Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2014-12-22

| Channels | #openstack-rally index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:26 yingjun joined #openstack-rally
02:30 yaguang joined #openstack-rally
02:46 openstackgerrit Li Yingjun proposed stackforge/rally: Move i18n to common lib  https://review.openstack.org/143347
02:56 openstackgerrit Li Yingjun proposed stackforge/rally: Move i18n to common lib  https://review.openstack.org/143347
03:14 jjmb1 joined #openstack-rally
03:34 openstackgerrit Li Yingjun proposed stackforge/rally: Move i18n to common lib  https://review.openstack.org/143347
03:59 yingjun joined #openstack-rally
04:04 Poornima joined #openstack-rally
04:18 mitz joined #openstack-rally
05:12 rmk joined #openstack-rally
05:20 rdas joined #openstack-rally
05:24 prmtl joined #openstack-rally
05:24 nkhare joined #openstack-rally
06:30 nmagnezi joined #openstack-rally
06:37 exploreshaifali joined #openstack-rally
06:43 neeti joined #openstack-rally
06:54 openstackgerrit Li Yingjun proposed stackforge/rally: Move files to common lib(Part 1)  https://review.openstack.org/143376
07:12 openstackgerrit Li Yingjun proposed stackforge/rally: Move files to common lib(Part 1)  https://review.openstack.org/143376
07:30 k4n0 joined #openstack-rally
07:51 tkammer joined #openstack-rally
08:09 yingjun joined #openstack-rally
08:10 yingjun_ joined #openstack-rally
08:53 oanufriev joined #openstack-rally
09:07 yingjun joined #openstack-rally
09:13 exploreshaifali joined #openstack-rally
09:24 andreykurilin_ joined #openstack-rally
09:30 andreykurilin__ joined #openstack-rally
09:33 tkammer joined #openstack-rally
10:18 tkammer joined #openstack-rally
10:33 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano benchmarks  https://review.openstack.org/137661
10:33 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano base for murano benchmarks  https://review.openstack.org/137650
10:57 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano base for murano benchmarks  https://review.openstack.org/137650
11:01 msdubov_ joined #openstack-rally
11:10 openstackgerrit Oleh Anufriiev proposed stackforge/rally: WIP 'sample' decorator  https://review.openstack.org/140033
11:14 redixin joined #openstack-rally
11:20 nmagnezi_ joined #openstack-rally
11:22 vkmc joined #openstack-rally
11:26 nmagnezi joined #openstack-rally
11:52 tnurlygayanov joined #openstack-rally
12:16 cdent joined #openstack-rally
12:29 openstackgerrit Sergey Skripnick proposed stackforge/rally: Exclude buggy version of pbr in requirements  https://review.openstack.org/143433
12:33 andreykurilin redixin: hi
12:33 redixin andreykurilin, sup
12:34 andreykurilin redixin: it seems to me, that this bug affects a lot of projects. maybe it would be better to proppose such change in global-requirements?
12:34 redixin i know, i just want to see if it fixes bug
12:35 andreykurilin redixin: oh)
12:35 andreykurilin redixin: btw, py27 and pep8 jobs are finished successfully:)
12:36 redixin not for me
12:36 redixin andreykurilin, it is not working on my debian-sid box
12:36 andreykurilin redixin: heh
12:36 redixin but seems problem is not pbr-related
12:38 andreykurilin redixin: hm. https://review.openstack.org/#/c/143351/
13:29 prmtl joined #openstack-rally
13:31 aswadr joined #openstack-rally
13:56 boris-42 redixin: ping
13:56 boris-42 redixin: what about proposing this fix to global-requirements?
13:56 boris-42 redixin: and pinging guys from infra
13:56 boris-42 redixin: even if they fix pbr (they will need to cut new release)
13:57 boris-42 redixin: which means that old one will be broken*
14:13 prmtl joined #openstack-rally
14:14 rook joined #openstack-rally
14:23 tkammer joined #openstack-rally
14:32 redixin boris-42, they are about to update existing versions. see https://review.openstack.org/#/c/143328/ and so on
14:42 boris-42 redixin: Ok nice
14:42 boris-42 redixin: I hope we will get this stuff in
14:43 andreykurilin chen: hi!
14:44 chen andreykurilin hi
14:44 andreykurilin chen: how are you?)
14:45 chen andreykurilin I'm good. :)
14:46 andreykurilin chen: great!) Do you have free time to contibute in Rally? I have several tasks related to porting Rally to py3
14:47 chen andreykurilin sure :)
14:49 andreykurilin chen: nice! Rally have a lot of py34 incompatible stuff. Several issues were reported here https://bugs.launchpad.net/rally/+bugs?field.tag=py3
14:51 chen andreykurilin o......
14:51 andreykurilin chen: it's not all incompatible stuff:)
14:53 andreykurilin chen: so feel free to assign to yourself any bug from this list
14:53 chen andreykurilin ok.IGot it!!!'ll try to fix these bug tomorrow.
14:53 chen andreykurilin now it's time for me to sleep!
