Monday, 2014-07-21

*** xianghui has quit IRC01:07
*** xianghui has joined #openstack-sahara01:12
*** bandarji has quit IRC01:22
*** Networkn3rd has joined #openstack-sahara02:40
*** Networkn3rd has quit IRC03:09
*** akuznetsov has joined #openstack-sahara04:58
*** Longgeek has joined #openstack-sahara05:15
*** akuznetsov has quit IRC06:14
*** k4n0 has joined #openstack-sahara06:16
*** akuznetsov has joined #openstack-sahara06:18
*** Longgeek has quit IRC07:26
*** Longgeek has joined #openstack-sahara07:28
*** akuznetsov has quit IRC07:41
*** skostiuchenko1 has left #openstack-sahara07:43
*** akuznetsov has joined #openstack-sahara07:54
*** IvanBerezovskiy has joined #openstack-sahara07:55
*** akuznetsov has quit IRC07:58
*** witlessb has joined #openstack-sahara08:02
*** akuznetsov has joined #openstack-sahara08:07
*** sreshetnyak has quit IRC09:12
*** sreshetnyak has joined #openstack-sahara09:13
*** tosky has joined #openstack-sahara09:42
*** Timotey has joined #openstack-sahara09:56
*** akuznetsov has quit IRC10:25
*** RockKuo_Office has quit IRC11:42
*** akuznetsov has joined #openstack-sahara11:55
toskyhi, I see a problem with the ambari repository of hortonworks used by DIB, so http://public-repo-1.hortonworks.com/ambari/centos6/1.x/GA/ambari.repo12:21
tosky$ wget http://public-repo-1.hortonworks.com/ambari/centos6/1.x/updates/repodata/repomd.xml12:21
tosky2014-07-21 14:18:25 ERROR 404: Not Found.12:21
toskydo you see it as well?12:21
*** k4n0 has quit IRC12:28
*** _elmiko is now known as elmiko12:44
mattf$ curl http://public-repo-1.hortonworks.com/ambari/centos6/1.x/updates/repodata/repomd.xml12:46
mattf<?xml version="1.0" encoding="UTF-8"?>12:46
mattf<Error><Code>NoSuchKey</Code><Message>The specified key does not exist.</Message><Key>ambari/centos6/1.x/updates/repodata/repomd.xml</Key><RequestId>27F87D64A35F456D</RequestId><HostId>ZDWvxH4HDMLiiPifd/QuQI4jCQlUBUm5Uide8Uq6ub9jSm/qeS9IDAfcajb65N+s</HostId></Error>%12:46
mattftosky, same here12:47
mattfhttp://public-repo-1.hortonworks.com/ambari -> 40412:47
mattfi don't see any hwx folks online to ask. i suggest dropping them an email12:48
toskymattf: Bob Nettleton and Erik Bergenholtz, right?12:50
elmikoyea12:51
elmikoRobLevas also works with Bob12:51
toskyoh, I don't have RobLevas' email12:52
elmikoi could hunt it up, but Bob and Erik should be sufficient12:52
toskyoki, thanks12:53
*** Networkn3rd has joined #openstack-sahara12:53
*** crobertsrh has joined #openstack-sahara13:02
*** sballe has joined #openstack-sahara13:54
*** tosky has quit IRC13:55
*** tmckay has joined #openstack-sahara14:04
*** Longgeek has quit IRC14:34
*** xchance has joined #openstack-sahara14:55
*** openstackgerrit has joined #openstack-sahara15:02
*** shakamunyi has joined #openstack-sahara15:54
*** ViswaV_ has joined #openstack-sahara15:58
*** ViswaV has quit IRC16:00
*** tosky has joined #openstack-sahara16:01
*** IvanBerezovskiy has left #openstack-sahara16:04
*** mattf is now known as _mattf16:12
openstackgerritOpenStack Proposal Bot proposed a change to openstack/python-saharaclient: Updated from global requirements  https://review.openstack.org/10659316:28
openstackgerritSergey Reshetnyak proposed a change to openstack/sahara: Add support testing mr job without log checking  https://review.openstack.org/10312916:29
*** Networkn3rd has quit IRC16:36
*** bradd1 has joined #openstack-sahara16:49
*** Networkn3rd has joined #openstack-sahara16:58
*** xchance has quit IRC17:01
*** ViswaV has joined #openstack-sahara17:03
*** ViswaV_ has quit IRC17:06
*** ViswaV_ has joined #openstack-sahara17:09
*** ViswaV has quit IRC17:12
*** _mattf is now known as mattf17:16
*** bandarji has joined #openstack-sahara17:41
*** shakamunyi_ has joined #openstack-sahara17:45
*** shakamunyi has quit IRC17:49
*** Networkn3rd has quit IRC18:10
*** Networkn3rd has joined #openstack-sahara18:27
*** bandarji has quit IRC18:35
*** bandarji has joined #openstack-sahara18:47
*** bandarji has quit IRC19:06
*** tosky has quit IRC19:22
*** ViswaV_ has quit IRC19:25
*** ViswaV has joined #openstack-sahara19:40
tellesnobregahey, do the numbers in sahara-image-elements mean something specific? for what i saw i will need to create a storm folder and a zookeeper one, i want to make sure about the numbers to get them right19:42
elmikotellesnobrega: you mean for the elements?19:43
tellesnobregayes19:43
elmikothey refer to the order in which the scripts are applied, iirc19:44
elmikosimilar to the way init scripts are numbered on some linux distros19:44
tellesnobregahuum19:46
elmikotellesnobrega: have you looked at the documentation for diskimage-builder?19:48
tellesnobregaim looking at the code19:49
elmikohttps://github.com/openstack/diskimage-builder check that out, there is some documentation about writing elements in there19:49
elmikomight be useful19:49
