[07:55:13] <grrrit-wm>	 (PS8) Nuria: Add autcomplete to tags on tag insertion [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/145039 (owner: Terrrydactyl)
[08:06:38] <grrrit-wm>	 (PS9) Nuria: Add autcomplete to tags on tag insertion [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/145039 (owner: Terrrydactyl)
[08:27:07] <grrrit-wm>	 (CR) Nuria: Refactor Aggregation constants (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146520 (owner: Milimetric)
[08:34:24] <grrrit-wm>	 (CR) Nuria: "Is this change supposed to be backwards compatible with our prior format? I do not think it needs to be (and from my tests it is not) but " (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146521 (https://bugzilla.wikimedia.org/67822) (owner: Milimetric)
[08:42:10] <grrrit-wm>	 (CR) Nuria: "This is so much better. It will be about 1 order of magnitude of file size savings. The dashboard will be a lot happier. Now we only need " [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146521 (https://bugzilla.wikimedia.org/67822) (owner: Milimetric)
[09:47:01] <wikibugs>	 Analytics / EventLogging: UniversalLanguageSelector-tofu logging too much data - https://bugzilla.wikimedia.org/67463#c1 (nuria)  Language team was logging data to that table at a very high rate without being aware they were doing so. Since their intention was to run an experiment for couple weeks they...
[12:22:57] <grrrit-wm>	 (CR) QChris: Avoid pickle max recursion while serializing chain (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146297 (https://bugzilla.wikimedia.org/67823) (owner: Milimetric)
[14:23:33] <grrrit-wm>	 (CR) Ottomata: Document 'refinery-' prefix for bin directory (1 comment) [analytics/refinery] - https://gerrit.wikimedia.org/r/145976 (owner: QChris)
[14:27:57] <grrrit-wm>	 (PS2) Ottomata: Document naming conventions [analytics/refinery] - https://gerrit.wikimedia.org/r/145421 (owner: QChris)
[14:28:04] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] Document naming conventions [analytics/refinery] - https://gerrit.wikimedia.org/r/145421 (owner: QChris)
[14:34:32] <grrrit-wm>	 (PS1) Ottomata: s/kraken/refinery in refinery-tools/README.txt [analytics/refinery/source] - https://gerrit.wikimedia.org/r/146778
[14:34:34] <grrrit-wm>	 (PS1) Ottomata: Build refinery version 0.0.1 against 2.3.0-cdh5.0.2 [analytics/refinery/source] - https://gerrit.wikimedia.org/r/146779
[14:34:52] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] s/kraken/refinery in refinery-tools/README.txt [analytics/refinery/source] - https://gerrit.wikimedia.org/r/146778 (owner: Ottomata)
[14:35:08] <ottomata>	 qchris: https://gerrit.wikimedia.org/r/#/c/146779/
[14:36:18] <qchris>	 ottomata: Looks good. Let me try to build it.
[14:39:01] <grrrit-wm>	 (CR) QChris: [C: 1 V: -1] "Building for me fails with" [analytics/refinery/source] - https://gerrit.wikimedia.org/r/146779 (owner: Ottomata)
[14:48:28] <grrrit-wm>	 (CR) Milimetric: Add autcomplete to tags on tag insertion (4 comments) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/145039 (owner: Terrrydactyl)
[14:48:55] <milimetric>	 nuria: ^ review posted with note for custom bindings
[14:49:03] <nuria>	 k
[14:50:00] <nuria>	 will grab it once i am done watching steve sanderson presentation
[14:53:01] <grrrit-wm>	 (CR) Milimetric: Avoid pickle max recursion while serializing chain (1 comment) [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146297 (https://bugzilla.wikimedia.org/67823) (owner: Milimetric)
[14:53:19] <ottomata>	 hmm, qchris, i'm having troulbe with a snappy dep building now
[14:53:25] <ottomata>	 i didn't before...
[14:55:15] <qchris>	 :-/
[15:01:21] <ottomata>	 ah needed to add it to archiva
[15:01:25] <ottomata>	 there we go
[15:01:27] <ottomata>	 ok try now qchris
[15:03:53] <qchris>	 Build worked. Let me check the cdh part.
[15:04:51] <grrrit-wm>	 (CR) QChris: [C: 2 V: 2] Build refinery version 0.0.1 against 2.3.0-cdh5.0.2 [analytics/refinery/source] - https://gerrit.wikimedia.org/r/146779 (owner: Ottomata)
[15:05:56] <ottomata>	 danke!