14:53 andreykurilin chen: thanks
14:53 andreykurilin chen: good night
14:54 chen andreykurilin I'll let you know if I have questions
14:54 andreykurilin chen: ok
14:58 tkammer joined #openstack-rally
15:07 nkhare joined #openstack-rally
15:14 paboldin joined #openstack-rally
15:37 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the flavor context  https://review.openstack.org/141670
15:37 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
15:37 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
15:37 openstackgerrit Pavel Boldin proposed stackforge/rally: Tests for rally gates  https://review.openstack.org/143469
16:06 paboldin https://nvd3-community.github.io/nvd3/
16:06 paboldin sorry, wrong link: https://nvd3-community.github.io/nvd3/examples/multiBarHorizontalChart.html
16:06 paboldin My pull request for adding error-bars has been merged. Can we switch to this version?
16:09 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano base for murano benchmarks  https://review.openstack.org/137650
16:11 openstackgerrit Merged stackforge/rally: DataBase: Delete BigText class  https://review.openstack.org/143133
16:11 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano base for murano benchmarks  https://review.openstack.org/137650
16:26 rvasilets How to import module that is in optional-requirements? You can look at https://review.openstack.org/#/c/137650/ I have an ImportError: No module named muranoclient
16:42 exploreshaifali joined #openstack-rally
16:44 andreykurilin rvasilets: you should use osloutils.importutils for such stuff
16:47 andreykurilin rvasilets: https://github.com/openstack/oslo.utils/blob/master/oslo/utils/importutils.py#L68
16:48 rvasilets Thx, I will  try
16:51 andreykurilin rvasilets: I think, we should do such stuff in next way: http://pastebin.com/8Dghn5Gr
16:52 rvasilets andreykurilin, Thx
16:57 andreykurilin paboldin: looks good)
16:58 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the flavor context  https://review.openstack.org/141670
16:58 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
16:58 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
16:58 openstackgerrit Pavel Boldin proposed stackforge/rally: Tests for rally gates  https://review.openstack.org/143469
17:13 openstackgerrit Roman Vasilets proposed stackforge/rally: Add murano base for murano benchmarks  https://review.openstack.org/137650
17:15 prmtl joined #openstack-rally
17:36 andreykurilin boris-42, redixin: we have another issue with requirements:(
17:37 andreykurilin boris-42, redixin: prettytable - http://logs.openstack.org/76/143376/2/check/gate-rally-dsvm-rally/ddfd9bc/logs/devstacklog.txt.gz
17:38 redixin T_T
17:39 prmtl joined #openstack-rally
17:54 jlk Is Oleh Anufriiev in the channel?
17:55 andreykurilin jlk, hi! he is offline now. his nickname - oanufriev
17:55 jlk alright, thanks andreykurilin
18:05 boris-42 jlk: what happend?)
18:05 boris-42 jlk: you are worried about context changes?)
18:05 boris-42 jlk: or your patch?
18:05 jlk boris-42: he asked me to chat on IRC about my patch for using admin port when creating users.
18:06 boris-42 jlk: ah
18:06 boris-42 jlk: could you expalin me as well
18:06 boris-42 jlk: cause not sure why user (without admin access) needs admin port
18:07 jlk alright, let me get through my stand up and I'll walk you through it
18:07 boris-42 jlk: sure
18:20 prmtl joined #openstack-rally
18:24 jlk boris-42: okay, stand up is done. I'm going to get some quick debugging where my patch is not in use, vs where my patch is in use.
18:35 jlk boris-42: okay, with out my patch, during a scenario that is boot and delete, there is a point in which we enter the user context, in consume()
18:35 jlk boris-42: there a user_endpoint is created, calling endpoint.Endpoint
18:35 aswadr left #openstack-rally
18:35 jlk no admin port data is passed, so the default admin port gets assumed
18:36 jlk that end point is assigned to a user in the users list
18:39 jlk something later tries to use that end point, now I'm trying to seek where that use is happening
18:40 prmtl joined #openstack-rally
18:41 rook joined #openstack-rally
18:50 harlowja joined #openstack-rally
19:07 jjmb joined #openstack-rally
19:10 prmtl joined #openstack-rally
19:11 boris-42 jlk:  hm very interesting..
19:12 jlk oh interesting
19:13 jlk it's in the config validation
19:13 jlk _validate_config_semantic
19:19 boris-42 jlk: hm
19:19 boris-42 jlk:  so i know where it is
19:19 boris-42 jlk: https://github.com/stackforge/rally/blob/master/rally/benchmark/validation.py#L352-L365
19:19 jlk in validat0r()?
19:19 boris-42 jlk: I belive this
19:19 boris-42 jlk: yep in one that check services
19:19 jlk yeah, and it's doing it via the user context
19:20 boris-42 jlk: fyi I am refactoring that part
19:20 boris-42 jlk: whole validation mechanism
19:20 jlk alright
19:21 jlk in the mean time, for the validation to work, is it safe to pass along the configured admin port for that end point?
