[01:26:38] <drdee>	 dschoon yes that's what i am saying blame ottomata , i also told him not to name that file like that
[01:27:07] <drdee>	 louisdang, we are still stuck with hue
[01:27:31] <drdee>	 we tried running the provided samples by hue and those didn't run either
[01:27:51] <drdee>	 maybe we should start with a new hue.ini file and start from scratch
[13:31:08] <milimetric>	 morning!
[13:44:51] <drdee>	 morning milimetric, average_drifter
[13:44:52] <milimetric>	 morning drdee
[13:53:04] <drdee>	 average_drifter time to build a debian package?
[14:21:05] <average_drifter>	 morning drdee
[14:21:10] <drdee>	 yoyo
[14:21:10] <average_drifter>	 squashing some commits into debianize now
[14:21:21] <average_drifter>	 uhm we might have a small problem
[14:21:37] <average_drifter>	 so basically we had a VERSION oneliner that would produce the next version
[14:21:53] <average_drifter>	 uhm, the problem is that it collides with the current one
[14:21:53] <drdee>	 yup
[14:21:55] <average_drifter>	 not sure why that is
[14:22:33] <drdee>	 are you guys having issues with gmail?
[14:23:11] <average_drifter>	 works for me
[14:23:16] <average_drifter>	 gmail I mean
[14:24:03] <drdee>	 k
[14:26:09] <drdee>	 quoting from david's email:
[14:26:09] <drdee>	 "To at least pair this with some good news: I'm comfortable promising a full-stack Limn release before the December meeting, with all the new d3 goodness, new edit UI, and new options. We should also have bar and area chart-types available. If we're lucky, a world-geo map type, but santa only brings presents to good kids"
[14:27:34] <drdee>	 milimetric ^^
[14:28:16] <milimetric>	 oh good
[14:28:16] <milimetric>	 when was this?
[14:28:54] <milimetric>	 I'm sorry if I was a space cadet and missed the email, am I cc-ed?
[14:32:28] <milimetric>	 drdee ^^
[14:50:32] <ottomata>	 phew, morning all, late morning on my  half day :p
[14:50:32] <ottomata>	 guess i'll be here the rest of the day
[14:51:46] <drdee>	 yo ottomata, have you installed the geoip libs on the cisco's?
[14:51:59] <drdee>	 i am looking for the geoip db's on an01
[14:52:02] <drdee>	 but can't find them
[14:53:02] <ottomata>	  /usr/share/GeoIP
[14:56:20] <drdee>	 ty
[14:56:23] <drdee>	 and about Hue........
[14:57:11] <drdee>	 maybe rollback the change you made regarding the tasktracker
[14:58:54] <ottomata>	 hm, what change was that?
[15:02:24] <average_drifter>	 drdee: can I change the automatic version generating code to include the revision count ?
[15:02:41] <drdee>	 average_drifter, sure
[15:02:41] <average_drifter>	 drdee: should we subscribe to semver.org ?
[15:02:46] <drdee>	 yes
[15:02:47] <average_drifter>	 I mean to their versioning scheme\
[15:02:50] <drdee>	 yes
[15:03:04] <drdee>	 ottomata, we set task tracker thing to no value or something like that
[15:04:48] <drdee>	 hey louisdang, i am having some issues with your page view script
[15:05:04] <louisdang>	 drdee, what is it?
[15:06:04] <drdee>	 did we have a specific gist for the page view code?
[15:06:58] <louisdang>	 no, just email
[15:07:27] <drdee>	 https://gist.github.com/747186677a8f7c4b92b5
[15:07:32] <drdee>	 the error is line 28
[15:07:57] <drdee>	 column 56>  Syntax error, unexpected symbol at or near ','
[15:08:14] <drdee>	 also, i mavenized the whole kraken source code
[15:08:20] <drdee>	 so you might want to resync with wmf-analytics/kraken
[15:08:24] <louisdang>	 I see, this isn't my newest version
[15:08:35] <drdee>	 ohhh argg
[15:08:36] <louisdang>	 yes I've done that
[15:08:46] <drdee>	 cool
[15:08:54] <drdee>	 can you add your version to the gist?
