[14:45] <mwak> Anyone is familiar with hadoop charm?
[14:46] <mwak> Trying to run it on arm but get some errors
[14:46] <mwak> http://pastebin.com/i20qtEgf
[14:46] <mwak> http://pastebin.com/JxMHzbdH
[14:52] <jcastro> mwak, you're looking for lazyPower ^^^
[15:00] <bloodearnest> rick_h_, hey, are you guys still gonna be in london week after next?
[15:02] <rick_h_> bloodearnest: yes
[15:02] <rick_h_> 19-23
[15:03] <jamespage> rick_h_, I'm assuming that you guys are aware that the machine view appears to get a little confused by suboridnate charms
[15:03] <jamespage> keeps generating them for placement
[15:03] <rick_h_> jamespage: /me checks bugs
[15:03] <jcastro> lazyPower, also can you see if the Pittsburgh Code and Supply talk can be recorded?
[15:04] <bloodearnest> rick_h_, so, the number of reasons for me to visit bluefin are increasing, you wanna schedule a slot to talk?
[15:04] <rick_h_> bloodearnest: yes, I would like to. Know what day works for you?
[15:05] <bloodearnest> rick_h_, weds?
[15:06] <rick_h_> bloodearnest: wfm
[15:07] <rick_h_> jamespage: closest I can find is https://bugs.launchpad.net/juju-gui/+bug/1369576 marked fix released months ago
[15:07] <mup> Bug #1369576: juju-gui MV shows subordinate services as unplaced units <juju-gui:Fix Released by hatch> <https://launchpad.net/bugs/1369576>
[15:07] <rick_h_> jamespage: so a bug report/screenshot/etc would be great
[15:07] <jamespage> rick_h_, will do
[15:08] <jamespage> rick_h_, thats exactly what i see
[15:08] <rick_h_> jamespage: ah ok, so regression perhaps or old juju-gui version?
[15:08] <jamespage> rick_h_, freshly deployed in the last hour
[15:08] <rick_h_> jamespage: ok, regression then
[15:08] <jamespage> so I would suspect regression
[15:08] <rick_h_> had to check :)
[15:09] <rick_h_> jamespage: I'll just reopen that bug and we'll take a look then if that's exactly what you're seeing
[15:33] <skay> I had a lot of python-django questions, so instead of asking here I posted to the list
[16:08] <jose> arosales: ping, mind a PM?
[16:42] <lazyPower> jcastro: jxreese says he'll try to make that happen.
[16:42] <arosales> jose sure :-)
[16:42]  * arosales doesn't mind at all
[16:43] <lazyPower> mwak: which charm is this? is this the 'hadoop' charm or 'hdp-hadoop'?
[16:44] <mwak> hadoop
[16:44] <mwak> https://manage.jujucharms.com/charms/trusty/hadoop
[16:45] <lazyPower> mwak: ok, i'm not nearly as familiar witht he hadoop charm. Looking at this - can you verify the hadoop service is up and running? if you execute `jps` you should see a few processes.
[16:45] <mwak> ubuntu@onlinelabs-15ae7c187615497fa4d5dc9f62332bae:~$ jps
[16:45] <mwak> 12936 Jps
[16:45] <mwak> 23665 NameNode
[16:45] <mwak> 19137 ResourceManager
[16:45] <lazyPower> hmm, it appears the services are running... interesting
[16:45] <mwak> yep
[16:45] <lazyPower> yet you're getting a connection timeout
[16:46] <lazyPower> single host deploy or multi-host?
[16:46] <mwak> got this :9000 failed on connection exception: java.net.ConnectException: Connection refused;
[16:46] <mwak> details
[16:46] <mwak> 15:47 < mwak> http://pastebin.com/i20qtEgf
[16:46] <mwak> 15:47 < mwak> http://pastebin.com/JxMHzbdH
[16:46] <mwak> multi-host deployment
[16:46] <lazyPower> ah i missed the first paste
[16:46] <mwak> np
[16:47] <lazyPower> can you pastebin me your site.xml?
[16:47] <lazyPower> should be in /etc/hadoop/conf
[16:47] <mwak> sure
[16:47] <mwak> hum
[16:47] <mwak> i ve no site.xml
[16:47] <mwak> in /etc/hoadoop/conf
[16:48] <lazyPower> thats for sure the location in hdp-hadoop - it might be different in the apache hadoop charm
[16:48] <mwak> http://pastebin.com/x4ghffEE
[16:48] <mwak> have the following in this dir
[16:51] <lazyPower> hmm... can you pastebin the core-site.xml as a start?
[16:53] <mwak> sure
[16:53] <mwak> http://pastebin.com/LRB4pBYd
[16:53] <lazyPower> mwak: does a hdfs dfsadmin -report listing work?