[15:18:25] <grrrit-wm>	 (PS1) Ottomata: Add refinery-tools-0.0.1.jar artifact and symlink [analytics/refinery] - https://gerrit.wikimedia.org/r/146787
[15:18:31] <ottomata>	 qchris ^ :)
[15:18:46] <qchris>	 Hahaha. I waited for that one :-D
[15:18:52] <ottomata>	 haha, i knew you were :)
[15:18:55] <ottomata>	 also, once that is done
[15:18:56] <ottomata>	 https://gerrit.wikimedia.org/r/#/c/133519/
[15:18:56] <ottomata>	 :)
[15:20:01] <qchris>	 Jajajaja ... you're justhe CR requests, so we cannot prepare for the ETL design discussion and have a easy game convincing us of your approach :-P
[15:20:18] <ottomata>	 oh haha
[15:20:21] <ottomata>	 are you prepping?
[15:20:22] <ottomata>	 i am not!
[15:20:42] <qchris>	 No. I am not preparing ... I am reviewing code :-P
[15:20:43] <ottomata>	 i want hive to work, and i want to be able to test stuff, and i don't want to have to manually copy the tools.jar to do so :)
[15:20:47] <ottomata>	 just doing things right!
[15:20:47] <ottomata>	 haha
[15:20:49] <ottomata>	 good!
[15:20:54] <ottomata>	 these are easy peasy reviews though
[15:20:58] <ottomata>	 just close your eyes and hit submit!
[15:24:18] <qchris>	 ll
[15:24:29] <qchris>	 Sorry.
[15:26:38] <ottomata>	 eh?
[15:27:07] <qchris>	 Sorry. Got pinged typing away 'll'.
[15:27:37] <qchris>	 The 'll' was not meant for IRC.
[15:29:36] <grrrit-wm>	 (CR) QChris: [C: 2 V: 2] Add refinery-tools-0.0.1.jar artifact and symlink [analytics/refinery] - https://gerrit.wikimedia.org/r/146787 (owner: Ottomata)
[15:32:24] <grrrit-wm>	 (PS5) Ottomata: Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519
[15:35:34] <grrrit-wm>	 (PS2) Milimetric: Refactor All Enums to a separate file [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146520
[15:38:59] <grrrit-wm>	 (PS6) Ottomata: Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519
[15:42:07] <grrrit-wm>	 (CR) Nuria: [C: 2] "This looks a lot better. Thank you. I think this can be merged." [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146520 (owner: Milimetric)
[15:42:45] <milimetric>	 grr, having rebasing issues, sorry nuria, I'm trying to get the other patch in
[15:43:21] <nuria>	 the 146520 will be merged by jenkins i think
[15:45:22] <grrrit-wm>	 (CR) Milimetric: [V: 2] "jenkins looks a little stuck" [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146520 (owner: Milimetric)
[15:46:47] <grrrit-wm>	 (PS4) Milimetric: Make coalesced report format more efficient [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146521 (https://bugzilla.wikimedia.org/67822)
[15:47:44] <milimetric>	 ok, better
[17:16:11] <grrrit-wm>	 (CR) BryanDavis: "recheck" [analytics/wikimetrics] - https://gerrit.wikimedia.org/r/146520 (owner: Milimetric)
[17:53:37] <ottomata>	 qchris: https://gerrit.wikimedia.org/r/#/c/133519/ :D
[17:54:18] <qchris>	 ottomata: :-D
[17:54:38] <ottomata>	 also, ok, i'm not sure what's wrong with my sequence fil eerror
[17:54:43] <ottomata>	 where are you testing, in labs?  vagrant?
[17:54:51] <qchris>	 In labs.
[17:54:52] <ottomata>	 how are you putting data in?
[17:54:57] <ottomata>	 which machines?  mine?
[17:55:00] <qchris>	 Through my cluster-scripts.
[17:55:09] <qchris>	 No. the worker nodes.
[17:55:12] <ottomata>	 qchris-worker?
[17:55:21] <qchris>	 qchris-worker2 typically.
[17:55:30] <qchris>	 But I am just bootstrapping the cluster again ...
[17:55:36] <qchris>	 So it's currently not up.
[17:55:47] <ottomata>	 oh
[17:55:55] <ottomata>	 ok, i noticed that earlier, was like, but hive and hfds aren't evne here!