[15:09:20] <drdee>	 i also put your code in a new package: org.wikimedia.analytics.kraken.pig
[15:10:43] <louisdang>	 ok
[15:10:43] <louisdang>	 drdee, https://gist.github.com/1ae782d76288ed8e3c4a
[15:11:27] <louisdang>	 it looks like Pig doesn't like reusing variable names when you group by
[15:11:53] <louisdang>	 so renaming the relation fixed that
[15:13:22] <drdee>	 ottomata, did you install the geoip stuff on all cisco's?
[15:14:58] <ottomata>	 shoudl be
[15:15:39] <louisdang>	 drdee, what does the time stamp  look like again? Like: 2012-11-02T06:00Z?
[15:16:38] <drdee>	 louisdang: 2012-01-25T18:41:25.356
[15:17:09] <drdee>	 ottomata…. weird pig complains it cannot find the geoip db
[15:17:09] <louisdang>	 ok the timestamp converter might not work
[15:17:43] <louisdang>	 drdee, are you using the akela GeoIP UDF?
[15:17:49] <louisdang>	 I had to modify it to get it to work with HDFS
[15:18:00] <drdee>	 i am using your script :)
[15:18:27] <drdee>	 org.wikimedia.analytics.kraken.pig.GeoIpLookup('/usr/share/GeoIP/GeoIP.dat');
[15:18:33] <louisdang>	 ok
[15:20:21] <louisdang>	 drdee, for the first argument in toMonth change that to: 'yyyy-MM-dd'T'HH:mm:ss.SSS' to make sure it works with your timestamp format
[15:21:53] <louisdang>	 drdee, "yyyy-MM-dd\'T'HH:mm:ss.SSS
[15:21:58] <louisdang>	 drdee, "yyyy-MM-dd\'T\'HH:mm:ss.SSS
[15:36:20] <drdee>	 ok
[15:38:01] <drdee>	 louisdang: java.lang.RuntimeException: could not instantiate 'org.wikimedia.analytics.kraken.pig.ConvertDateFormat' with arguments '[yyyy-MM-ddTHH:mm:ss.SSS, yyyy-MM]'
[15:38:27] <louisdang>	 it looks like it's missing the ' ' between T
[15:41:29] <drdee>	 got it, those ' need to be escaped else it won't work either
[15:41:42] <louisdang>	 ok
[15:41:48] <drdee>	 i still have the :java.io.FileNotFoundException: File does not exist: /usr/share/GeoIP/GeoIP.dat
[15:41:59] <drdee>	 which is very strange
[15:42:05] <louisdang>	 is it in HDFS?
[15:42:07] <drdee>	 because the file does exist,could it be a permission issue
[15:42:15] <drdee>	 no
[15:42:17] <drdee>	 it's local fs
[15:42:23] <drdee>	 should it be on hdfs?
[15:42:37] <louisdang>	 if you're using mapreduce mode yeah
[15:43:03] <louisdang>	 or maybe putting file:// in front of it will use the local file
[15:43:20] <louisdang>	 since in mapreduce mode the default context is hdfs
[15:43:53] <drdee>	  java.lang.IllegalArgumentException: Resource name must be relative
[15:43:59] <drdee>	 when trying file://
[15:44:32] <louisdang>	 yeah I don't know exactly how to use a local file in mapreduce mode
[15:44:57] <louisdang>	 can you put it in hdfs and change the argument?
