morning Snaps [08:05:00] * ori-l pokes drdee [08:05:17] man, that's way too early for you [08:05:27] or are you in CET? [08:05:32] i know, i usually don't get going for another hour or so [08:05:33] no, PST [08:05:42] you are kidding [08:05:49] never [08:05:52] :D [08:05:53] right [08:06:13] or you still have to go to bed? [08:06:23] yes, obviously [08:06:39] my timezone math is slow [08:07:01] you found the numbers you needed? [08:07:11] yep [08:07:15] kool [08:07:19] i'm forwarding you an email about it actually [08:07:23] ohohoh [08:07:30] ;) [08:07:42] today is 360 review day [08:07:47] I AM SO EXCITED! [08:10:37] yes, i am also trying not to pass out with excitement [08:12:08] ugh [08:12:14] i only forwarded you the last message [08:17:17] Morning drdee [08:18:19] hey Snaps [08:18:32] what's brewing in Sweden? [08:18:50] beer, one would hope, but Ive seen no such evidence :/ [08:19:21] I talked to Mark yesterday about tracking down the performance bottleneck(s) of your current solution [08:20:05] he'll provide some profiling data later this week, which will be very useful in deciding what route to go with varnishkafka: stay with the varnishncsa codebase or do something more from scratch [08:22:33] and then I'll make a proper stab this weekend [08:28:42] great! is there an email thread or was this on IRC? [08:46:49] it was in a secret irc chat [09:13:37] Snaps: okay :) can you maybe send a quick email to summarize that conversation so we have a 'paper trail' and please CC me, Andrew and Faidon as well [09:14:57] oki [12:34:40] hey manybubbles [12:34:43] hi drdee [12:34:55] hi! [15:48:01] average, aroun? [15:48:12] yo ottomata [15:48:17] yo milimetric [15:48:36] drdee: here [15:48:43] yo [15:48:47] yoyo [15:48:51] hiya [15:49:54] can you look at https://github.com/wikimedia/kraken/blob/master/kraken-generic/src/test/java/org/wikimedia/analytics/kraken/pageview/PageviewTest.java and review the unit-tests for the isWikistatsMobileReportPageview() function [15:50:01] all the tests pass now [15:50:15] yes [15:50:19] but i wanna make sure that my implementation matches your implement atin [15:55:02] milimetric; all unit-tests pass now for the page view implementations [15:55:11] you can also specify the modus in your pig script [15:55:15] oh yay [15:55:31] https://github.com/wikimedia/kraken/commit/b3bcb942bd50f8b9141e1ccb333d793d7cea01b1 [15:55:44] gj drdee [15:56:45] drdee: any insights from messing with the pageview implementation [15:57:00] not yet :( [15:57:05] k [15:57:15] i asked average to review my implementation of his definition [15:57:26] feel free to have a look at https://github.com/wikimedia/kraken/blob/master/kraken-generic/src/test/java/org/wikimedia/analytics/kraken/pageview/PageviewTest.java [15:57:26] cool [15:57:56] i also pushed https://github.com/wikimedia/kraken/commit/b3bcb942bd50f8b9141e1ccb333d793d7cea01b1 [15:58:05] so you can specify the modus in your pig script [15:58:12] we should be able to run some jobs shortly [15:58:17] nice [15:58:29] I was just looking at those [15:58:55] aight, send me your complaints :D [15:59:11] no complaints so far [16:07:28] oh hey erosen [16:07:32] yo yo [16:07:50] i'm all yours today, wanna hangout and squash this zero hurdle? [16:07:59] yes i do [16:08:04] now work? [16:08:06] k, to the batcave [17:02:20] erosen: scrum [17:46:41] oh hey, drdee, we were going to remove webrequest log data from stat1001 this week, weren't we? [17:46:45] sorry [17:46:46] stat1* [17:51:00] ottomata: was the duplication period was 15-20 of may? [17:51:17] was the duplication period 15th to 20th of may? [20:43:35] erosen, whoa i just saw your message [20:43:42] internet has been real crappy at this cafe, sorry [20:43:49] no, may 21-28 [20:43:52] 15-20 was no data [20:48:56] milimetric: bah! your sampeld data is still duplicated [20:48:58] GRRRR [20:49:07] i'm only running packet loss job on wikipedia-mobile data set [20:49:08] ah balls [20:49:10] it is def unduplicated [20:49:20] but i dunno why the sampled didn't unduplicate [20:49:21] i'll have to rerun that [20:49:23] growl [20:49:48] 16-20 no data (inclusive) [20:49:56] that's true [20:49:58] can't do much about that [20:49:59] 21-28 duplicated (inclusive) [20:50:01] right [20:50:06] no correcting above [20:50:17] ? [20:50:24] I was just correcting above for erosen [20:50:30] oh right [20:50:31] 'cause you said 15-20 no data [20:50:33] yeah [20:50:37] yeah something like that [20:50:47] its 16? [20:50:47] yeha [20:51:09] no prob about the duplication, I don't think it's high priority. I mean it's no longer duplicated and I don't think they care about the old numbers [20:51:27] we can talk to tomasz later to see if it's a big deal for him [20:52:02] or tfinc: care to weigh in? We tried to de-duplicate the data that goes into the mobile platform analysis to fix the May 21-May 28 time period [20:52:04] eh? [20:52:04] oh really? [20:52:06] but failed [20:52:17] i mean, first attempt failed [20:52:19] let us know if you need it urgently or else we'll just put it on lower priority [20:52:20] i must have done something wrong [20:52:24] milimetric: in the middle of reviews now. please email me if its urgent. [20:52:30] or the job i ran to dedupliate failed [20:52:32] it shoud lwork [20:53:21] ok welp, milimetric, my compy is about to die, and the internet here basically ran out (i'm on my phone), and none of my stuff is working right now anyway, so i'm going to peace out for the eve [20:53:28] k, well just don't spend too much time [20:53:34] since we're not sure about the priority [20:53:37] i mean, i'd love not to bother [20:53:40] ok [20:53:47] i'll leave it unless someone else wants it then [20:53:51] k [20:53:55] erosen's 0 data i think should be ok though [20:53:56] have a good nite otto [20:53:59] good [20:54:00] zero* [20:54:02] that's very important [20:54:04] he needs to double check [20:54:10] sure, we'll do that [20:54:12] ok danke [20:54:14] alright, i'm outy [20:54:15] latas [20:54:18] danke *you* [20:54:20] New patchset: Stefan.petrea; "Fixing problem with cronjob" [analytics/wikistats] (master) - https://gerrit.wikimedia.org/r/67010 [20:54:43] average: did you find it? [20:54:54] your cron problem? [20:55:53] ottomata: no, but started putting the stuff into gerrit [20:56:41] ah ok [20:56:48] ok, i got to run, i am all yours for helping tomorrow though [20:57:47] ottomata: thank you [21:04:31] New patchset: Stefan.petrea; "Fixing problem with cronjob [IN PROGRESS]" [analytics/wikistats] (master) - https://gerrit.wikimedia.org/r/67010 [21:26:19] milimetric, erosen: sent pivot table with zero data for may [21:26:26] k [22:00:01] aite, i'm out of here for a bit, will try to do the reportcard later tonight