[16:54] <mwak> negative
[16:54] <mwak> http://pastebin.com/CkGtpYpu
[16:54] <mwak> connection refused
[17:00] <lazyPower> mwak: i'm grepping through notes i've got over here - i'm not sure why its missing site.xml and why nothing appears to be started when jps is reporting the services are active. I'm composing an email to our resident big data expert who's out on vacation this week to see if i cant get a quick troubleshooting list to run through
[17:00] <lazyPower> mwak: no ETA on the response but i'll take point as POC until we get you resolved.
[17:01] <mwak> lazyPower: alright
[17:02] <mwak> note that it is running on arm 32bit servers
[17:05] <lazyPower> mwak: would you happen to be from our friends over at onlinelabs
[17:07] <mwak> hum, not sure to understand
[17:07] <jose> arosales: sorry, my ISP cut my connection and it's back on, shooting now
[17:08] <lazyPower> mwak: no worries - just curious.
[17:09] <lazyPower> mwak: can you reach out in private with your email so i can cc you on this?
[17:09] <mwak> oh sure
[17:09] <mwak> PM
[17:10] <lazyPower> confirmed - thanks mwak. i'll get you on a comms about this with the big data team
[17:10] <mwak> alright, if they want an access to a similar environment, ping I will give them an access
[17:10] <mwak> could be more confortable
[17:12] <lazyPower> mwak: additionally - we haven't been testin in an arm32 lab ourselves - so we might be pioneering with the charm(s)
[17:12] <mwak> :)
[18:13] <lazyPower> mwak: are any of these pastbin links set to self destruct?
[18:13] <lazyPower> i'm referencing them in my comms - jsut making sure i dont need to copy them out to a non-self-destructing pastebin before i move forward
[18:20] <mwak> lazyPower: no
[18:22] <lazyPower> perfect. sending now
[18:24] <lazyPower> mwak: alternatively - if you want to join in the development discussion / participate in the future of the charms - we have a bigdata-developer mailing list over here: https://launchpad.net/~bigdata-dev  that you can register for with your launchpad account. I'll reach out if/when i hear back from Amir - our lead developer on the charms. He's our resident big data guru, and is scheduled to be back by Tuesday of next week - but he might get back to
[18:24] <lazyPower>  us via email before then.
[20:39] <johnny_shieh> Hi, is there online documentation and a URL site (which I guess the documentation would point me to) to a sandbox ubuntu system where I can test out my charm?
[20:40] <lazyPower> johnny_shieh: certainly - we have a vagrant box at your disposal
[20:40] <lazyPower> https://jujucharms.com/docs/config-vagrant
[20:40] <lazyPower> johnny_shieh: and https://jujucharms.com/docs/howto-vagrant-workflow
[20:41] <johnny_shieh> cool, thanks!
[20:41] <lazyPower> aisrael: wait didn't we change the workflow to no longer require sshuttle?
[20:43] <aisrael> lazyPower: Hm. I guess not.
[20:43] <lazyPower> aisrael: well, i meant the process not the doc :D
[20:44] <lazyPower> i'll file a bug against teh docs for this so we dont forget we need to update that page
[20:44] <aisrael> lazyPower: Yeah, we've totally got a better process for it now. I kind of wonder if it can be baked into the vagrant image itself.
[20:45] <lazyPower> aisrael: https://github.com/juju/docs/issues/218
[20:45] <lazyPower> nice
[20:46] <aisrael> lazyPower: excellent, thanks! Dropped the link to the blog post in there.
[20:46] <lazyPower> johnny_shieh: when you get to reaching inside the vagrant image - might want to take a look at the bug post above as it has some workflow items that will save you a headache with sshuttle
[20:46] <johnny_shieh> okay, thanks.
[20:52] <johnny_shieh> hey, if I'm still on Mavericks, I'm okay with the shuttle issue right?
[20:52] <johnny_shieh> I mean, I can use shuttle
[20:52] <aisrael> johnny_shieh: Yep, you can.
[20:53] <johnny_shieh> thx
[20:54] <aisrael> johnny_shieh: You may notice some network slowness over sshuttle, fwiw.
[20:54] <johnny_shieh> okay
[20:55] <lazyPower> yeah, the encryption overhead and funky packet latency
[20:56] <aisrael> It essentially tears down every packet and reassembles it on the other side of the ssh connection.
[23:35] <mbruzek> Hello beisner, lazyPower asked me to send you this MP link. https://code.launchpad.net/~mbruzek/charms/precise/rabbitmq-server/add-tests/+merge/246025
[23:36] <mbruzek> Please review the amulet tests for rabbitmq-server and let me know if you have any feedback.
[23:36] <mbruzek> Thanks.