[15:50:46] <louisdang>	 the original akela script said to put it in hdfs too
[16:00:57] <louisdang>	 drdee, it looks like you have to modify the UDF to use the distributed cache to use the file locally: http://ofps.oreilly.com/titles/9781449302641/writing_udfs.html and http://stackoverflow.com/questions/5106679/how-do-i-read-static-files-in-a-pig-udf
[16:07:05] <ottomata>	 i remember having that problem with the akela stuff when I was playing with it
[16:07:21] <ottomata>	 I had a really hard time getting it to load the GeoIP stuff properly
[16:07:52] <ottomata>	 and was getting really confused between local file load and distributed cache load
[16:07:52] <drdee>	 yeah it's not easy :)
[16:07:52] <ottomata>	 but I tried tons of variations
[16:07:52] <ottomata>	 i think that's when I stopped working on it :p
[16:07:59] <louisdang>	 yeah I did too, modifying the UDF seemed to work for me
[16:08:12] <louisdang>	 as long as I the dat file in HDFS
[16:08:57] <louisdang>	 but I think it could be inefficient in a real system if you have a bunch of map tasks accessing the same file
[16:09:46] <drdee>	 i am not getting this
[16:09:46] <louisdang>	 so you can modify the UDF to use the distributed cache as it says here: http://ofps.oreilly.com/titles/9781449302641/writing_udfs.htm
[16:10:10] <drdee>	 if i enter an absolute path it says Resource name must be relative
[16:12:01] <louisdang>	 can you load it into HDFS?
[16:12:07] <drdee>	 the file is both on hdfs and local
[16:12:19] <louisdang>	 did you change the argument to the relative path on HDFS?
[16:12:34] <louisdang>	 say 'GeoIP.dat'
[16:13:54] <louisdang>	 or './GeoIP.dat'
[16:13:54] <drdee>	 File does not exist: GeoIP.dat
[16:13:59] <drdee>	 but how would it know where to find it?
[16:14:13] <louisdang>	 I just put it in my user folder
[16:14:17] <louisdang>	 user/louis/
[16:14:31] <louisdang>	 and used 'GeoIP.dat' and that seemed to work
[16:15:59] <drdee>	 okay it's running but this is ridiculous
[16:16:45] <drdee>	 it assumes a relative path from your home folder on hdf
[16:16:46] <drdee>	 s
[16:16:54] <drdee>	 okay it's not running anymore
[16:16:55] <drdee>	 ob job_1351787518765_0040 has failed! Stop running all dependent jobs
[16:17:10] <drdee>	 ERROR 2997: Unable to recreate exception from backed error: AttemptID:attempt_1351787518765_0040_m_000016_2 Info:Container killed by the ApplicationMaster.
[16:18:02] <louisdang>	 yeah I got that too, put languages.txt in HDFS and change the argument of ParseWikiUrl to 'languages.txt'
[16:18:30] <drdee>	 thx
[16:18:47] <drdee>	 and languages should be where?
[16:19:18] <louisdang>	 one sec didn't work for me
[16:19:18] <louisdang>	 I put it in my user folder though like GeoIP.dat
[16:31:05] <louisdang>	 drdee, there's a fix for this but it involves modifying ParseWikiUrl can I send you the new code?
[16:31:21] <drdee>	 just send a pull request ;)
[16:33:39] <louisdang>	 ok
[16:34:22] <louisdang>	 I get an error using maven: error: generics are not supported in -source 1.3
[16:37:57] <drdee>	 when is that error happening?
[16:38:16] <louisdang>	 compiling
[16:40:47] <drdee>	 are you compiling with java 7 again?
[16:41:54] <drdee>	 louisdang: see v
[16:42:02] <drdee>	 http://stackoverflow.com/questions/7597950/maven-error-generics-are-not-supported-in-source-1-3-i-am-using-1-6
[16:46:08] <drdee>	 louisdang, does that solve the problem?
[16:46:24] <louisdang>	 no
[16:48:43] <drdee>	 :)
[16:49:13] <drdee>	 did you update the pom.xml?
[16:49:44] <louisdang>	 would just editing and saving it work?
[16:51:25] <drdee>	 the pom you mean?
[16:51:34] <drdee>	 yes that would work
[16:52:02] <drdee>	 hold on i can fix it
[16:52:05] <drdee>	 then you have to sync, okay?
[16:52:09] <louisdang>	 ok
[16:55:03] <drdee>	 louisdang; done
[16:55:09] <drdee>	 fetch the new pom.xml and try again
[17:00:35] <ottomata>	 haaaangout!
