Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2015-04-12

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

All times shown according to UTC.

Time Nick Message
00:38 rook_ joined #openstack-rally
00:44 darkhuy2 joined #openstack-rally
01:39 rook_ joined #openstack-rally
01:47 ilbot3 joined #openstack-rally
01:47 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, yfried, msdubov, rediskin, andreykurilin, amaretskiy  ☁ Documentation: https://rally.readthedocs.org/en/latest/ ☁ To publish changes to Rally:  https://rally.readthedocs.org/en/latest/contribute.html
01:59 baker joined #openstack-rally
02:36 darkhuy2 joined #openstack-rally
02:37 darkhuy2 Boris-42:ping
02:39 rook_ joined #openstack-rally
02:40 davideagnello joined #openstack-rally
03:16 baker joined #openstack-rally
03:23 sputnik13 joined #openstack-rally
03:40 rook_ joined #openstack-rally
04:09 prashantS joined #openstack-rally
04:17 baker joined #openstack-rally
04:41 rook_ joined #openstack-rally
04:44 tfreger joined #openstack-rally
05:05 psd joined #openstack-rally
05:17 prashantS joined #openstack-rally
05:38 yfried joined #openstack-rally
05:41 baker joined #openstack-rally
05:42 rook_ joined #openstack-rally
05:47 serverascode_ joined #openstack-rally
05:48 prashantS joined #openstack-rally
05:58 davideagnello joined #openstack-rally
06:18 prashantS joined #openstack-rally
06:25 e0ne joined #openstack-rally
06:38 nmagnezi joined #openstack-rally
06:40 e0ne joined #openstack-rally
06:43 rook_ joined #openstack-rally
06:50 e0ne joined #openstack-rally
07:29 baker joined #openstack-rally
07:30 openstackgerrit Wataru Takase proposed stackforge/rally: Support multiple EC2 APIs  https://review.openstack.org/156991
07:43 rook_ joined #openstack-rally
07:46 boris-42 darkhuy: ping
07:47 davideagnello joined #openstack-rally
07:51 darkhuy_ joined #openstack-rally
07:51 darkhuy_ boris-42:ping
07:56 boris-42 darkhuy_: yep yep
07:56 darkhuy_ finally :-D
07:56 boris-42 darkhuy_: I am here
07:56 darkhuy_ boris-42: this time zone difference is so difficult :-p
07:56 boris-42 darkhuy_: it's not only TZ
07:57 boris-42 darkhuy_: I was celebrating Rally moves to OpenStack =)
07:57 darkhuy_ boris-42: I saw! i was reading your guys meeting logs
07:58 darkhuy_ boris-42: anywho just wanted to introduce myself as a hopeful new contributor :-P
08:01 boris-42 darkhuy_: that is nice=)
08:01 boris-42 darkhuy_: so what you would like to work on?)
08:02 darkhuy_ boris-42: Of all the modules so far, rally seems most interesting to me. Most of my experience (only 2 years :-p) has been in performance and benchmarking automation
08:02 darkhuy_ So rally kind of hit right at home
08:04 boris-42 darkhuy_: that's is nice
08:04 boris-42 darkhuy_: so we have a lot of interesting topics as well a lot of simple and complex tasks
08:05 boris-42 darkhuy_: I believe the best way to get familiar is to work on some plugins
08:07 darkhuy_ Anything specific? I've been working on some fairly low priority bugs, but they were just simple ones to get my feet wet
08:12 boris-42 darkhuy_: here is rally roadmap https://docs.google.com/a/mirantis.com/spreadsheets/d/16DXpfbqvlzMFaqaXAcJsBzzpowb_XpymaK2aFY2gA2g/edit?pli=1#gid=0
08:12 boris-42 darkhuy_: you can as first step get a bit familiar
08:12 boris-42 darkhuy_: then you can take a part in current disucssions regarding to new features/improving old
08:13 boris-42 darkhuy_: and you can take this one task https://bugs.launchpad.net/rally/+bug/1437515
08:13 boris-42 darkhuy_: it's quite easy to start
08:14 darkhuy_ boris-42: sounds good
08:17 boris-42 darkhuy_: btw if you are related to performance
08:17 boris-42 darkhuy_: take a look at this https://github.com/boris-42/profimp
08:17 boris-42 darkhuy_: this can help us to speed up starting rally
08:17 boris-42 darkhuy_: by optimization of imports
08:18 darkhuy_ looks cool!
