| *** alazarev has quit IRC | 00:10 | |
| *** alazarev has joined #openstack-sahara | 00:12 | |
| *** alazarev has joined #openstack-sahara | 00:23 | |
| *** zhiyan_ is now known as zhiyan | 00:27 | |
| *** _elmiko has quit IRC | 00:31 | |
| *** _elmiko has joined #openstack-sahara | 00:43 | |
| *** alazarev has quit IRC | 00:52 | |
| *** alazarev has joined #openstack-sahara | 03:13 | |
| *** alazarev has quit IRC | 03:24 | |
| *** zhiyan is now known as zhiyan_ | 03:30 | |
| *** akuznetsov has joined #openstack-sahara | 03:30 | |
| *** zhiyan_ is now known as zhiyan | 03:41 | |
| *** alazarev has joined #openstack-sahara | 03:49 | |
| *** akuznetsov has quit IRC | 04:09 | |
| *** akuznetsov has joined #openstack-sahara | 04:20 | |
| openstackgerrit | A change was merged to openstack/sahara: Opening stable/icehouse for next stable release https://review.openstack.org/99059 | 04:52 |
|---|---|---|
| openstackgerrit | A change was merged to openstack/sahara: Use in-memory sqlite DB for unit tests https://review.openstack.org/98923 | 04:52 |
| *** alazarev has quit IRC | 04:54 | |
| *** alazarev has joined #openstack-sahara | 04:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/sahara: Updated from global requirements https://review.openstack.org/99096 | 05:00 |
| *** alazarev has quit IRC | 05:34 | |
| *** akuznetsov has quit IRC | 05:40 | |
| *** alazarev has joined #openstack-sahara | 06:02 | |
| *** akuznetsov has joined #openstack-sahara | 06:11 | |
| *** alazarev has quit IRC | 06:18 | |
| *** akuznetsov has quit IRC | 06:25 | |
| *** akuznetsov has joined #openstack-sahara | 06:48 | |
| *** julienvey has joined #openstack-sahara | 07:01 | |
| *** akuznetsov has quit IRC | 07:02 | |
| *** akuznetsov has joined #openstack-sahara | 07:15 | |
| *** witlessb has joined #openstack-sahara | 07:22 | |
| *** sreshetnyak has quit IRC | 08:16 | |
| *** vrovachev has quit IRC | 08:16 | |
| *** julienvey has quit IRC | 08:35 | |
| *** julienvey has joined #openstack-sahara | 08:35 | |
| openstackgerrit | Daniele Venzano proposed a change to openstack/sahara: Implement scaling for Spark clusters https://review.openstack.org/99295 | 09:03 |
| *** zhiyan is now known as zhiyan_ | 09:23 | |
| *** tosky has joined #openstack-sahara | 09:36 | |
| *** julienve_ has joined #openstack-sahara | 09:39 | |
| *** julienvey has quit IRC | 09:43 | |
| *** sreshetnyak has joined #openstack-sahara | 10:13 | |
| *** akuznetsov has quit IRC | 10:21 | |
| *** witlessb has quit IRC | 10:30 | |
| *** witlessb has joined #openstack-sahara | 10:31 | |
| *** themistymay has joined #openstack-sahara | 10:57 | |
| SergeyLukjanov | I've canceled todays irc meeting due to the holiday in russia | 11:01 |
| SergeyLukjanov | mattf, tmckay, croberts ^^ | 11:01 |
| SergeyLukjanov | please, note, that we'll be accessible starting from Monday | 11:02 |
| themistymay | Have a good one in Russia! | 11:21 |
| *** shakayumi has joined #openstack-sahara | 11:24 | |
| *** julienve_ has quit IRC | 11:35 | |
| mattf | SergeyLukjanov et al, have a good long weekend! | 11:35 |
| *** julienvey has joined #openstack-sahara | 11:36 | |
| SergeyLukjanov | themistymay, mattf, thanks folks | 11:36 |
| *** julienve_ has joined #openstack-sahara | 11:38 | |
| *** julienvey has quit IRC | 11:41 | |
| *** shakayumi has quit IRC | 11:50 | |
| themistymay | SergeyLukjanov, mattf: of course the week I am back from business trips is the one when we cancel the meeting *rage rage rage* | 12:18 |
| *** akuznetsov has joined #openstack-sahara | 12:38 | |
| mattf | almost like it was planned | 12:41 |
| *** openstackgerrit has quit IRC | 12:46 | |
| *** openstackgerrit has joined #openstack-sahara | 12:47 | |
| *** _elmiko is now known as elmiko | 12:52 | |
| *** julienve_ has quit IRC | 12:55 | |
| *** julienvey has joined #openstack-sahara | 12:56 | |
| *** themistymay has quit IRC | 13:10 | |
| *** tmckay has joined #openstack-sahara | 13:11 | |
| *** julienve_ has joined #openstack-sahara | 13:36 | |
| *** julienvey has quit IRC | 13:38 | |
| *** bradd1 has joined #openstack-sahara | 13:40 | |
| *** tellesnobrega has quit IRC | 13:49 | |
| *** tellesnobrega has joined #openstack-sahara | 14:01 | |
| *** tellesnobrega has quit IRC | 14:08 | |
| *** tellesnobrega has joined #openstack-sahara | 14:12 | |
| *** bradd11 has joined #openstack-sahara | 14:16 | |
| *** bradd1 has quit IRC | 14:18 | |
| *** bradd11 has quit IRC | 14:23 | |