[17:00:36] <ottomata>	 https://plus.google.com/hangouts/_/2e8127ccf7baae1df74153f25553c443bd351e90
[17:02:48] <louisdang>	 working
[17:07:58] <average_drifter>	 drdee: can you please explain this part from a version of the debianize.sh https://gist.github.com/ceb8145f99dec64356b0
[17:08:21] <drdee>	 sure give me 5 minutes
[17:14:57] <dschoon>	 I am late for no good reason.
[17:14:57] <dschoon>	 Thanks Muni.
[17:15:00] <dschoon>	 Are we still a-scrummin?
[17:15:47] <drdee>	 just finished
[17:15:55] <drdee>	 average_drifter i just replied to your gist question
[17:16:12] <dschoon>	 ah well.
[17:16:23] <dschoon>	 yesterday i did amazing things that i guess y'all will never know about
[17:16:41] <drdee>	 but dschooon milimetric let's talk limn in https://plus.google.com/hangouts/_/2e8127ccf7baae1df74153f25553c443bd351e90
[17:16:44] <dschoon>	 one sec
[17:16:47] <dschoon>	 now i am getting coffee
[17:16:49] <drdee>	 k
[17:18:00] <average_drifter>	 drdee: oh so you mean if the repo dir has a different name than the actual repo name
[17:18:10] <average_drifter>	 drdee: so you want the repo name
[17:18:18] <average_drifter>	 drdee: right ?
[17:18:31] <drdee>	 assuming that the repo name matches the package name, yes
[17:22:59] <ottomata>	 drdee, just ran puppet on an01
[17:23:08] <ottomata>	 buncha hive changes, eh?
[17:23:24] <drdee>	 i hope you puppetized them :)
[17:23:56] <ottomata>	 nope
[17:24:01] <ottomata>	 but I got them from the changes diff
[17:24:06] <ottomata>	 so I can
[17:24:14] <drdee>	 k
[17:28:48] <ottomata>	 yeah, drdee, i will puppetize these now, but go ahead and try your query
[17:28:49] <ottomata>	 in hue
[17:28:50] <ottomata>	 see if it works
[17:29:04] <drdee>	 YES LET"S DO IT
[17:29:04] <ottomata>	 oh, q
[17:29:12] <ottomata>	 hive.intermediate.compression.codec
[17:29:19] <dschoon>	 hokay
[17:29:19] <dschoon>	 back
[17:29:19] <ottomata>	 org.apache.hadoop.io.compress.SnappyCodec
[17:29:20] <ottomata>	 do I need to puppetize snappy now?
[17:29:24] <ottomata>	 hiiiii dschoon!
[17:29:35] <dschoon>	 hola
[17:29:43] <dschoon>	 drdee, milimetric did you want to chat?
[17:29:51] <dschoon>	 i'm in that hangout
[17:29:51] <drdee>	 ottomata, no
[17:29:51] <milimetric>	 k
[17:29:57] <drdee>	 snappy is already built-in hadoop
[17:29:59] <ottomata>	 ok
[17:30:23] <ottomata>	 it was just the snappy cli stuff that you were manually installing, right?
[17:30:30] <drdee>	 yup
[17:33:27] <drdee>	 OTTOMATA IT IS WORKING!!!!!
[17:33:38] <ottomata>	 yeehaawwwww
[17:33:44] <ottomata>	 thanks louisdang for his confs
[17:33:53] <ottomata>	 I used those to slowly check a few pieces til I found that
[17:34:25] <louisdang>	 cool
[17:34:30] <louisdang>	 what was wrong?
[17:39:07] <ottomata>	 hue gets really confused whne mapred.job.tracker
[17:39:13] <ottomata>	 even though we aren't running a jobtracker (since we are using yarn)
[17:39:35] <ottomata>	 drdee and I thought we ahd tried this before, but we used a value of 'ignoreme' instead of a real url like you have
[17:39:35] <ottomata>	 i don't exactly know why
[17:39:47] <ottomata>	 but putting the real url there (even though there is no jobtracker) works
[17:40:04] <louisdang>	 I see
[17:40:20] <louisdang>	 drdee: https://github.com/wmf-analytics/kraken/pull/6
[17:42:00] <ottomata>	 drdee, do I need to create the hive_metrics database?
