Wednesday, 2014-08-27

« Tuesday, 2014-08-26 Index Thursday, 2014-08-28 »
*** yograterol has quit IRC00:15
*** vkmc has quit IRC00:41
*** alcabrera|afk is now known as alcabrera01:32
*** staphae has quit IRC02:02
*** chandankumar has joined #openstack-zaqar02:34
*** chandankumar has quit IRC02:34
*** yograterol has joined #openstack-zaqar02:40
*** echevemaster_ has quit IRC02:45
*** echevemaster has joined #openstack-zaqar02:45
*** alcabrera is now known as alcabrera|afk02:58
*** AAzza_afk has joined #openstack-zaqar06:04
*** AAzza_afk is now known as AAzza06:04
*** yograterol has quit IRC06:09
*** chandankumar has joined #openstack-zaqar06:14
*** flaper87|afk is now known as flaper8706:15
flaper87echevemaster: hey man, have you had the time to do the package renaming thing?06:16
echevemasterflaper87: I'm doing it. but still I'm not finished it, why?06:19
flaper87echevemaster: just keeping track of things, no preasures at all06:19
echevemasterno problem man, I hope finish it this week, count on it :) flaper8706:20
*** AAzza is now known as AAzza_afk06:46
flaper87flwang: ping07:18
flaper87flwang: https://review.openstack.org/#/c/106346/ pls, again07:18
flaper87I had to address some minor nits in the comments07:18
*** AAzza_afk is now known as AAzza07:29
*** AAzzal has joined #openstack-zaqar07:52
*** AAzza has quit IRC07:54
*** AAzzal is now known as AAzza07:54
*** dguerri_ has joined #openstack-zaqar08:01
*** dguerri has quit IRC08:10
*** wpf has quit IRC08:10
AAzzaflaper87: ping08:12
flaper87AAzza: pong08:12
AAzzai have several concerns/problems with this api change: "Message href's now always include claim_id in the query string if the message is claimed"08:13
flaper87AAzza: shoot08:13
AAzzafirst the on_get on message instance controller doesn't handle the claim_id from params. so the check should be added here to verify that it is correspond the real claim_id?08:14
AAzzai saw that line in wiki in the get message section -  "If either the message ID or the claim ID is malformed or nonexistent, no message is returned. "08:15
flaper87ah damn, I've to run. Will you be around later today?08:16
AAzzayeah, will be)08:16
*** wpf has joined #openstack-zaqar08:16
AAzzafor several hours now08:16
AAzzaand in the evening)08:16
flaper87aewsome, that's great. I'd also like to discuss this with kgriffs08:16
flaper87sorry for dropping off08:17
AAzzaflaper87: np, run-run)08:17
*** flaper87 is now known as flaper87|afk08:17
*** flwang1 has joined #openstack-zaqar08:50
*** AAzza_ has joined #openstack-zaqar08:58
*** AAzza has quit IRC09:00
*** AAzza has joined #openstack-zaqar09:00
AAzzaflaper87|afk: created etherpad to track discussion about v1.1 response changes https://etherpad.openstack.org/p/zaqar-api-v1-1-response-changes09:00
*** AAzza_ has quit IRC09:04
*** echevemaster has quit IRC09:24
*** chandankumar has quit IRC09:32
*** chandankumar has joined #openstack-zaqar09:40
*** AAzza is now known as AAzza_afk10:17
*** chandankumar has quit IRC10:26
*** chandankumar has joined #openstack-zaqar10:27
*** flwang1 has quit IRC10:34
*** AAzza_afk has quit IRC10:35
*** flwang1 has joined #openstack-zaqar10:35
*** AAzza_afk has joined #openstack-zaqar10:35
*** AAzza_afk is now known as AAzza10:35
*** AAzza has quit IRC10:39
*** AAzza_afk has joined #openstack-zaqar10:40
*** AAzza_afk is now known as AAzza10:40
*** flaper87|afk is now known as flaper8710:47
*** flaper87 is now known as flaper87|afk11:14
*** flaper87|afk is now known as flaper8711:23
*** flwang1 has quit IRC11:34
*** flwang2 has joined #openstack-zaqar11:35
*** flaper87 is now known as flaper87|afk11:42
*** flaper87|afk is now known as flaper8711:44
*** vkmc has joined #openstack-zaqar12:09
vkmcmorning!12:10