tellesnobregacool19:50
tellesnobregathanks19:50
elmikoand i think the numbering only matters for scripts within an element, not between elements.19:50
*** ViswaV has quit IRC20:02
*** bandarji has joined #openstack-sahara20:07
crobertsrhtmckay or anyone:  Any thoughts on getting more useable info (like a job name/cluster name) in the Job Executions object?  I ask because of https://bugs.launchpad.net/horizon/+bug/1342293  (the one where we want to get rid of the uuid on the job executions table).20:07
elmikocrobertsrh: is any of that "useful" info stored in JobExecution.info or JobExecution.job_configs?20:09
crobertsrhI didn't see it off hand20:10
elmikoit doesn't look like there is a quick way to go from Job to JobExecution, data-wise20:10
crobertsrhThe job name happens to be in appPath (or something like that), but that would probably not be nice to parse and use20:10
elmikoagreed20:10
elmikowe almost need a JobConnector model to help associate Jobs with Executions20:11
tmckaycrobertsh, hey20:12
crobertsrhThe JobId is in the execution, but not enough info for the UI to display20:12
tmckaymissed the bing20:12
crobertsrhnp20:12
tmckayyou can get the Job info from the Job Id in the execution20:12
elmikocrobertsrh: agreed20:12
tmckaypull the id, then request the Job, then get the stuff out20:12
elmikothat makes way too much sense20:13
crobertsrhRight, I was hoping to avoid making an extra job_info call for each job_execution at render time20:13
crobertsrhSuch a thing was deemed "not a great pattern" by the horizon people20:14
tmckaywhat the double call?20:14
tmckayI suppose Sahara could do it for you20:14
crobertsrhThat would be ideal20:15
tmckayas long as the REST police don't get upset about it20:15
tmckayI could see injecting the name of the job into the job execution on return20:15
crobertsrhI couldn't find any other place in the UI that did anything like what I was hoping to do20:15
tmckayit doesn't need to be added to the database20:15
crobertsrhthat lead me to believe it was morally wrong from the horizon perspective20:15
crobertsrhRight, I don't think it should be in the db20:16
elmikolol, morally wrong...20:17
tmckaywould we add another column to the page?20:17
crobertsrhAdd this new column, the uuid would go away20:17
tmckaywell, what if I submit SparkPi 10 times20:18
crobertsrhI could take job name + start time and use it as the "name"20:18
crobertsrhor job name + start time + cluster name (which we would also need)20:18
tmckayIf I click on the name, in the details I get the id, yes?20:18
crobertsrhYeah, id would still show up there20:18
crobertsrhI think Job Executions is the only place where uuid shows up at top level20:19
tmckayIt would be easy to add extra fields I think to the object while it's being rendered on the return20:19
tmckayThe only question is whether or not that would break anything, having extra fields in there20:19
crobertsrhOr....just give the Job Execution an actual name20:19
elmikonot the Job's name, but their own unique name?20:20
crobertsrhYeah, some unique name, "somewhat readable" for each execution20:20
openstackgerritA change was merged to openstack/sahara-specs: Add Cloudera Distribution of Hadoop in Sahara  https://review.openstack.org/10489820:21
elmikoi like the idea of "job name + start time + cluster"20:21
tmckayThe job name, cluster, and start time ought to be pretty good for identification20:21
crobertsrhI would think so20:21
tmckayheh, execution has cluster id too but no cluster name20:26
crobertsrhYes :)20:26
tmckayI know it's possible to mess with these things on the way out, let me take a look.20:27
*** Networkn3rd has quit IRC20:29
*** ViswaV has joined #openstack-sahara20:30
*** ViswaV_ has joined #openstack-sahara20:31
tmckaycroberts, we have something that allows filtering of fields on return but not something that allows adding info20:34
tmckaynot sure of the best way to do that20:34
*** ViswaV has quit IRC20:35
*** ViswaV_ has quit IRC20:44
*** ViswaV has joined #openstack-sahara20:55
*** crobertsrh is now known as _crobertsrh20:58
*** mattf is now known as _mattf20:59
*** ViswaV has quit IRC21:12
*** elmiko is now known as _elmiko21:20
*** tmckay has quit IRC21:23
*** shakamunyi_ has quit IRC21:24
*** RobLevas has joined #openstack-sahara21:38
*** tmckay has joined #openstack-sahara21:49
*** bandarji has quit IRC22:05
*** bradd1 has quit IRC22:16
*** ViswaV has joined #openstack-sahara22:33
*** ViswaV_ has joined #openstack-sahara22:34
*** ViswaV has quit IRC22:37
*** bradd1 has joined #openstack-sahara22:52
*** shakamunyi_ has joined #openstack-sahara22:53
*** witlessb has quit IRC23:02
*** bradd1 has quit IRC23:24
*** shakamunyi_ has quit IRC23:33
*** ViswaV_ has quit IRC23:58
*** ViswaV has joined #openstack-sahara23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!