[17:42:16] <drdee>	 yes please do
[17:42:30] <ottomata>	 did you create it manually?
[17:43:04] <drdee>	 probably
[17:43:05] <ottomata>	 hm, actually, it doesn't exist
[17:43:06] <ottomata>	 hm
[18:08:56] <ottomata>	 drdee, can I test hive metrics somehow?
[18:09:10] <drdee>	 just run a hive query
[18:09:24] <drdee>	 it should store metrics in mysql
[18:14:24] <drdee>	 louisdang: got a new version of the ParseWikiUrl function?
[18:17:27] <louisdang>	 drdee, yes sent pull request
[18:18:14] <drdee>	 ok
[18:20:34] <drdee>	 dschoon, milimetric, got kicked out again but i think we were done
[18:20:49] <dschoon>	 ok
[18:21:56] <ottomata>	 drdee, do I have to create the tables too?
[18:22:25] <drdee>	 i thought that the jdbc string had a option like createTables=true
[18:22:33] <ottomata>	 i've seen create database, hmm
[18:23:30] <ottomata>	 have you actually seen metrics inserted yet?  or are you just guessing? :p
[18:23:46] <ottomata>	 did you test the hive chnages that you made?
[18:26:01] <drdee>	 well we were still trying to get hue to work :)
[18:26:06] <ottomata>	 haha, ok
[18:26:22] <drdee>	 louisdang: you can give hue a try again, it seems that ottomata has solved the problem
[18:26:34] <ottomata>	 were these changes that you made in attempt to make hue work?
[18:26:37] <ottomata>	 should I keep these
[18:26:38] <ottomata>	 ?
[18:26:58] <drdee>	 louisdang: but the pig script is still giving an error: Unable to recreate exception from backed error: AttemptID:attempt_1351787518765_0048_m_000035_1 Info:Container killed by the ApplicationMaster.
[18:27:05] <drdee>	 ottomata, what changes?
[18:27:21] <louisdang>	 hm
[18:27:21] <ottomata>	 https://gist.github.com/4002920
[18:27:26] <louisdang>	 drdee, is there anything more in the log?
[18:28:04] <drdee>	 ottomata, no those were just hive optimizations and customization options
[18:28:28] <ottomata>	 shoudl we keep them?
[18:28:47] <drdee>	 yes please :)
[18:29:00] <drdee>	 i spend the whole weekend reading on them
[18:29:16] <ottomata>	 ok cool
[18:29:53] <ottomata>	 i'm fine with most of them, not so sure about the stats ones…just cause I haven't gotten them to actually do anything yet
[18:29:53] <ottomata>	 and i'm not having much google luck
[18:30:52] <milimetric>	 dschoon: pushed
[18:31:41] <dschoon>	 cool. will check it out
[18:36:07] <drdee>	 louisdang: ERROR 2078: Caught error from UDF: org.wikimedia.analytics.kraken.pig.ParseWikiUrl, Out of bounds access [String index out of range: 3]
[18:36:26] <ottomata>	 yo doooods, i need to get some food, i'll be back on in like an hour
[18:36:38] <drdee>	 happy snacking!
[18:37:08] <drdee>	 me snacking too
[18:38:54] <louisdang>	 drdee, I see the problem fixing it
[18:40:12] <drdee>	 thx
[18:43:23] <dschoon>	 ottomata: any word on progress on the dells?
[18:46:45] <dschoon>	 milimetric: you need to check in some files
[18:46:45] <dschoon>	 404 on jquery.custom
[18:47:18] <milimetric>	 my bad - pushed
[18:47:26] <milimetric>	 forgot to git add .
[18:48:58] <dschoon>	 s'all good.
[18:49:33] <dschoon>	 okay, let's see what we can get done.
[18:50:37] <louisdang>	 drdee, try this: https://github.com/wmf-analytics/kraken/pull/7
[19:53:19] <ottomata>	 baaaaack, things are far in a beach town :/
[19:57:45] <ottomata>	 dschoon, you around?