*** flaper87 is now known as flaper87|afk12:14
*** flaper87|afk is now known as flaper8712:18
*** chandankumar has quit IRC12:21
*** flwang2 has quit IRC12:37
*** sriram has joined #openstack-zaqar12:41
vkmcsriram, o/12:52
flaper87heeeeeeeeeeeeeeeeeeeeeeeeeeeeeeey12:52
flaper87vkmc: how are you doing?12:52
vkmcflaper87, yooooooo o/12:52
sriramhello there! :)12:52
vkmcflaper87, fine and you? :)12:52
flaper87vkmc: allllll goood12:52
flaper87vkmc: one thing, before I forget. I'm moving all wiki pages that have (Zaqar) in the name under wiki/Zaqar/$NAME12:53
vkmcflaper87, awesome! can I give you a hand with that?12:53
flaper87the whole (Zaqar) thing is aweful and it doesn't make sense for it to exist outside the Zaqar section in the wiki12:53
flaper87vkmc: if you happen to open one, just move it12:53
vkmcI dunno if there is an automagical way to do renaming in the wiki12:54
flaper87vkmc: that's what I'm doing12:54
flaper87There's not :(12:54
vkmcok, so let's divide the load a bit12:54
flaper87there's not even a way to keep yourself logged in12:54
* flaper87 kicks wiki's ass12:54
vkmcdo some renaming backtracking12:54
vkmcyeah that's a bummer :@12:54
flaper87vkmc: it's not urgent, do it if you happen to see one12:55
vkmck12:55
flaper87vkmc: also, if you can review this patch, it'd be great: https://review.openstack.org/#/c/112009/12:57
flaper87I'm updating the wiki as we speak12:57
flaper87damn, can I say something really really bad?12:58
flaper87I hate doing this12:58
flaper87yes yes, I know, it's a bad example12:59
flaper87..>12:59
flaper87>.>12:59
vkmcsure I'll check that too13:02
vkmcnerdy alert... you probably already know about it, but yesterday I was reading about hybrid interpreters and I get to know the dis Python library13:03
vkmchttps://docs.python.org/2/library/dis.html13:03
vkmcsuper cool13:03
flaper87ah yeah, I've played with dis a bit, nothing serious though13:05
vkmcdoes it have a real use... apart from knowing how the python interpreter works?13:06
*** mpanetta has joined #openstack-zaqar13:12
flaper87vkmc: well, it can certainly be used in attackts13:17
flaper87by analysing the bytecode you know what the program is doing and you can definitely get useful informations out of it13:17
flaper87so, I'd say, it's lot to do with reverse-engineering13:18
* flaper87 documented flavor's PUT. GET, DELETE, LIST missing13:18
*** AAzza is now known as AAzza_afk13:19
vkmc+1 :)13:20
flaper87DOOOOOOOOOOOOOONE https://wiki.openstack.org/wiki/Zaqar/specs/api/v1.1#Queue_Flavors13:24
*** mpanetta_ has joined #openstack-zaqar13:25
*** mpanetta has quit IRC13:25
vkmc\o/13:25
flaper87Bonus, I realized that: https://bugs.launchpad.net/zaqar/+bug/136217413:26
vkmcflaper87, why that?13:27
flaper87vkmc: what?13:27
flaper87I mean, why do we need to raise an error?13:28
vkmcwhy you cannot delete a pool if it's used by some flavor13:28
vkmcI'd thought that the other way round13:28
vkmcthat is... you cannot delete a flavor is a pool is using it13:28
flaper87oh but pools don't use flavors. Flavors use pools13:29
flaper87In fact, there's a pool field in the flavor13:29
vkmctrue that13:29
*** AAzza_afk has quit IRC13:29
flaper87Thing is that, if you delete a pool and then a user tries to use that flavor, then there won't be a pool to store that queue in13:29
flaper87there a couple of ways around this without creating a hard constraint13:30
*** AAzza_afk has joined #openstack-zaqar13:30
*** AAzza_afk is now known as AAzza13:30
flaper87however, since this is an "admin only" operation, I think it'd be better to teach the admin the right workflow to do this13:30
