Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2014-08-03

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

All times shown according to UTC.

Time Nick Message
00:10 openstackgerrit Rafi Khardalian proposed a change to stackforge/rally: Add support for Nova live migrations  https://review.openstack.org/111444
01:10 yingjun joined #openstack-rally
04:40 penguinRaider__ joined #openstack-rally
05:07 openstackgerrit veena proposed a change to stackforge/rally: Renames *.rst files in /rally/doc/samples/tasks  https://review.openstack.org/111484
06:51 openstackgerrit A change was merged to stackforge/rally: Renames *.rst files in /rally/doc/samples/tasks  https://review.openstack.org/111484
08:26 jjmb joined #openstack-rally
08:28 tzabal left #openstack-rally
12:25 jjmb joined #openstack-rally
13:40 yingjun joined #openstack-rally
13:42 jjmb joined #openstack-rally
14:45 yingjun joined #openstack-rally
14:58 rook joined #openstack-rally
15:00 jjmb joined #openstack-rally
15:03 jjmb1 joined #openstack-rally
15:04 jjmb2 joined #openstack-rally
15:16 jjmb joined #openstack-rally
15:31 yingjun joined #openstack-rally
15:38 jjmb joined #openstack-rally
16:05 boris-42 rook ping
17:35 boris-42 penguinRaider__ around?
17:36 penguinRaider__ boris-42, yeah
17:36 boris-42 penguinRaider__ I have an idea
17:37 boris-42 penguinRaider__ what you can do
17:37 penguinRaider__ boris-42, cool :-)
17:37 boris-42 penguinRaider__ benchmark that will just run HTTP request
17:37 boris-42 penguinRaider__ and check status and so on
17:39 penguinRaider__ boris-42, I dont really understand it, will it be a scenario ?
17:39 boris-42 penguinRaider__ yep
17:40 boris-42 penguinRaider__ scenario that assept as a arg
17:40 boris-42 penguinRaider__ url and probably expected response
17:40 boris-42 penguinRaider__ that just execute it
17:42 penguinRaider__ boris-42, like try to get a url and compare response with expected response?
17:42 boris-42 penguinRaider__ yep
17:42 boris-42 penguinRaider__ I am going to remove requirement from using task
17:46 penguinRaider__ boris-42, ^^ I dont get the last statement ?
17:46 boris-42 penguinRaider__ oh sorry deployment*
17:46 boris-42 penguinRaider__ task won't require deployments
17:50 penguinRaider__ boris-42, I dont have much idea about deployment code what do they do ?
17:57 boris-42 penguinRaider__ ...
17:57 boris-42 penguinRaider__ nevermind
17:57 boris-42 penguinRaider__ just make http benchmark
18:11 Topic for #openstack-rally is now ☁ Rally IRC chat logs http://irclog.perlgeek.de/openstack-rally ☁ Key persons to ask:  boris-42, rediskin, andreykurilin, k4n0  ☁ Open tasks https://trello.com/b/DoD8aeZy/rally ☁ To publish changes to Rally:  https://wiki.openstack.org/wiki/Rally/Develop#How_to_contribute  ☁ To run unit tests locally install tox<=1.6.1 and run "tox"  ☁ Rally human2human weekly updates:  https
18:11 Topic for #openstack-rally is now ☁ Rally IRC chat logs http://irclog.perlgeek.de/openstack-rally ☁ Key persons to ask:  boris-42, rediskin, andreykurilin, k4n0  ☁ Open tasks https://trello.com/b/DoD8aeZy/rally ☁ To publish changes to Rally:  https://wiki.openstack.org/wiki/Rally/Develop#How_to_contribute  ☁ To run unit tests locally install tox<=1.6.1 and run "tox"  ☁ Rally human2human weekly updates:  https
18:12 Topic for #openstack-rally is now ☁ Rally IRC chat logs http://irclog.perlgeek.de/openstack-rally ☁ Key persons to ask:  boris-42, rediskin, andreykurilin, k4n0  ☁ Open tasks https://trello.com/b/DoD8aeZy/rally ☁ To publish changes to Rally:  https://wiki.openstack.org/wiki/Rally/Develop#How_to_contribute   ☁ Rally weekly updates:  https://wiki.openstack.org/wiki/Rally/Updates
19:20 boris-42 zhiyan around ?
20:00 rook boris-42: ?
20:03 boris-42 rook I just wanna say that now our chat is logged* I don't remember who exactly asked for this=)