[17:56:26] <qchris>	 Yep. ... But getting distracted by meetings and gerrit stuff :-D
[17:58:59] <ottomata>	 aye
[17:59:11] <ottomata>	 i'm running them on my cluster :)
[18:00:04] <qchris>	 So how can I help debug?
[18:00:18] <qchris>	 Like trying to get data in from analytics1011?
[18:00:49] <qchris>	 Or could you post the error message?
[18:02:23] <ottomata>	 https://gist.github.com/ottomata/5444e31ef204501cceba
[18:02:31] <ottomata>	 thta's jsut from a select count(*) ...
[18:03:40] <Ironholds>	 oh yay that error!
[18:03:49] <ottomata>	 you've seen that Ironholds?
[18:04:16] <Ironholds>	 yeah, I got it running over files that did not exist.
[18:04:27] <Ironholds>	 it's worrying how many hadoop error messages I can interpret now
[18:04:39] <Ironholds>	 (old query with wrong month=foo parameter, pointing at partitions that were no longer containing files)
[18:04:45] <qchris>	 I know thaht error.
[18:04:57] <qchris>	 Let me find it in my notes ...
[18:07:25] <ottomata>	 HMMMM
[18:07:31] <ottomata>	 so that maybe means the location is wrong?!
[18:08:16] <qchris>	 ottomata ... there is no table wmf on the cluster ... you're using otto?
[18:08:46] <ottomata>	 yes, but i'm also considering renaming the 'wmf' database to 'external'
[18:08:50] <ottomata>	 so there is an external db right now
[18:08:52] <ottomata>	 but i'm using otto
[18:09:10] <qchris>	 Ok.
[18:10:10] <qchris>	 webrequest_text/hourly ... shouldn't that be hourly/webrequest_text
[18:11:47] <ottomata>	 no
[18:11:54] <ottomata>	  /wmf/data/external/webrequest/webrequest_text/hourly/2014/07/16/18/webrequest_text.18.10.23116.3434569163.1405533600000
[18:11:59] <qchris>	 ok.
[18:13:32] <ottomata>	 this is how i created a partition
[18:13:32] <ottomata>	 alter table webrequest add partition (webrequest_source='text',year=2014,month=07,day=15,hour=14)   location '/wmf/data/external/webrequest/webrequest_text/hourly/2014/07/15/14';
[18:13:40] <ottomata>	 sorry, i'm looking at the external.webrequest table now
[18:13:45] <ottomata>	 since that's the one I really want to work
[18:13:56] <qchris>	 Ok ... switching over.
[18:16:34] <qchris>	 'select * from webrequest where year > 0 limit 1;' works for me on that table.
[18:18:09] <qchris>	 ottomata: Could you please create me a home directory in hdfs?
[18:18:12] <qchris>	 I am getting
[18:18:17] <ottomata>	 ja
[18:18:22] <qchris>	 Permission denied: user=qchris, access=WRITE, inode="/user":hdfs:hadoop:drwxrwxr-x
[18:18:51] <qchris>	 Thanks.
[18:18:53] <ottomata>	 i haven't remade everyone's user dirs yet because I think we want to put some thought into the stats user
[18:18:55] <ottomata>	 but ja, there you go
[18:19:05] <qchris>	 Thanks.
[18:20:03] <qchris>	 ottomata: Even 'select count(*) from webrequest where year > 0 limit 1;' worked for me.
[18:20:11] <qchris>	 What query is giving you the error?
[18:20:44] <ottomata>	 you know, i got that to work sometimes too, maybe it always works that way on the webrequest table
[18:20:46] <ottomata>	 lemme see
[18:22:00] <qchris>	 "sometimes" ... that sounds fun.
[18:22:28] <ottomata>	 ja ok, count works, it wasn't working in labs or somewhere else, moving back to a real query, checking...
[18:22:33] <ottomata>	 ja
[18:22:34] <ottomata>	 SELECT
[18:22:34] <ottomata>	   referer,
[18:22:34] <ottomata>	   COUNT(DISTINCT ip) AS hits
[18:22:34] <ottomata>	 FROM
[18:22:34] <ottomata>	   webrequest
[18:22:34] <ottomata>	 WHERE
[18:22:35] <ottomata>	    webrequest_source = 'mobile' AND year = 2014 AND  month= 07 AND  day = 14 AND  hour = 14
[18:22:35] <ottomata>	 GROUP BY referer
[18:22:36] <ottomata>	 ORDER BY hits DESC
[18:22:36] <ottomata>	 LIMIT 50;
[18:22:43] <ottomata>	 try that
[18:22:49] * qchris tries
[18:23:06] <ottomata>	 DAY = 14
[18:23:10] <ottomata>	 THERE IS NO DAY = 14 partition
[18:23:11] <ottomata>	 !!!