flaper87Does that make sense?13:30
flaper87My forbidding pool deletion when a flavor is using it, we're telling the admin that flavors should be first migrated to other pools13:31
flaper87I think we'll have to add a "status" field to the flavor13:31
flaper87or something to know whether it's still valid or not13:31
flaper87For example, and admin may want to disable a flavor because they want to dismiss a storage pool13:32
vkmcit does yeah13:32
flaper87In that case, the admin would set the flavor as disabled, then migrate everything13:32
flaper87or the admin could also create a new pool, migrate pools and then re-enable the flavor13:32
flaper87Things to do during Juno13:33
flaper87erm, Kilo13:33
flaper87:P13:33
vkmccool :) I got confused for a moment13:34
*** itisit has joined #openstack-zaqar13:36
*** malini has joined #openstack-zaqar13:37
* flaper87 gives vkmc an de-confussing glass of coffee13:39
flaper87glass because the cup is too small13:39
vkmcthanks for clearing it up :)13:40
* vkmc takes a bit sip13:40
flaper87vkmc: now go and review my patch13:40
flaper87oh, fucking jenkins13:40
vkmcI'd need more than coffee for that13:42
vkmclet's say... a gummybear13:42
*** AAzza has quit IRC13:48
*** AAzza_afk has joined #openstack-zaqar13:49
*** AAzza_afk is now known as AAzza13:49
*** chandankumar has joined #openstack-zaqar13:50
*** cpallares has joined #openstack-zaqar14:03
vkmcflaper87, does it make sense to add a test controlling that it's not possible to add one queue with two flavors?14:18
vkmcI mean, you register a queue with a flavor, and then you register the same queue with another flavor14:18
vkmcor that cannot happen?14:18
*** jeffrey4l has joined #openstack-zaqar14:19
*** chandankumar has quit IRC14:24
flaper87vkmc: I believe that cannot happen14:38
flaper87you can't have 2 `flavor` keys in the queue metadata14:38
flaper87and once you create the queue, you can't 're-create' it14:38
flaper87There's no queue migration yet, so updating the queue flavor won't work yet14:38
*** amitgandhinz has joined #openstack-zaqar14:38
*** chandankumar has joined #openstack-zaqar14:41
vkmcflaper87, awesome14:44
*** kgriffs has joined #openstack-zaqar14:48
kgriffsthe IRC server I was connected to is having issues today14:49
*** malini|afk has joined #openstack-zaqar14:49
kgriffsreconnected and now I see everyone again.14:49
*** alcabrera|afk is now known as alcabrera14:53
vkmchi kgriffs, alcabrera o/14:53
kgriffso/14:53
alcabreraheya vkmc. :)14:53
kgriffsmalini: I fixed ZNC so your account is using a good IRC server now too (I think)14:54
malinithanks kgriffs! i'll try tht14:54
*** openstackgerrit has joined #openstack-zaqar14:54
*** flaper87 is now known as flaper87|afk14:56
*** flaper87|afk is now known as flaper8714:56
vkmcbbl14:56
openstackgerritCindy Pallares proposed a change to openstack/zaqar: First patch and first review docs  https://review.openstack.org/11536914:57
*** yograterol has joined #openstack-zaqar15:03
kgriffsflaper87: ping15:04
flaper87kgriffs: pong15:04
kgriffsso, I wanted to see if I understood the way policies work in OpenStack15:04
kgriffsThe policies are defined in JSON files on the keystone server, right?15:05
kgriffsor are they defined on each individual service?15:05
* kgriffs is ashamed to not know this15:05
jeffrey4lthey are define in each individual service15:06
jeffrey4le.g nova/etc/policy.json15:06
kgriffskk15:08
flaper87kgriffs: policies are defined in each service15:18
flaper87ah, what jeffrey4l said15:18
kgriffsgot it. that's what I thought but then I was reading this and it made me wonder if I had that right. http://docs.openstack.org/admin-guide-cloud/content/keystone-user-management.html15:18