[20:00:23] <dschoon>	 meeting
[20:00:23] <dschoon>	 brb
[20:00:55] <ottomata>	 ok
[20:03:39] <drdee>	 louisdang: trying again
[20:03:48] <louisdang>	 alright
[20:04:20] <drdee>	 seems to look good!
[20:04:30] <drdee>	 48 maps completed
[20:04:36] <louisdang>	 nice
[20:04:46] <drdee>	 yes very nice and all thanks to you!
[20:04:49] <drdee>	 have you given hue a spin?
[20:04:53] <drdee>	 it should be working now
[20:05:02] <louisdang>	 yes the hive query works
[20:05:51] <louisdang>	 do you think you can fix the pig shell on hue?
[20:07:31] <drdee>	 that should work
[20:07:36] <drdee>	 what is the error you get?
[20:08:03] <louisdang>	 2012-11-02 20:07:49,863 [main] INFO org.apache.pig.Main - Apache Pig version 0.10.0-cdh4.1.1 (rexported) compiled Oct 16 2012, 12:27:24
[20:08:03] <louisdang>	 2012-11-02 20:07:49,863 [main] INFO org.apache.pig.Main - Logging error messages to: /dev/null
[20:08:03] <louisdang>	 2012-11-02 20:07:49,917 [main] ERROR org.apache.pig.Main - ERROR 2997: Encountered IOException. /bin/false/.pig_history (Not a directory)
[20:08:03] <louisdang>	 Details at logfile: /dev/null
[20:08:25] <drdee>	 grumble
[20:08:59] <drdee>	 ottomata, can you help louisdang troubleshooting this pig shell error in hue?
[20:11:26] <ottomata>	 ok
[20:11:51] <ottomata>	 oh interesting
[20:11:52] <ottomata>	 hm, ok
[20:18:25] <ottomata>	 louisdang, what are you running?
[20:19:07] <louisdang>	 right now?
[20:19:21] <louisdang>	 a hive script through beeswax
[20:19:36] <ottomata>	 you are having trouble with pig shell, right?
[20:19:49] <ottomata>	 i want to reproduce
[20:19:58] <louisdang>	 ok
[20:20:39] <louisdang>	 i'm using pig through http://analytics1001.wikimedia.org:8888/shell/create?keyName=pig
[20:21:15] <ottomata>	 right
[20:21:29] <ottomata>	 that happens to you when you just load the shell?
[20:22:43] <louisdang>	 yes
[20:23:30] <ottomata>	 hmm
[20:23:42] <ottomata>	 that's weird
[20:23:42] <ottomata>	 ohhhhhh
[20:23:42] <ottomata>	 interesting
[20:23:42] <ottomata>	 i know
[20:24:40] <ottomata>	 try now
[20:27:22] <louisdang>	 ottomata, I get Error creating temp dir in hadoop.tmp.dir /tmp/hadoop-louisdang due to Permission denied now
[20:28:03] <ottomata>	 hmm
[20:28:54] <ottomata>	 and now?
[20:29:18] <louisdang>	 works now
[20:29:27] <ottomata>	 yay cool
[20:30:11] <louisdang>	 thanks ottomata
[20:31:37] <drdee>	 average_drifter: ping
[20:32:39] <louisdang>	 brb
[20:36:35] <average_drifter>	 drdee: pong
[20:36:47] <average_drifter>	 drdee: uhm, I'm struggling with the debianization
[20:36:54] <drdee>	 what's the problem?
[20:37:37] <average_drifter>	 merged debianize.sh in udp-filters with the one in debianize/
[20:39:01] <drdee>	 k
[20:39:07] <ottomata>	 i spent a bit of time making the debian/ directory work, and I wasn't using debianize.sh to do that
[20:39:07] <ottomata>	 i don't htink
[20:39:10] <ottomata>	 but maybe this was too long ago
[20:40:49] <drdee>	 yeah i think so, average_drifter and i worked on this a couple of weeks ago
[20:41:23] <ottomata>	 drdee, i've got all those changes to hive-site.xml puppetized except for the stats ones
[20:41:30] <drdee>	 ok
[20:41:31] <ottomata>	 well, i mean
[20:41:45] <ottomata>	 they are useable in puppet, but off by default
[20:41:48] <ottomata>	 if we figure out what that is and how to make it work, we can turn it o n
[20:42:00] <drdee>	 i will try to get the metrics stuff to work in the coming days
[20:42:37] <ottomata>	 ok
[20:49:06] <louisdang>	 drdee, did you fix feedback.pig?
