Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2015-02-23

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

All times shown according to UTC.

Time Nick Message
00:31 dmorita joined #openstack-rally
00:52 rook joined #openstack-rally
01:13 yingjun joined #openstack-rally
01:13 boris-42 joined #openstack-rally
01:22 yingjun joined #openstack-rally
02:33 yingjun joined #openstack-rally
02:47 ilbot3 joined #openstack-rally
02:47 Topic for #openstack-rally is now ☁ Rally RoadMap: http://goo.gl/JZkmwY ☁ Rally IRC chat logs http://irclog.perlgeek.de/openstack-rally ☁ Key persons to ask:  boris-42, msdubov, rediskin, andreykurilin, amaretskiy  ☁ Documentation: https://rally.readthedocs.org/en/latest/ ☁ To publish changes to Rally:  https://rally.readthedocs.org/en/latest/contribute.html
03:54 rook joined #openstack-rally
04:52 himangi joined #openstack-rally
05:10 rdas joined #openstack-rally
05:24 tfreger joined #openstack-rally
05:26 exploreshaifali joined #openstack-rally
05:40 Mehrnaz joined #openstack-rally
05:41 Mehrnaz hi every body! Good morning! :)
05:41 Mehrnaz I have a question about running tempest by rally!
05:41 anshul joined #openstack-rally
05:41 Mehrnaz How can I run a sample test of tempest by rally?
05:42 Mehrnaz I've used "rally verify start" command, but after a while, It got an error and I couldn't see logs.
05:43 rook joined #openstack-rally
05:43 anshul joined #openstack-rally
05:45 Mehrnaz No one can help me ! :(
05:50 Mehrnaz exploreshaifali: ^
05:52 exploreshaifali Mehrnaz: you don't want to run benchmarks? Just tempest test ?
05:53 Mehrnaz exploreshaifali: yes for now I just want to run tempest test! :)
05:54 exploreshaifali I have no idea how to do that, but boris-42 could help you
05:54 exploreshaifali if he is around
05:55 Mehrnaz exploreshaifali: ok,  thanks for your response. :)
05:55 exploreshaifali Mehrnaz: np :)
05:55 exploreshaifali Mehrnaz: I am sorry, could not help
05:56 Mehrnaz exploreshaifali: don't mention it! :)
05:56 exploreshaifali :)
06:07 rdas joined #openstack-rally
06:08 meteorfox boris-42: you there?
06:16 openstackgerrit Carlos L. Torres proposed stackforge/rally: Sets new users' endpoint_type to default to admin's  https://review.openstack.org/157988
06:26 neeti joined #openstack-rally
06:58 psd joined #openstack-rally
07:13 yfried joined #openstack-rally
07:17 yfried_ joined #openstack-rally
07:32 rook joined #openstack-rally
07:33 pboldin joined #openstack-rally
07:49 e0ne joined #openstack-rally
07:51 Miouge joined #openstack-rally
07:59 exploreshaifali boris-42: around ?
08:05 openstackgerrit Prabhjyot Singh Sodhi proposed stackforge/rally: Adding Security group Scenarios for Neutron  https://review.openstack.org/157458
08:06 nmagnezi joined #openstack-rally
08:06 psd joined #openstack-rally
08:07 fhubik joined #openstack-rally
08:11 psd joined #openstack-rally
08:18 yfried_ boris-42: hi
08:19 yfried_ boris-42: re reviews
08:37 Miouge_ joined #openstack-rally
08:48 yfried_ boris-42: andreykurilin: what's the role for rally.conf? I'm seeing some stuff that could/should be part of the scenario json/yaml and I'm not sure what the rules are
09:10 amaretskiy joined #openstack-rally
09:11 aarefiev joined #openstack-rally
09:12 openstackgerrit Ravikumar Venkatesan proposed stackforge/rally: Adds Nova keypair tests  https://review.openstack.org/156989
09:14 psd_ joined #openstack-rally
09:20 rook joined #openstack-rally
09:26 boris-42 yfried_: hey there
09:27 boris-42 meteorfox: me around now=)
09:28 openstackgerrit Roman Vasilets proposed stackforge/rally: Adding Multiple Request Scenario  https://review.openstack.org/117705
09:29 openstackgerrit Roman Vasilets proposed stackforge/rally: Adding Multiple Request Scenario  https://review.openstack.org/117705
09:34 andreykurilin_ joined #openstack-rally
09:36 nmagnezi amaretskiy, ping
09:37 nmagnezi amaretskiy, regarding you comment https://review.openstack.org/#/c/154825/13/rally/benchmark/scenarios/vm/utils.py
09:37 nmagnezi amaretskiy, i have a question :)
09:38 nmagnezi boris-42, hi, where can i find list of core members in Rally?