kgriffsthe docs aren't really clear on where policies live15:18
kgriffsbut yeah, I didn't recall a Keystone API for getting them, and wanted to be sure15:18
* kgriffs second-guesses himself too often these days15:18
*** jeffrey4l has quit IRC15:18
AAzzakgriffs: flaper87: I have several questions about api-v1-1-response changes? can we discuss it today?15:18
AAzzatoday means anytime during next 3-4 hours :)15:18
kgriffsAAzza: what's up?15:19
*** dguerri_ has quit IRC15:22
*** peoplemerge has quit IRC15:22
AAzzakgriffs: I have etherpad of thoughs https://etherpad.openstack.org/p/zaqar-api-v1-1-response-changes15:22
AAzzakgriffs: most actual is the first one - about the claim_id in the message href15:22
*** dguerri_ has joined #openstack-zaqar15:22
*** peoplemerge has joined #openstack-zaqar15:22
AAzzakgriffs: the most important problem, i'm not sure how to do it without changes to storage driver. maybe there is a way, i don't see15:23
kgriffslooking15:23
*** jeffrey4l has joined #openstack-zaqar15:26
*** amitgandhinz has quit IRC15:27
flaper87kgriffs: whenever you've time, could you take a look here? https://review.openstack.org/#/c/112009/15:27
*** amitgandhinz has joined #openstack-zaqar15:27
*** alcabrera is now known as alcabrera|afk15:27
*** amitgandhinz has quit IRC15:30
kgriffsAAzza: ok, do those comments make sense so far? we can continue discussion here in IRC15:30
*** amitgandhinz has joined #openstack-zaqar15:30
AAzzakgriffs: yes, lets continue there :)15:30
AAzzakgriffs: so there is no reason to do this change?15:31
AAzzakgriffs: if there is indifferent for GET?15:31
AAzzakgriffs: or15:32
AAzzakgriffs: I'm stupid15:32
kgriffsWell, it may make sense to ensure that only the thing that claimed the message can delete it15:32
AAzzakgriffs: got what you mean)15:32
kgriffswait15:32
kgriffslet me check something15:32
*** openstack has joined #openstack-zaqar16:19
openstackgerritKurt Griffiths proposed a change to openstack/zaqar: Add observer role to benchmark tool  https://review.openstack.org/11638416:24
*** ametts has joined #openstack-zaqar16:26
*** earnThis has joined #openstack-zaqar16:57
vkmcearnThis, o/16:57
earnThisHi all16:59
vkmcearnThis, how are you?16:59
earnThisdoing well, back to annoy you with more questions haha, yourself?17:00
vkmcdoing well too :)17:00
vkmcnah you don't annoy me17:00
vkmcI saw you asked about running tests17:00
earnThisYup I got tox running but got an invocationerror after it ran for a little17:01
vkmcwell if you simply call tox you will run tests for all the environments in the tox.ini file17:03
vkmcyou can choose to run some of them, calling tox with 'tox -e py27,pep8'17:03
vkmccheck if you get errors like that too and, if so, paste me the output17:04
earnThisyup i called it with tox -e py2717:04
vkmccool17:04
*** earnThis has quit IRC17:13
*** earnThis has joined #openstack-zaqar17:24
vkmckgriffs, do you know which are the memory requirements for the tests environment?17:29
vkmckgriffs, earnThis is getting this when running tox http://paste.openstack.org/show/101111/17:29
*** echevemaster has joined #openstack-zaqar17:31
*** kgriffs is now known as kgriffs|afk17:31
cpallaresearnThis: How much memory do you have?17:33
earnThisa little over 2 G17:33
vkmclooks like ~700MB ++ did the trick17:34
cpallaresearnThis: Are you using it on devstack?17:34
earnThiscpallares, no17:35
earnThisit seems to be working though p27 just succeeded, pep8 running now17:36
vkmc\o/17:36