[20:49:29] <drdee>	 i first want to run the page views stuff
[20:49:36] <drdee>	 but you can try it yourself now :)
[20:49:48] <drdee>	 just upload the files using hue  to your home folder and give it a  spin
[20:50:45] <louisdang>	 ok
[20:50:57] <louisdang>	 drdee, I need permission to access the file browser
[20:51:08] <drdee>	 1 sec
[20:51:50] <drdee>	 try agin
[20:53:07] <louisdang>	 drdee, works
[20:56:45] <drdee>	 louisdang: geocoding does not seem to work, i only see either '--' as output or a couple of spaces
[20:58:55] <louisdang>	 ok let me try it on hue
[20:59:11] <louisdang>	 since the output is fine using example.log on my computer
[20:59:27] <louisdang>	 are the month, languages, isMobile fine?
[20:59:44] <drdee>	 yes, those look good
[21:00:02] <louisdang>	 drdee, it could be that we're using different dat files. I'm using GeoIPCity.dat from maxmind's website
[21:00:03] <louisdang>	 where did you get yours?
[21:00:50] <drdee>	 we should also sort the output: project, month, site, country
[21:01:07] <louisdang>	 project is the language?
[21:01:11] <drdee>	 yes
[21:01:26] <drdee>	 we have subscription, i used the general GeoIP.dat
[21:01:43] <louisdang>	 ok
[21:01:45] <drdee>	 i assumed that they had the same interface,
[21:02:02] <louisdang>	 yeah I think they would
[21:02:03] <drdee>	 can it work with both types of files?
[21:05:41] <louisdang>	 I don't know. Should we try the newer GeoIP API?
[21:07:47] <drdee>	 what do you mean?
[21:11:13] <louisdang>	 drdee, right now the count.pig script is registering: REGISTER 'geoip-1.2.5.jar'
[21:11:29] <louisdang>	 but there's a geoip-1.2.8 on maxmind's website
[21:12:01] <drdee>	 i don't think it really matters
[21:25:00] <ottomata>	 dschoon, still meeting?
[21:28:41] <louisdang>	 drdee, can you try running getLocation() on an address using GeoIP API and see what you get?
[21:31:58] <dschoon>	 yes, ottomata
[21:34:17] <ottomata>	 hi, ok, i'm responding to your email
[21:34:18] <ottomata>	 about varnish log format
[21:38:48] <mwalker>	 drdee: as jeff_green went home for the day, I was hoping I could get one of y'all to set up a 1:1 gather on meta for the two strings "RecordImpression?" and "BannerRandom?"
[21:39:24] <mwalker>	 it'll be a firehose... but I only need it to run for, say, 5 minutes
[21:41:19] <dschoon>	 okay, i need to eat a food
[21:41:26] <dschoon>	 (meeting over) i will be back soon.
[21:54:58] <dschoon>	 back
[21:55:16] <dschoon>	 ottomata left?
[21:55:33] <drdee>	 seems so
[22:02:40] <louisdang>	 drdee, I can't tell what's wrong with the geocoding right now
[22:03:40] <drdee>	 ok, i'll look into it as well
[22:04:26] <louisdang>	 to make sure that the problem is with the API or not you can try running some of the addresses through getLocation()
[22:18:02] <louisdang>	 drdee, I ran the hive query you originally asked for btw
[22:25:28] <drdee>	 cool
[22:56:24] <drdee>	 all righty folks enjoy your weekend!
[22:56:32] <drdee>	 i am outta here
[22:57:08] <dschoon>	 see ya man
[22:57:09] <dschoon>	 you too