09:38 yfried_ boris-42: what's the role for rally.conf? I'm seeing some stuff that could/should be part of the scenario json/yaml and I'm not sure what the rules are
09:38 yfried_ ^ repost
09:40 boris-42 yfried_: I agree with you that most of stuff in config should be configured via task
09:41 boris-42 yfried_: especially ssl stuff
09:41 boris-42 yfried_: but this work is blocked unit we introduce new input task foramt
09:41 boris-42 yfried_: I already have some ideas
09:41 boris-42 yfried_: about new format, I will share them soon
09:41 boris-42 nmagnezi: hi hi
09:41 yfried_ boris-42: I'm more concerned with various timers
09:42 nmagnezi boris-42, hi Boris :-)
09:42 boris-42 nmagnezi: https://review.openstack.org/#/admin/groups/211,members here is the group with core members
09:42 yfried_ boris-42: ie timeout, intervals...
09:42 boris-42 yfried_: yep that as well should me moved under input task
09:42 boris-42 yfried_: I think that we should have 1 context class for all timeouts/intervals
09:43 boris-42 yfried_: or somehow we should put it in deployment info
09:44 boris-42 Mehrnaz: everybody is sleeping at that moment
09:44 boris-42 Mehrnaz: could you run "rally -vd verify start" ?
09:45 yfried_ boris-42: so what's the guideline when reviewing new context/scenario? -1 if it should be in json?
09:46 boris-42 yfried_:  ?
09:46 boris-42 yfried_: what you mean by -1 if it should be in json?
09:47 boris-42 yfried_: you mean if they add new values to rally.conf?
09:47 yfried_ boris-42: some one is creating a new scenario class and adding timers to conf. I think it should be in scenario's json. is that cause for -1?
09:47 yfried_ boris-42: yes
09:47 boris-42 yfried_: so for now it's okay
09:48 boris-42 yfried_: but we should definitely work on refactoring this part
09:49 boris-42 yfried_: do you want to work on this?)
09:51 boris-42 yfried_: by the way related to your patch
09:51 boris-42 yfried_: about atomic action mixin
09:51 boris-42 yfried_:  I would like to merge it with my new class Plugin
09:51 boris-42 yfried_: when I finish it
09:51 boris-42 yfried_: it should be during this week
09:52 yfried_ boris-42: I was unable to work on it (need more unittests)
09:52 yfried_ boris-42: sorry
09:53 boris-42 yfried_: no worries I would like to put that Mixin in plugin.py
09:53 boris-42 yfried_: and add it
09:53 boris-42 yfried_: so at this moment I am switching to common base all plugins base that we have
09:53 yfried_ boris-42: I've no idea what you are saying :)
09:53 boris-42 yfried_: DeployEngines, ServerProviders, Scenarios, Runners, Context, SLA
09:54 yfried_ boris-42: if you want to consume my code into you patch, feel free
09:54 yfried_ boris-42: ohhh...