[18:23:15] <ottomata>	 GRRRRR am i that blind?!
[18:23:37] <qchris>	 Ha.
[18:23:55] <ottomata>	 AH
[18:23:55] <ottomata>	 no, phew
[18:23:57] <ottomata>	 it still errosr
[18:23:59] <ottomata>	 with day = 15
[18:24:03] <ottomata>	 same error
[18:24:30] <ottomata>	 i'm going to explictly add another partition and use it, just to be sure
[18:26:35] <qchris>	 ottomata: Quoting does the trick. When quoting the partitions it starts working.
[18:26:48] <ottomata>	 in the query or in the add partitoins statement
[18:26:48] <qchris>	 like:
[18:26:51] <qchris>	 SELECT  referer,   COUNT(DISTINCT ip) AS hits FROM   webrequest WHERE    webrequest_source = 'mobile' AND year = '2014' AND  month= '07' AND  day = '15' AND  hour = '15' GROUP BY referer ORDER BY hits DESC LIMIT 50;
[18:27:10] <qchris>	 in the query.
[18:29:16] <ottomata>	 MMMMM
[18:29:30] <ottomata>	 it does work
[18:29:33] <qchris>	 \o/
[18:29:34] <ottomata>	 welllllLLLLL ok then!
[18:29:40] <ottomata>	 quotes it is!
[18:29:50] <qchris>	 :-D
[18:29:53] <ottomata>	 thank you qchris, not sure why that didn't work when I tried it before, mabye my labs or otto table was all weird anyway
[18:30:25] <qchris>	 yw
[18:31:02] <ottomata>	 well, good!
[18:31:24] <ottomata>	 qchris, you ok with the 'external' database name? I think its better, and will allow us to keep the raw external data separate from what we intend people to use
[18:41:06] <grrrit-wm>	 (PS1) Ottomata: Import webrequest_upload into HDFS as well [analytics/refinery] - https://gerrit.wikimedia.org/r/146842
[18:41:22] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] Import webrequest_upload into HDFS as well [analytics/refinery] - https://gerrit.wikimedia.org/r/146842 (owner: Ottomata)
[18:41:39] <ottomata>	 qchriiiis, gimme some pluses on https://gerrit.wikimedia.org/r/#/c/133519/ :D
[18:41:42] <ottomata>	 i want to deploy!
[18:41:48] <grrrit-wm>	 (PS7) Ottomata: Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519
[18:44:28] <qchris>	 sorry, ottomata.
[18:45:06] <qchris>	 About the "external" database name ... not sure. What do we want to convey there?
[18:45:26] <qchris>	 To me, it's more like "raw"
[18:45:40] <qchris>	 because ... being "external" is not really descriptive ... is it?
[18:47:06] <qchris>	 Maybe "wmf-raw"?
[18:47:13] <ottomata>	 ha, i had raw once, milimetric didn't lik eit
[18:47:14] <ottomata>	 liek it
[18:47:18] <ottomata>	 (oh cant' type)
[18:47:31] <qchris>	 Does milimetric like "external"?
[18:47:39] <ottomata>	 that's why 'external' is in the file path
[18:48:03] <milimetric>	 er?
[18:48:06] <qchris>	 (I thought 'external' in the path meant 'this data is used in external tables')
[18:48:16] <milimetric>	 I didn't like raw?
[18:48:27] <qchris>	 milimetric: likes 'raw'?
[18:48:57] <milimetric>	 wait, what does raw refer to?
[18:49:03] <milimetric>	 untouched from camus?
[18:49:11] <qchris>	 yes (I hope)
[18:49:20] <milimetric>	 then yeah, I think that's great
[18:49:43] <milimetric>	 I think I didn't like "raw" to be something else, like, after geocoding or sanitizing
[18:50:11] <qchris>	 Cool.
[18:50:24] <milimetric>	 'cause then it's like maybe medium-rare or rare but not raw :)
[18:50:24] <qchris>	 Makes sense.