| *** mattf is now known as _mattf | 14:23 | |
| *** bradd1 has joined #openstack-sahara | 14:31 | |
| *** _mattf is now known as mattf | 14:56 | |
| *** akuznetsov has quit IRC | 15:11 | |
| SergeyLukjanov | themistymay, more than that, we have a holidays in russia special for it ;) | 15:14 |
| *** akuznetsov has joined #openstack-sahara | 15:19 | |
| *** bandarji has joined #openstack-sahara | 15:24 | |
| *** bandarji has quit IRC | 15:24 | |
| *** mattf is now known as _mattf | 15:28 | |
| *** _mattf is now known as mattf | 15:29 | |
| *** bogdando has quit IRC | 15:55 | |
| *** bogdando has joined #openstack-sahara | 15:56 | |
| *** bandarji has joined #openstack-sahara | 15:59 | |
| *** akuznetsov has quit IRC | 16:03 | |
| *** elmiko is now known as _elmiko | 16:17 | |
| *** themistymay has joined #openstack-sahara | 16:21 | |
| *** bandarji has quit IRC | 16:32 | |
| *** akuznetsov has joined #openstack-sahara | 16:34 | |
| *** openstackgerrit has quit IRC | 16:38 | |
| *** bandarji has joined #openstack-sahara | 16:38 | |
| *** bogdando has quit IRC | 17:00 | |
| *** alazarev has joined #openstack-sahara | 17:11 | |
| *** themistymay has quit IRC | 17:12 | |
| *** _elmiko is now known as elmiko | 17:14 | |
| *** tosky has quit IRC | 17:14 | |
| *** mattf is now known as _mattf | 17:31 | |
| *** _mattf is now known as mattf | 17:32 | |
| *** themistymay has joined #openstack-sahara | 17:47 | |
| *** themistymay has quit IRC | 17:58 | |
| *** tosky has joined #openstack-sahara | 18:02 | |
| alazarev | It looks like meeting was canceled bacause of national holiday in Russia | 18:12 |
| *** openstackgerrit has joined #openstack-sahara | 18:12 | |
| elmiko | alazarev: yes | 18:13 |
| elmiko | national no meeting day :) | 18:13 |
| *** alazarev has quit IRC | 18:17 | |
| tmckay | SergeyLukjanov, ping. Happen to be around? | 18:24 |
| tmckay | mattf, ping too | 18:24 |
| *** alazarev has joined #openstack-sahara | 18:29 | |
| *** themistymay has joined #openstack-sahara | 18:35 | |
| *** bandarji has quit IRC | 18:35 | |
| *** themistymay has quit IRC | 18:37 | |
| tmckay | hey, anyone care that delete doesn't actually stop a job? | 18:37 |
| tmckay | because it doesn't | 18:37 |
| tmckay | anyone want it fixed in juno1, or is juno2 good enough? | 18:38 |
| tmckay | I think I'm talking to myself. | 18:38 |
| * tmckay verifies on latest trunk | 18:38 | |
| tosky | tmckay: and maybe backport? | 18:45 |
| alazarev | juno1 is today, so no chance for juno1 | 18:46 |
| alazarev | and… should it stop job? | 18:46 |
| tmckay | yes. I can't believe we missed this. No test anywhere actually checks the oozie client for job status after a delete, so we never noticed | 18:46 |
| tmckay | that's the question | 18:46 |
| tmckay | currently, there is no way to stop a job | 18:46 |
| tmckay | unless you use the REST API yourself from the linux command line | 18:46 |
| tmckay | so, delete can cancel-and-delete, or the client can implement cancel and the UI can call both, or ? | 18:47 |
| tmckay | probably needs some discussion | 18:47 |
| alazarev | we should revise job lifecycle | 18:48 |
| tmckay | I ran into this playing with refactoring, realized the cancel is never actually being called, and then that the client doesn't even implement it | 18:48 |
| alazarev | because deletion of cluster does stracnge things too: https://bugs.launchpad.net/sahara/+bug/1329109 | 18:49 |
| alazarev | and yesterday I realized that external hdfs feature even doesn’t compile (python syntax error) | 18:50 |
| tmckay | ooo | 18:51 |
| alazarev | regarding job excecution deletion: I could emagine both scenarious | 18:51 |
| alazarev | when user wants to just remove reference from sahara and whan he wants to completely cancel job | 18:52 |
| alazarev | we have undocumented ‘cancel job’ API, right? | 18:53 |
| tmckay | Hmm, I think it's documented. Let me check the REST api. But, it's not exposed in the client, and mattf's V2 CR doesn't list it | 18:54 |
| tmckay | alazarev, yes, cancel is in section 6.4 | 18:55 |
| alazarev | I was against exclusion it from v2, I as I remember we decided that more discussion needed | 18:55 |
| tmckay | So, there is technically a workaround. Someone can cancel with curl or httpie | 18:55 |
| tmckay | but not from the web UI or client. :-( | 18:56 |
| tmckay | I just created a bug, feel free to add comments | 18:56 |
| alazarev | we can expose it to client and wen UI | 18:56 |