*** jimhoagland_ has joined #openstack-zaqar17:48
*** jimhoagland_ has left #openstack-zaqar17:49
*** kgriffs|afk is now known as kgriffs17:52
*** amitgandhinz has joined #openstack-zaqar17:53
echevemastervkmc: are you there?18:05
vkmcI am18:06
echevemasterHi, I'm packaging zaqar in Fedora, but the latest version complains of pbr, I've applied a patch in version.py, but does not work, Do you have any clue about it?18:07
vkmchm, I do not18:08
vkmccan you paste the output?18:08
vkmckgriffs, malini may know about this? ^18:08
echevemasterthis is the patch http://pkgs.fedoraproject.org/cgit/openstack-marconi.git/tree/0001-Remove-runtime-dependency-on-PBR.patch18:08
echevemaster"versioning in this project requires either an sdist tarball" is the exception.18:10
echevemasterthe previous version before renaming works fine with the patch, but this not.18:11
echevemastervkmc:18:12
vkmcwe didn't rename the Icehouse branch18:13
*** mpanetta_ is now known as mpanetta18:13
vkmcwhich I see is the version you are packaging18:13
vkmcright?18:13
echevemaster 2014.2.b2 is the version that I'm packaging18:14
echevemasterthis commit b3f1b48d0fdbf978f8a4c94f48c5f3bc41dd436518:14
vkmcthanks18:14
vkmclet me take a look into it18:15
echevemasterthanks vkmc, please if you encounter something there, please ping me or contact me in echevemaster at fedoraproject dot org18:16
echevemasterwhile I'm trying to solve.18:16
vkmcwill do18:17
vkmcthanks for working on this18:17
echevemaster:)18:17
*** kgriffs is now known as kgriffs|afk18:32
vkmcechevemaster, now that you mention, I remember hitting that same error two weeks ago18:35
vkmcand flaper87|afk suggest me to uninstall pbr, remove all pbr matches in a find command and then run pip install again18:35
vkmcthe problem is because of the renaming18:36
vkmcs/suggest/suggeste18:36
vkmcd18:36
echevemasteroh ok. pbr appears in setup.py tests* and config* (I'll remove those all matches and  check it)18:38
vkmcok18:38
vkmclet me know if that works18:38
echevemastersure.18:39
vkmcbbl18:41
*** kgriffs|afk is now known as kgriffs18:48
*** echevemaster has quit IRC19:14
*** cpallares has quit IRC19:20
*** malini has quit IRC19:44
*** malini|afk is now known as malini19:44
*** flwang1 has joined #openstack-zaqar19:57
kgriffsgoing to go offline for a bit to focus on the redis driver20:01
*** kgriffs is now known as kgriffs|afk20:02
vkmckgriffs|afk, why FlavorDoesNotExist return HTTP 503? https://review.openstack.org/#/c/112009/11/zaqar/queues/storage/pooling.py20:04
vkmchappy hacking :)20:07
*** AAzza is now known as AAzza_afk20:12
*** flwang1 has quit IRC20:12
*** malini1 has joined #openstack-zaqar20:40
*** malini1 has quit IRC20:41
*** malini1 has joined #openstack-zaqar20:41
*** sriram has quit IRC20:50
*** cpallares has joined #openstack-zaqar21:07
*** cpallares has quit IRC21:07
*** amitgandhinz has quit IRC21:53
*** amitgandhinz has joined #openstack-zaqar21:53
*** amitgandhinz has quit IRC21:58
*** mpanetta has quit IRC22:00
earnThishold on..http://paste.openstack.org/show/101255/22:00
earnThisthose are the files I want, have they been committed if they are still in git status22:01
vkmcI assume that was for me :p22:02
vkmcearnThis, &22:02
vkmc^22:02
vkmcyou can reset the health.py file with 'git reset HEAD zaqar/queues/transport/wsgi/v1_1/health.py'22:03
vkmcand then you probably can change to your bug branch22:04
*** malini1 has quit IRC22:04
*** sriram has joined #openstack-zaqar22:08
*** reed has joined #openstack-zaqar22:24
*** earnThis has quit IRC22:38
*** sriram has quit IRC22:45
*** itisit has quit IRC23:12
*** yograterol has quit IRC23:29
« Tuesday, 2014-08-26 Index Thursday, 2014-08-28 »

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