[18:50:30] <qchris>	 :-D
[18:50:46] <ottomata>	 ummm, ok raw then is fine with me?
[18:50:50] <ottomata>	 database name:
[18:50:54] <ottomata>	 wmf-raw? or just raw?
[18:51:12] <ottomata>	 i will switch path to /wmf/data/raw...
[18:51:33] <qchris>	 What will the database for processed data be called?
[18:51:33] <ottomata>	 i think jsut 'raw' is fine
[18:51:58] <ottomata>	 wmf
[18:51:58] <ottomata>	 ?
[18:52:13] <qchris>	 It is not too obvious thaht 'raw' will get turned into 'wmf'.
[18:52:18] <ottomata>	 haha
[18:52:21] <ottomata>	 refined
[18:52:22] <ottomata>	 ?
[18:52:22] <ottomata>	 hah
[18:52:28] <qchris>	 Hence, 'wmf-raw'.
[18:52:35] <ottomata>	 ok ok
[18:52:39] <qchris>	 But yes ... raw -> refined would work.
[18:52:39] <ottomata>	 wmf-raw is fine with me
[18:53:07] <ottomata>	 i like 'wmf-raw' and 'wmf' better, s'ok?
[18:53:12] <qchris>	 Sure.
[19:09:12] <grrrit-wm>	 (PS1) Ottomata: Import webrequest data via camus into /wmf/data/raw directory [analytics/refinery] - https://gerrit.wikimedia.org/r/146846
[19:13:52] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] Import webrequest data via camus into /wmf/data/raw directory [analytics/refinery] - https://gerrit.wikimedia.org/r/146846 (owner: Ottomata)
[19:14:36] <ottomata>	 doodee dooo, wouldn't qchris love to review this one?
[19:14:39] <ottomata>	 https://gerrit.wikimedia.org/r/#/c/133519/
[19:14:58] <ottomata>	 sorry qchris, just want to do a deploy and it'd be nice to have this go along with it so I don't have to do another one later
[19:15:02] <qchris>	 :-D
[19:15:15] <ottomata>	 id eploy this, re do the database, paths, etc., start importing upload, and bam, good to go for this part
[19:15:50] <ottomata>	 if you think there is a lot more discussion on sequence-file tool then we can postpone and I can do another deploy later i guess
[19:16:06] <qchris>	 LI do not think that there'll be much more discussion.
[19:16:14] <qchris>	 Let me just go through it ...
[19:16:19] <qchris>	 Doing it right now.
[19:16:53] <qchris>	 '# --jar default should be in ../../artifacts/ directory'
[19:16:58] <qchris>	 Looks like it should be
[19:17:02] <qchris>	 # --jar default should be in ../artifacts/ directory
[19:17:04] <ottomata>	 oh in the comment
[19:17:04] <ottomata>	 ok
[19:17:38] <grrrit-wm>	 (PS8) Ottomata: Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519
[19:18:19] <qchris>	 PS7 and PS8 agree for me.
[19:18:20] <ottomata>	 anythign else?
[19:18:28] <ottomata>	 that was a rebase
[19:18:34] <qchris>	 Oh. Ok.
[19:18:51] <qchris>	 So the '../../artifacts' is the only thing that I see.
[19:18:53] <ottomata>	 will submit PS9 if nothign else
[19:18:53] <ottomata>	 ok
[19:19:04] <grrrit-wm>	 (PS9) Ottomata: Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519
[19:19:32] <grrrit-wm>	 (CR) QChris: [C: 2 V: 2] Add bin/sequence-file wrapper for refinery-tools [analytics/refinery] - https://gerrit.wikimedia.org/r/133519 (owner: Ottomata)
[19:20:09] <ottomata>	 thank you!
[19:20:22] <qchris>	 Anything else to merge before you can deploy?
[19:20:27] <ottomata>	 naw think that's it
[19:20:30] <qchris>	 Cool.
[19:27:51] <grrrit-wm>	 (PS1) Ottomata: Fix typo in camus/camus.webrequest.properties [analytics/refinery] - https://gerrit.wikimedia.org/r/146852
[19:28:05] <grrrit-wm>	 (CR) Ottomata: [C: 2 V: 2] Fix typo in camus/camus.webrequest.properties [analytics/refinery] - https://gerrit.wikimedia.org/r/146852 (owner: Ottomata)
[19:33:49] <ottomata>	 qchris, i know I am Mr. interrupter today, but i'm just curious!