| alazarev | *web | 18:56 |
| alazarev | I think it is useful function | 18:57 |
| alazarev | and what do you think cluster deletion should do? E.g. there was running job. User deletes cluster. What should happen with job? | 18:57 |
| tmckay | Yes, sometimes I want a job to go away because I releaize that I made a mistake | 18:57 |
| tmckay | I would think "cancel all jobs" then terminate the cluster. | 18:58 |
| alazarev | what do you mean by ‘cancel’? send ‘kill’ request to oozie? just mark all jobs as killed? | 18:59 |
| tmckay | alazarev, I can imagine cancel and delete both as well. I launch a job, realize I got the arguments wrong, cancel the job run, and use relaunch to run again. If cancel-and-delete is one action, then the UI can't really do relaunch. | 18:59 |
| tmckay | send kill request to oozie | 19:00 |
| tmckay | but, if you delete the cluster, maybe that doesn't matter | 19:00 |
| tmckay | It always seems cleaner to me to do ordered shutdown. | 19:00 |
| tmckay | But, maybe in this case making the cluster disappear is fine. Jobs go away by definiition. The execution records could stay in Sahara, as they do now for completed or failed jobs. | 19:01 |
| tmckay | The cancel/delete semantic influences all this, I think. | 19:02 |
| tmckay | If an execution record remains after a job runs, then it should always remain, whether or not the job succeeded, was killed by an error, was canceled by hand, or was ended because the cluster went away | 19:03 |
| alazarev | agree, for cluster deletetion my suggestion is to just move to KILLED | 19:05 |
| tmckay | I think that work. It might be in the middle of writing data, but so what? | 19:06 |
| alazarev | I thought a little more about deletion of job excecution… and I would say we need both - canceling during deletion and just canceling | 19:06 |
| tmckay | I think so, if we're going to have relaunch in the UI. Otherwise, the UI has to synthesize relaunch. Remember on it's own all the details. | 19:07 |
| tmckay | I like relaunch, I use it all the time when I'm playing | 19:08 |
| *** alazarev has quit IRC | 19:17 | |
| *** ViswaV has joined #openstack-sahara | 19:22 | |
| *** bandarji has joined #openstack-sahara | 19:37 | |
| *** elmiko_ has joined #openstack-sahara | 19:38 | |
| *** akuznetsov has quit IRC | 19:56 | |
| *** julienve_ has quit IRC | 20:08 | |
| *** boris-42 has quit IRC | 20:18 | |
| *** boris-42 has joined #openstack-sahara | 20:18 | |
| *** alazarev has joined #openstack-sahara | 20:35 | |
| *** openstackgerrit has quit IRC | 20:35 | |
| *** openstackgerrit has joined #openstack-sahara | 20:35 | |
| *** julienvey has joined #openstack-sahara | 20:38 | |
| *** julienvey has quit IRC | 20:40 | |
| *** julienvey has joined #openstack-sahara | 20:40 | |
| *** julienvey has quit IRC | 20:45 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/sahara: Updated from global requirements https://review.openstack.org/99096 | 21:08 |
| *** julienvey has joined #openstack-sahara | 21:15 | |
| *** mattf is now known as _mattf | 21:24 | |
| *** _mattf is now known as mattf | 21:26 | |
| *** julienvey has quit IRC | 21:40 | |
| *** julienvey has joined #openstack-sahara | 22:04 | |
| *** julienvey has quit IRC | 22:18 | |
| *** ViswaV has quit IRC | 22:20 | |
| *** ViswaV has joined #openstack-sahara | 22:22 | |
| *** bandarji has quit IRC | 22:35 | |
| *** elmiko is now known as _elmiko | 22:40 | |
| openstackgerrit | Andrew Lazarev proposed a change to openstack/sahara: Update job execution status on cluster deletion https://review.openstack.org/99788 | 22:46 |
| openstackgerrit | Andrew Lazarev proposed a change to openstack/sahara: Fixed status update for job execution https://review.openstack.org/99790 | 22:52 |
| alazarev | does anyone knows why we have job_execution.info.status structure instead of obvious job_execution.status? | 22:53 |
| *** bandarji has joined #openstack-sahara | 23:03 | |
| *** bandarji1 has joined #openstack-sahara | 23:09 | |
| *** bandarji has quit IRC | 23:09 | |
| *** alazarev has quit IRC | 23:14 | |
| *** ViswaV has quit IRC | 23:23 | |
| *** alazarev has joined #openstack-sahara | 23:23 | |
| *** tosky has quit IRC | 23:40 | |
| openstackgerrit | Andrew Lazarev proposed a change to openstack/sahara-dashboard: Added start and end times to job execution details page https://review.openstack.org/99804 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!