09:54 yfried_ I think I understand now
09:54 boris-42 yfried_: so in this base we will add one more mixing
09:54 boris-42 yfried_: your mixing AtomicAction
09:54 yfried_ boris-42: you want my Mixin as part of your common base
09:54 boris-42 yfried_: yep yep
09:54 yfried_ boris-42: greate
09:55 boris-42 yfried_: so let's just wait a bit to merge my abse
09:55 yfried_ boris-42: seems reasonable
09:55 boris-42 yfried_: and you will rebase your patch on it
09:55 yfried_ boris-42: ack
09:55 boris-42 yfried_: and we will merge your patch
09:57 openstackgerrit Andrey Kurilin proposed stackforge/rally: Remove all non-incubated oslo code  https://review.openstack.org/152847
09:58 oanufriev joined #openstack-rally
09:58 yfried_ boris-42: re reviews
09:59 amaretskiy joined #openstack-rally
09:59 yfried_ boris-42: there's a tendency to push large patches with not much explanations
09:59 e0ne joined #openstack-rally
09:59 yfried_ boris-42: not very reviewer friendly
10:13 psd_ joined #openstack-rally
10:14 boris-42 yfried_: it's valid point to -1
10:14 openstackgerrit Ravikumar Venkatesan proposed stackforge/rally: Adds Nova keypair tests  https://review.openstack.org/156989
10:14 boris-42 yfried_: if it is not clear what in patch happens
10:20 yfried_ boris-42: it's not a cause for -1 usually. but it discourages new ppl from reviewing
10:21 yfried_ boris-42: I would drop a lot of the reviews I did yesterday and today, if I didn't promise you I would stick with it...
10:23 boris-42 yfried_: it's okay to put -1 if in code something is unclear
10:23 boris-42 yfried_: sometimes large patches doesn't require to much description
10:23 boris-42 yfried_: sometimes not
10:23 boris-42 =)
10:23 boris-42 yfried_: sometimes yes
10:25 yfried_ boris-42: I'm just raising an issue - as cores, you might wanna enforce stricter guidelines on patches to be more reviewer-friendly or you will have more big patches and less reviewers :)
10:32 boris-42 yfried_: yep I am actually asking people to add more details in commit message
10:32 boris-42 yfried_: or to split them when they are too big
10:33 boris-42 yfried_: so if it is unclear/to big for you be free to put -1 and ask to split/add more in commit message
10:33 boris-42 yfried_: as I said this is enough reason to put -1 always
10:39 rook joined #openstack-rally
10:49 Mehrnaz boris-42: hi! thanks for your response! our servers are down now! I'll try it as soon as servers come up!
11:05 jaypipes joined #openstack-rally
11:16 himangi joined #openstack-rally
11:28 aix joined #openstack-rally
11:34 psd_ joined #openstack-rally
11:36 neeti joined #openstack-rally
12:02 psd joined #openstack-rally
12:05 openstackgerrit Prabhjyot Singh Sodhi proposed stackforge/rally: Adding Security group Scenarios for Neutron  https://review.openstack.org/157458
12:05 tfreger left #openstack-rally
12:06 Miouge_ joined #openstack-rally
12:10 Miouge joined #openstack-rally
12:21 Miouge joined #openstack-rally
12:25 exploreshaifali joined #openstack-rally
12:41 cdent joined #openstack-rally
12:42 openstackgerrit Wataru Takase proposed stackforge/rally: Add name pattern filter for resource cleanup  https://review.openstack.org/139643
12:44 Miouge joined #openstack-rally
12:45 psd joined #openstack-rally
13:02 pboros joined #openstack-rally
13:13 mwagner_lap joined #openstack-rally
13:15 openstackgerrit Ivan Kolodyazhny proposed stackforge/rally: WIP. Implement create_and_upload_volume_to_image scenario for Cinder  https://review.openstack.org/152670
13:29 openstackgerrit Wataru Takase proposed stackforge/rally: Add boot server scenario for EC2 API  https://review.openstack.org/147550
13:38 openstackgerrit Roman Vasilets proposed stackforge/rally: Adding Multiple Request Scenario  https://review.openstack.org/117705
13:53 openstackgerrit Pavel Boldin proposed stackforge/rally: Add FIO disk benchmark for VMs  https://review.openstack.org/152388