[19:33:55] <ottomata>	 how's the add partition reworking going?
[19:34:17] <qchris>	 Hahaha.
[19:34:27] <qchris>	 It's not gonna happen today.
[19:34:46] <qchris>	 Sorry.
[19:34:54] <qchris>	 (Although it is simple)
[19:34:55] <ottomata>	 oh i know
[19:34:59] <ottomata>	 i'm just curious about it
[19:35:05] <ottomata>	 did bundles end up making sense?
[19:35:31] <qchris>	 Yes. But we might need to use the same coordinator 4 times.
[19:35:40] <qchris>	 But that might change once everything is in place.
[19:36:00] <ottomata>	 like, instead of a bundle?
[19:36:16] <qchris>	 No. Like the bundle injecting
[19:36:19] <ottomata>	 hm aye
[19:36:23] <ottomata>	 k
[19:36:26] <qchris>	 the same coordinator with 4 different properties.
[19:36:55] <qchris>	 Let me rephrase that ...
[19:37:22] <qchris>	 "Like the bundle injecting the same coordinator 4 times. Each time with a slightly modified properties file"
[19:37:53] <ottomata>	 aye
[19:38:14] <ottomata>	 we could also just submit 4 different coordinators with different properites files manually too, but the bundle makes it nicer to admin via the cli, right?
[19:38:18] <ottomata>	 groups the 4 jobs together?
[19:38:39] <ottomata>	 hm, since the partitions need to be strings, i wonder if I should create the table partitioned by strings
[19:38:42] <ottomata>	 PARTITIONED BY (
[19:38:42] <ottomata>	   `webrequest_source` string,
[19:38:42] <ottomata>	   `year` int,
[19:38:42] <ottomata>	   `month` int,
[19:38:42] <ottomata>	   `day` int,
[19:38:42] <ottomata>	   `hour` int)
[19:38:42] <ottomata>	 hm
[19:38:51] <ottomata>	 i guess it works like that...
[19:39:22] <qchris>	 Right, the bundle bundles them together.
[19:39:45] <qchris>	 I am not sure about the strings vs int partitions.
[19:39:53] <qchris>	 ints feel right here ...
[19:40:10] <qchris>	 the fact that we somewhat have to treat them as strings is bad though.
[19:40:35] <qchris>	 But I did not hunt the issue down.
[19:40:49] <ottomata>	 aye, don't worry, ok, i'll mess with it
[19:40:50] <qchris>	 Not sure ... maybe it is on purpose, maybe a hive issue, maybe a cdh issue.
[19:41:07] <qchris>	 I'd just keep ints for now, and quote them when querying.
[19:55:23] <ottomata>	 ok, if I define the partition columns as strings, and add the partitions as strings
[19:55:30] <ottomata>	 then querying via strings and via ints works
[19:55:36] <ottomata>	 so I think that is the better way to go
[20:01:13] <grrrit-wm>	 (PS1) Ottomata: Add webrequest table schema [analytics/refinery] - https://gerrit.wikimedia.org/r/146878
[21:12:27] <marktraceur>	 leila: You should register leilatretikov.
[21:12:45] <leila>	 hahaha!
[21:12:53] <leila>	 seriously! ;-)
[21:12:53] <YuviPanda>	 :P
[21:13:00] <YuviPanda>	 leila: you should make him put money in the jar
[21:13:01] <leila>	 and she should register LilaZia
[21:13:14] <leila>	 why? did he do the bad thing?
[21:13:16] <marktraceur>	 I'm not mistaking anything! I'm fully cognizant of the difference.
[21:13:18] <YuviPanda>	 ah
[21:13:20] <YuviPanda>	 right.
[21:13:31] <YuviPanda>	 oh, I'm in -analytics.
[21:13:33] <YuviPanda>	 nevermind then :)
[21:13:47] <marktraceur>	 What
[23:14:04] <wikibugs>	 Analytics / Refinery: Story: AnalyticsEng has daily PageView counts - https://bugzilla.wikimedia.org/67804#c1 (Kevin Leduc) Following discussions and note taking here: http://etherpad.wikimedia.org/p/analytics-design  Requirements: * Use the cluster * Sample Report (tsv) | Hour | Project  | Total Pagev...
[23:14:34] <wikibugs>	 Analytics / Refinery: Story: AnalyticsEng has hourly PageView counts - https://bugzilla.wikimedia.org/67804 (Kevin Leduc)