19:21 boris-42 jlk: it depends
19:22 boris-42 jlk: there is valiation that works only with admin access
19:22 jlk yeah, so it strikes me that just passing along the admin port configuration to the end point creation shouldn't disrupt anything. All it does is mean that admin actions use the correct port.
19:22 jlk it doesn't force all things to use admin context
19:23 boris-42 jlk: but I don't think that it makes any sense in case of non admin user
19:23 boris-42 jlk: non admin user with admin prot can't do anything
19:23 jlk I don't know where that would happen.
19:24 jlk currently you have to run rally as an admin user
19:30 jlk I'm also looking at bug 1397160 because it breaks sites with custom admin port and https auth_url
19:32 jlk boris-42: if you want to switch subjects, I really think this one is ill thought out.
19:34 boris-42 jlk: yep you have to run rally as a an admin user, believe me I am hard working on dropping that condtion*
19:34 boris-42 jlk: but task is quite huge*
19:34 boris-42 jlk: about bug that you are trying to fix I am more interested in understanding full situation
19:37 jjmb joined #openstack-rally
19:37 boris-42 jlk: btw can we just "not" pass admin port for non admin users?
19:37 boris-42 jlk: will it work?
19:37 jlk no
19:38 jlk boris-42: let me rephrase
19:39 jlk boris-42: what happens during the validation, is the validation call is constructing a url to the admin interface, and so it pulls the endpoint object's admin_port to construct the URL. That falls back to the default admin port if one isn't configured. I suspect this is "working" for most people because they don't have a custom admin port
19:39 jlk since we have a custom admin port, this validation call fails for is as it attempts to hit a port that isn't listening.
19:39 boris-42 jlk: hm be back in second
19:40 jlk okay
19:48 boris-42 jlk:  me back
19:48 boris-42 jlk: so let me try to rephrase what I am trying to say
19:48 boris-42 jlk: I know that we can INIT client without admin_port argument*
19:49 boris-42 jlk: so we can try to remove hardcoded here default value https://github.com/stackforge/rally/blob/master/rally/deploy/engines/existing.py#L127
19:49 jlk oh I see.
19:49 jlk hrm.
19:49 prmtl joined #openstack-rally
19:49 boris-42 jlk: and if it is None during the osclients inti
19:49 boris-42 jlk: we can make .pop
19:50 boris-42 jlk: so it won't be passed to client at all
19:51 jlk what does that accomplish though? Because then it would fall to osclients code default admin port would it not?
19:51 jlk or would it discover that from the service registry?
19:52 boris-42 jlk: actually to be honest that magic that we have in oclinets scares me
19:52 boris-42 =)
19:52 boris-42 jlk: especially this part https://github.com/stackforge/rally/blob/master/rally/osclients.py#L102-L113
19:53 jlk well yes, that's a source of more of my frustration
19:53 boris-42 jlk: so that code is something terrible imho I would like to refactor it
19:53 boris-42 but I don't know how
19:54 boris-42 cause any thouching of it is breaking something
19:56 jlk is the service catalog supposed to return back the admin end point?
19:58 jlk YES
19:58 jlk keystone catalog shows: |   adminURL  | https://openstack.example.com:35358/v2.0 |
19:58 jlk so we should be able to discover admin end point by checking the service catalog
19:59 jlk which you can get from auth_url
19:59 * jlk has to conduct a phone interview
19:59 boris-42 jlk: +2
19:59 boris-42 jlk: approved=)
20:01 jlk hah, sigh
20:57 harlowja_ joined #openstack-rally
21:09 andreykurilin_ joined #openstack-rally
22:30 pboldin joined #openstack-rally
22:30 pboldin Seems like everything is failing: http://logs.openstack.org/66/138466/11/check/gate-rally-dsvm-rally/22232fa/console.html
22:34 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the flavor context  https://review.openstack.org/141670
22:34 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
22:34 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
22:34 openstackgerrit Pavel Boldin proposed stackforge/rally: Tests for rally gates  https://review.openstack.org/143469
22:48 pboldin joined #openstack-rally
22:52 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the flavor context  https://review.openstack.org/141670
22:52 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
22:52 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
22:52 openstackgerrit Pavel Boldin proposed stackforge/rally: Tests for rally gates  https://review.openstack.org/143469
22:52 * pboldin stupido
23:07 pboldin well, the `pip` changed it behaviour
23:07 pboldin http://logs.openstack.org/71/141671/8/check/gate-rally-dsvm-neutron-rally/0cd8619/logs/devstacklog.txt.gz
23:08 pboldin so, now it denies to install a 0.7.2 for >0.7
23:32 nkhare joined #openstack-rally
23:42 pboldin and the bug is in the devstack
23:44 pboldin (and it is already fixed in the upstream)
23:45 pboldin https://review.openstack.org/#/q/I2134c7d8f58f8b83f33150c9ed86d87f8ccba2f3,n,z

| Channels | #openstack-rally index | Today | | Search | Google Search | Plain-Text | summary