13:53 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
13:53 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
14:10 openstackgerrit Roman Vasilets proposed stackforge/rally: Adding Multiple Request Scenario  https://review.openstack.org/117705
14:22 openstackgerrit Wataru Takase proposed stackforge/rally: Add list servers scenario for EC2 API  https://review.openstack.org/156228
14:30 openstackgerrit Merged stackforge/rally: Adds Nova keypair tests  https://review.openstack.org/156989
14:31 openstackgerrit Andrey Kurilin proposed stackforge/rally: Remove all non-incubated oslo code  https://review.openstack.org/152847
14:48 openstackgerrit Pavel Boldin proposed stackforge/rally: Add FIO disk benchmark for VMs  https://review.openstack.org/152388
14:48 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the generic script-based benchmark for the VMs  https://review.openstack.org/141671
14:48 openstackgerrit Pavel Boldin proposed stackforge/rally: Add the context benchmark_image  https://review.openstack.org/138466
14:56 e0ne joined #openstack-rally
15:20 Miouge joined #openstack-rally
15:26 Miouge joined #openstack-rally
15:32 openstackgerrit Sergey Skripnick proposed stackforge/rally: Test murano-benchmarks vs mos (do-not-merge)  https://review.openstack.org/158338
15:35 openstackgerrit Sergey Skripnick proposed stackforge/rally: Test murano-benchmarks vs mos (do-not-merge)  https://review.openstack.org/158338
15:38 exploreshaifali joined #openstack-rally
15:44 openstackgerrit Wataru Takase proposed stackforge/rally: Add boot server scenario for EC2 API  https://review.openstack.org/147550
15:45 Miouge joined #openstack-rally
15:47 exploreshaifali boris-42, gooood morning!!
15:47 exploreshaifali boris-42, around ?
16:05 psd_ joined #openstack-rally
16:10 Miouge joined #openstack-rally
16:12 yfried_ joined #openstack-rally
16:25 andreykurilin_ joined #openstack-rally
16:27 boris-42 exploreshaifali: yep
16:31 mwagner_ joined #openstack-rally
16:31 exploreshaifali boris-42, the parameters we are taking into account are duration, errors; does your robot mind can think of others too ?
16:32 exploreshaifali :P
16:33 e0ne joined #openstack-rally
16:36 boris-42 exploreshaifali: these arguments are just begging
16:36 boris-42 exploreshaifali: to start from something simple
16:37 openstackgerrit Sergey Skripnick proposed stackforge/rally: Test murano-benchmarks vs mos (do-not-merge)  https://review.openstack.org/158338
16:38 exploreshaifali boris-42, okay, so what are other parameters we need to consider?
16:39 boris-42 exploreshaifali: atomic actions, avarge/min/max/precnetile durations
16:40 boris-42 exploreshaifali: as well we can do some analyze of this data
16:42 Miouge joined #openstack-rally
16:48 meteorfox boris-42: is there something else I have to do to get my changeset merged? https://review.openstack.org/157988
16:49 boris-42 meteorfox: just that
16:49 boris-42 meteorfox: sorry for nitpicking but it was hell to replace all ' -> "
16:50 boris-42 meteorfox: I don't want to repeat it, actually we have CI job but it failed for some reason...
16:50 boris-42 meteorfox: here is the hell http://stackalytics.com/?release=kilo&metric=commits&project_type=all&module=rally&company=&user_id=jacobliberman%40gmail.com
16:51 meteorfox boris-42: ? I don't follow
16:52 meteorfox boris-42: ok, you are talking about the string quotes
16:52 boris-42 meteorfox: yep
16:52 boris-42 meteorfox: we have special job Mirantis Rally CI pep8 that checks them
16:52 boris-42 meteorfox: for some not clear reason it stopped working..
16:53 psd_ boris-42, Hey!
16:54 meteorfox boris-42: sorry about that. is just that every project seems to have their own nitpickings about string quotes. A few projects out there, like SaltStack, for example only use single quotes
16:54 boris-42 meteorfox: so we decided to use " because in most cases we used them..