08:21 boris-42 darkhuy_: hm are you Hugh?)
08:21 darkhuy_ boris-42: yes
08:21 boris-42 darkhuy_: ahh=) I saw a lot of your patches recently =)
08:21 boris-42 darkhuy_: nice work
08:21 darkhuy_ boris-42: only simple ones :-P the learning curve on openstack code is very har
08:23 darkhuy_ boris-42: so in the case of you comment just now, should i just abandon the changes, as I think the other code under review is much more in-depth
08:34 boris-42 darkhuy_: seems like so
08:34 boris-42 darkhuy_: sorry for wasting your time, somehow that bug was not attached to that guys
08:35 boris-42 darkhuy_: that are already working..
08:36 darkhuy_ np
08:36 boris-42 darkhuy_: btw where you are from?
08:36 darkhuy_ uswest, san jose
08:38 boris-42 darkhuy_: oh I am going to relocate to mtv
08:38 darkhuy_ boris-42: oh nice!
08:38 boris-42 darkhuy_: ya=)
08:38 darkhuy_ boris-42: very close!
08:38 boris-42 darkhuy_: if I get visa finally=)
08:38 boris-42 darkhuy_: yep yep=)
08:40 andreykurilin joined #openstack-rally
08:41 boris-42 darkhuy_: lol soon you are going to be top contirbutor in rally
08:41 boris-42 darkhuy_: +2 patches for one day =)
08:41 darkhuy_ boris-42: haha im sure as bug get harder, that will not be possible
08:44 rook_ joined #openstack-rally
08:44 yfried boris-42: hi
08:46 boris-42 yfried: hi there
08:46 boris-42 yfried: so now you are core of official OpenStack project=)
08:48 yfried boris-42: and I'm one step closer to complete my master plan
08:48 yfried boris-42: and TAKE OVER THE WORLD!
08:48 yfried boris-42: jk
08:48 boris-42 yfried: master plan ?
08:48 boris-42 yfried:  hehe=)
08:49 yfried boris-42: re https://review.openstack.org/#/c/154885/
08:49 boris-42 yfried: uh that one
08:49 boris-42 yfried: I hope nmagnezi is not angry ..
08:49 yfried boris-42: you've had an idea of how to change that to create a new context
08:49 yfried boris-42: he isn't. but he had to drop this as he has joined neutron-dev team
08:49 nmagnezi boris-42, morning :-)
08:50 boris-42 nmagnezi: hi there
08:50 yfried boris-42: I've agreed to take this off his hands
08:50 nmagnezi boris-42, why would i be angry? :-)
08:50 boris-42 nmagnezi: I mean I blocked a bit your work that sux
08:50 boris-42 nmagnezi: but I am ready to help to provide this functionallity
08:51 yfried boris-42: since it's been a while and I've been on vacation, I was hoping you would explain (again) how you'd prefer this to go, and if you had an example for that, it would be great
08:51 boris-42 yfried: so my common idea is next
08:51 boris-42 Do the common work regarding to network (that can be done in both nova-network and neutron in network context)
08:52 boris-42 All specific steps in separated
08:52 boris-42 so users should specify network dual stack context
08:52 boris-42 network + dual stack contexts*
08:53 boris-42 this will be simpler to validate in future when we move to plugin base
08:53 boris-42 like dual stack context will require neutron
08:53 boris-42 if we put together network context will require neutron *
08:54 boris-42 yfried: ^
08:55 boris-42 yfried: if there is NO WAY to split dual stack from current network context, I would prefer to create separated context
08:55 boris-42 network_dual_stack
08:56 nmagnezi boris-42, yfried will probably takeover those patches.. my tasks and focus are going to change soon.
08:56 yfried boris-42: not arguing - just brainstorming:
08:56 nmagnezi boris-42, and no that's no because of your feedback :-)
08:56 openstackgerrit Andrey Kurilin proposed stackforge/rally: add aborted and broken flags to sla checker  https://review.openstack.org/172596
08:56 openstackgerrit Andrey Kurilin proposed stackforge/rally: WIP: return from_context_uploaded scenarios  https://review.openstack.org/172702
08:56 yfried boris-42: network ctx creates the network for the tenant
08:58 yfried boris-42: it does that with net-wrapper
08:58 boris-42 yfried: just one thing, I am not saying that code that we have in master is ideal..