16:54 boris-42 meteorfox: could you please fix it and we will get merged your patch
16:55 meteorfox boris-42: noted, I'll double check my commit for single quotes, and change them
16:59 boris-42 meteorfox: thank you
16:59 boris-42 meteorfox: acutlaly you can install
16:59 boris-42 meteorfox: flake8-double-quotas
16:59 boris-42 meteorfox: it will check them automatically
16:59 boris-42 psd_: hey hey
17:02 psd_ joined #openstack-rally
17:03 psd_ boris-42, need some help, as andreykurilin pointed out, there is no security_group option in neutron-pythonclient, (re  https://review.openstack.org/#/c/157458/)
17:06 openstackgerrit Wataru Takase proposed stackforge/rally: Support multiple EC2 APIs  https://review.openstack.org/156991
17:07 exploreshaifali psd_, there what andreykurilin is saying that "there is no security_groups option in neutron client"
17:07 exploreshaifali psd_, this is your line of code  "sg = self.clients("neutron").security_groups.create("
17:07 psd_ exploreshaifali, yups, I got that
17:07 exploreshaifali there *.security_groups* is not needed
17:08 exploreshaifali rather "create_security_group()" can be used
17:09 exploreshaifali psd_, so it could be "sg = self.clients("neutron").create_security_group(????)"
17:09 yfried_ joined #openstack-rally
17:09 Miouge joined #openstack-rally
17:10 psd_ exploreshaifali, thanks!
17:10 exploreshaifali psd_, only if it will work :P
17:11 psd :)
17:15 Miouge joined #openstack-rally
17:15 openstackgerrit Carlos L. Torres proposed stackforge/rally: Sets new users' endpoint_type to default to admin's  https://review.openstack.org/157988
17:20 boris-42 psd: hm
17:20 boris-42 psd: you need to take a look how to delete security group in neutron client
17:21 boris-42 psd: neutron client is not similar to nova
17:22 psd okay
17:25 jlk joined #openstack-rally
17:33 openstackgerrit Wataru Takase proposed stackforge/rally: Add name pattern filter for resource cleanup  https://review.openstack.org/139643
17:40 ibiris joined #openstack-rally
17:59 meteorfox boris-42: change for double quotes is up, also, I've verified in our environment, that I can finally run the scenarios against the internal URLs
18:05 boris-42 meteorfox: nice
18:10 meteorfox boris-42: I want your opinion on something. I know Rally include several sample tasks, which some are very basic. Obviously, I understand that anybody using Rally, can create their own tasks, based on these samples. What would be your recommendation in keeping these new tasks in order? Would you say that a separate git repo for just tasks is a good idea?
18:33 harlowja joined #openstack-rally
18:42 jay_t_c joined #openstack-rally
18:46 jay_t_c boris-42: are you around today?
18:55 yfried_ joined #openstack-rally
19:30 psd joined #openstack-rally
19:36 boris-42 meteorfox: I dislike the idea of having separated repo
19:37 boris-42 meteorfox: I really don't see any benefit.
19:37 meteorfox boris-42: true, so do you think tasks should be included in the main Rally repo, and have a folder for each vendor?
19:38 boris-42 meteorfox: so my idea is to have 1 or few tasks that will cover all stuff
19:38 boris-42 meteorfox: they will be parametrized
19:38 boris-42 meteorfox: so they will work for everybody
19:40 boris-42 meteorfox: so but before maybe it makes sense to change input format?