08:58 yfried boris-42: current create_network in net-wrap
08:58 boris-42 But new code really produce/shows/displays a lot of issues
08:59 yfried boris-42: has add_router kwarg
08:59 boris-42 so we need to refactor current code or do new code in other way
08:59 boris-42 yfried: so just one more time the whole idea of net_wrapper was to have common interface between nova-network and neutron
09:00 yfried boris-42: which is the root of all the problems, since it isn't comatible with nova
09:00 boris-42 yfried: I agree with you on this
09:00 boris-42 yfried: but there is part of functioanllity that exist in both*
09:00 yfried boris-42: so you'd like to drop the add_router kwarg from create-network
09:00 yfried ?
09:01 boris-42 yfried: so amaretsky was suggesting another thing
09:01 yfried boris-42: ?
09:02 boris-42 yfried: so first of all to call this code like services * not wrappers
09:03 boris-42 yfried:  I partially agree with him, that we should have some kind of common code
09:03 boris-42 yfried: for all services that will unify work with various versions and so on
09:03 boris-42 yfried: like we have currently in benchmark/*/utils.py and benchmark/wrappers/*
09:03 darkhuy_ boris-42: yfried: gnite :)
09:04 boris-42 yfried: that will resolve first question regarding to where to place neutron specific functionallity that can do high level operations
09:04 boris-42 yfried: maybe we should create spec for this?
09:05 boris-42 yfried: because it requires a lot of refactoring ?
09:06 yfried boris-42: agreed re spec. I'm not following the whole "service" idea and the differences
09:07 openstackgerrit Merged stackforge/rally: Remove DeployEngines in favor of DeploymentEngine  https://review.openstack.org/172675
09:08 boris-42 yfried: let's split current task to 2
09:08 boris-42 yfried: one is refactoring wrappers to avoid that bad stuff
09:09 boris-42 yfried: another one is context (absolutelly independet task)
09:09 boris-42 yfried: so now we are talking about wrappers/* and utils.py
09:09 boris-42 yfried: I would like to have in benchmark/os_services/* directory
09:09 boris-42 inside it to have all code from utils and wrappers
09:09 boris-42 like we can have
09:10 boris-42 os_services/keystone/ directory
09:10 boris-42 that will contain common.py, v2.py, v3.py
09:11 boris-42 in common py we will put service that unifies v2 and v3 (and does everything that can be done in both)
09:11 boris-42 in v2 and v3 we are putting all functionality of services
09:12 boris-42 atomic action stuff will be build-in in base class of services
09:12 boris-42 so this services will be quite easy to reuse in our benchmark scenario and context
09:13 boris-42 yfried: ^
09:18 baker joined #openstack-rally
09:19 yfried boris-42: going to lunch. back in 30min
09:20 yfried boris-42: I like that
09:20 yfried boris-42: talk more when I'm back from lunch
09:24 boris-42 yfried|lunch: ok
09:31 openstackgerrit Boris Pavlovic proposed stackforge/rally: Add EC2 api scenario that list all servers  https://review.openstack.org/168991
09:32 openstackgerrit Merged stackforge/rally: Replaced deprecated 'use deployment' with 'deployment use'  https://review.openstack.org/172191
09:42 yfried boris-42: ok, so I'm back
09:43 yfried boris-42: how do you want to proceed with the whole "service" thing, and what do you want to do with the problematic patch meanwhile?
09:45 rook_ joined #openstack-rally
09:58 subscope_ joined #openstack-rally
10:11 boris-42 yfried: so could you start spec
10:11 boris-42 yfried: we will get merged spec and we will know what to do with these paches
10:13 yfried boris-42: spec for unifying all of Benchmark/<service>/utils.py under Benchmark/services/<service>.py?
10:14 boris-42 yfried: and wrappers*
10:14 boris-42 yfried: ye
10:17 yfried boris-42: so, what would happen to the wrapper under that tree?
10:18 boris-42 yfried: I already tried to explain
10:19 boris-42 https://www.irccloud.com/pastebin/L8iGyYwH
10:19 boris-42 ^ there will be service/<name>/common.py
10:19 boris-42 that will be actually wrapper
10:19 boris-42 and call other particular services
10:23 boris-42 yfried: ^
10:29 yfried boris-42: that's fine for keystone versions
10:29 yfried boris-42: how do you handle neutron/nova ?