19:40 meteorfox boris-42: ok, but most likely I'll have very specific scenarios, where concurrency for boot_and_delete is very high, and might use un-published images and/or flavors
19:40 meteorfox boris-42: do you think I should keep that in my own repo
19:40 boris-42 meteorfox: flavors and images are arguments
19:40 boris-42 meteorfox: tasks are parametrized
19:41 meteorfox boris-42: right I know, but my concern is the particular tasks with those specific parameters
19:41 boris-42 meteorfox: let me show one patch
19:42 boris-42 meteorfox: https://github.com/stackforge/rally/commit/a2d28d1f9f469b1699d31814a957505935803a17
19:42 boris-42 meteorfox: I think you didn't understand what I mean by parametrized*
19:43 boris-42 meteorfox: ^ take a look at that patch
19:44 meteorfox boris-42: ah I see
19:44 boris-42 meteorfox: so input tasks are jinja2 templates
19:44 meteorfox right
19:44 boris-42 meteorfox: we have great flexibility to write single "template"
19:44 boris-42 meteorfox: that will depend on few arguments and work for everybody
19:45 meteorfox boris-42: I see, this should work
19:46 boris-42 meteorfox: so we don't need to keep N tasks for different vendors and can do this in upstream of rally
19:46 meteorfox boris-42: makes sense
19:46 meteorfox boris-42: is this patch in upstream ?
19:47 boris-42 meteorfox: yep we merged it 1.5 months ago
19:48 meteorfox boris-42: awesome, then I'll use this approach.
19:48 meteorfox boris-42: quick question, kind of unrelated, can I work on a patch to include median on the summary report
19:48 meteorfox ?
19:51 e0ne joined #openstack-rally
19:52 openstackgerrit joined #openstack-rally
19:52 meteorfox boris-42: or was there a reason this wasn't included?
20:00 openstackgerrit Wataru Takase proposed stackforge/rally: Add name pattern filter for resource cleanup  https://review.openstack.org/139643
20:06 boris-42 meteorfox: median?
20:06 meteorfox boris-42: like 50 percentile
20:07 meteorfox yes
20:08 boris-42 meteorfox: yep sure
20:08 boris-42 meteorfox: be free to include it
20:08 meteorfox boris-42: cool, I assume I need to also make sure it works with the HTML report
20:11 boris-42 meteorfox: HTML report is separated from CLI
20:13 meteorfox boris-42: ok, separate change sets then
20:13 boris-42 meteorfox: you can make together* I mean that you need to change in both places
20:14 meteorfox boris-42: sure, this should be easy
20:19 meteorfox boris-42: can you show me an example of the command for --task-args? Here's what I did, but it arguments were ignored, and it ran with the defaults: https://gist.github.com/meteorfox/e7e1ce7c6cdb2f0865f5
20:19 andreykurilin_ joined #openstack-rally
20:24 boris-42 meteorfox: this won't work
20:24 boris-42 meteorfox: {% set nr_concurrent = nr_concurrent or 1 %}
20:24 boris-42 will work
20:24 meteorfox boris-42: ok, I could also omit the set, right?
20:29 psd joined #openstack-rally
20:29 psd joined #openstack-rally
20:34 harlowja_ joined #openstack-rally
20:38 exploreshaifali joined #openstack-rally
21:38 psd_ joined #openstack-rally
21:52 abhirc joined #openstack-rally
21:57 boris-42 meteorfox: yep
21:57 boris-42 meteorfox: set is required only if you want to have default value
22:17 psd__ joined #openstack-rally
22:26 aix joined #openstack-rally
22:29 openstackgerrit Merged stackforge/rally: Sets new users' endpoint_type to default to admin's  https://review.openstack.org/157988
22:30 harlowja joined #openstack-rally
22:45 openstackgerrit Merged stackforge/rally: Add new script for gate jobs  https://review.openstack.org/144633
23:00 openstackgerrit John Wu proposed stackforge/rally: Add base for Swift API Benchmarks  https://review.openstack.org/154225
23:15 openstackgerrit Wataru Takase proposed stackforge/rally: Add boot server scenario for EC2 API  https://review.openstack.org/147550
23:16 harlowja joined #openstack-rally
23:34 psd_ joined #openstack-rally
23:35 openstackgerrit Andrey Kurilin proposed stackforge/rally: WIP: found python2.7 executor for verification  https://review.openstack.org/158490
23:41 abhirc joined #openstack-rally
23:44 psd joined #openstack-rally

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