10:29 boris-42 yfried: in the same way
10:30 yfried boris-42: you don't have a shared ancestor for them
10:30 boris-42 yfried: benchmark/netowork/ will have common.py and nova_network and neutron
10:30 boris-42 yfried: you have
10:30 boris-42 current network wrapper
10:30 boris-42 is common for both
10:30 yfried boris-42: I'll write a draft
10:31 boris-42 yfried: current arch has issue
10:31 yfried boris-42: could you plz also help me with the motivation? since you and amaretskiy are the ones who came up whit that?
10:31 boris-42 yfried: motivition?
10:31 yfried boris-42: "reason for submitting the spec"
10:32 boris-42 yfried: for this task?
10:32 boris-42 yfried: this task is to BIG to make mistakes
10:32 boris-42 yfried: so we need to discuss all stuff before doing any code
10:32 boris-42 yfried: and there are some unclear areas
10:33 yfried boris-42: no I mean - the problems with the current code
10:33 boris-42 yfried: so problem with current code is obvious, and many rally devs pointed us
10:34 boris-42 yfried: that it is hard to use code from scenarios utils in context
10:34 boris-42 yfried: as well this wrappers are some kind of strange solutions that makes things unclear
10:34 openstackgerrit Merged stackforge/rally: Remove keystone v3 jobs from MOS  https://review.openstack.org/172001
10:45 rook_ joined #openstack-rally
11:07 baker joined #openstack-rally
11:23 psd joined #openstack-rally
11:46 rook_ joined #openstack-rally
11:59 yfried joined #openstack-rally
12:47 rook_ joined #openstack-rally
12:53 rook_ joined #openstack-rally
12:53 baker joined #openstack-rally
13:14 davideagnello joined #openstack-rally
14:09 openstackgerrit Pradeep K Surisetty proposed stackforge/rally: Add keystone create_and_delete_tenant scenario  https://review.openstack.org/172716
14:40 cdent joined #openstack-rally
14:59 andreykurilin joined #openstack-rally
15:00 prashantS joined #openstack-rally
15:16 prashantS joined #openstack-rally
15:48 darkhuy2 joined #openstack-rally
16:41 yfried joined #openstack-rally
16:51 davideagnello joined #openstack-rally
17:28 mwagner_lap joined #openstack-rally
17:49 e0ne joined #openstack-rally
17:56 e0ne joined #openstack-rally
18:18 prashantS joined #openstack-rally
18:38 ptoohill boris-42: Yes, we have tempest testing inside our project and from what it seems is that we would need the tempest test along with a way to tag for our project to run via rally. Thank you for the response.
18:39 ptoohill tests inside rally*
18:40 prashantS joined #openstack-rally
18:46 ptoohill So, to clear that up a bit; We were asked to move tempest test inside our project 'neutron-lbaas', so im curious if Rally can point or 'verify' tests that live under neutron-lbaas. Currently we have it set up to run via tox i.e 'tox -e tempest'. I am curious if we could run 'rally verify start neutron_lbaas.tests.temepest' or something similar?
18:47 ptoohill Its possible im asking for something thats not the focus of Rally, but any input is appreciated.
19:37 e0ne joined #openstack-rally
20:45 boris-42 ptoohill around?
20:46 ptoohill yep, hello
20:49 ptoohill boris-42:
21:40 prashantS joined #openstack-rally
21:45 andreykurilin joined #openstack-rally
21:48 boris-42 ptoohill: so
21:48 boris-42 ptoohill: okay the question is clear for me
21:49 boris-42 ptoohill: so we are thinking about making this possible
21:49 boris-42 http://boris-42.me/rally-verfiy-as-the-control-plane-for-gabbi-tempest-in-tree-functional-tests/
21:49 boris-42 ptoohill: and here is spec https://review.openstack.org/#/c/166487/
21:49 boris-42 ptoohill: so you can take a part in discussion
21:50 boris-42 ptoohill: but there is no way for now to do it out of box
21:51 boris-42 ptoohill: I am going to sleep a bit
21:52 boris-42 ptoohill: see you later
21:54 prashantS joined #openstack-rally
22:09 prashantS joined #openstack-rally
22:16 prashantS joined #openstack-rally
22:38 ptoohill Thank you much boris-42!
23:12 prashantS joined #openstack-rally
23:24 prashantS joined #openstack-rally

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