20:03 boris-42 rook but logs can be found http://irclog.perlgeek.de/openstack-rally here
20:06 rook oh, i never asked, but good to know!
20:11 boris-42 rook yep sometimes is useful to have history of chats
20:11 rook boris-42 sure
20:12 boris-42 rook ok now i am fully back from any kind of vacation and remote work=)
20:13 rook Cool, i will be online but AFK in training all this coming week
20:13 rook then PTO the next week, then travel the following ;)
20:18 boris-42 rook nicce
20:18 boris-42 rook have a good vacation=)
20:18 rook lol not a vaca but jus away
20:18 boris-42 rook heh
20:19 boris-42 rook I hope benchmark image context will be ready
20:19 boris-42 rook so you'll be able to work on cbtool integration=)
20:19 rook the image prep tool - yeah, I have been pushing forward without that.
20:19 boris-42 rook you don't need that?
20:20 rook Nope - Prep the image in glance before hand.
20:20 boris-42 rook this is bad idea
20:20 boris-42 =)
20:20 rook boris-42 but I have been looking at the image prep being proposed.
20:20 boris-42 rook it's hard to use => nobody uses =(
20:21 boris-42 rook so it will be better to have a context that will prepare image
20:21 boris-42 rook from any specified
20:21 boris-42 rook I like 1 click solutions
20:21 boris-42 =)
20:21 boris-42 rook some day we will have webUI
20:21 boris-42 rook and it will be really 1 click=)
20:22 rook boris-42 ha - we will see, I have high hopes for the image prep, but I fear it will be done serially and be very slow to stand up environments.
20:22 boris-42 rook what do you mean by serial?
20:22 rook not to mention, it will have a hard requirement that guests have external access or a repository somewhere of packages to pull from.
20:22 boris-42 rook for every tenant?
20:22 rook boris-42 no
20:23 rook boris-42 most benchmarks I have worked iwth are not a single guest
20:23 boris-42 rook yep
20:23 rook boris-42 for example Hadoop workloads are hundreds of guests.
20:23 boris-42 rook so why it's the issue? to… prepare 1 time image
20:23 rook boris-42 things we can work through
20:23 boris-42 rook and then run hundreds VMs from this snapshot?
20:23 rook boris-42 so put all hadoop packages in a single guest?
20:24 rook that could bloat things (not just in the terms of hadoop)
20:24 boris-42 rook you need different guests?
20:24 rook we ill need a "role" built into this logic.
20:24 boris-42 rook for single benchmark?
20:24 boris-42 rook then you can prepare few different images
20:24 rook Think more than guests, people want to trim down to Containers... and in containers the users do not want to have unnecessary bloat.
20:25 rook boris-42 this functionality is there?
20:25 * rook is reviewing the code again.
20:26 rook I didn't see a way to support multiple snapshots
20:26 boris-42 rook so we can extend that
20:26 boris-42 rook the idea is that we will have base class that do all logic about preparing image
20:26 boris-42 rook and have subclasses
20:26 boris-42 rook that installs specific for benchmark stuff
20:26 boris-42 rook but good point
20:26 boris-42 rook about multple snapshots
20:27 rook boris-42 roger - I will review things again and provide feedback =)
20:27 rook boris-42 but I have a pregnant wife calling my name, trying not to catch hell... ttyl =)
20:27 boris-42 rook =)))))
20:27 boris-42 rook see you=)
20:27 rook later
20:42 boris-42 rmk I commented your patch about endpoint_type
20:42 boris-42 rmk there is a better solution=)
20:46 openstackgerrit Boris Pavlovic proposed a change to stackforge/rally: Remove magic() method from context manager  https://review.openstack.org/104962
21:35 rook joined #openstack-rally
21:47 rook joined #openstack-rally
22:35 rook joined #openstack-rally
22:58 rook joined #openstack-rally
22:58 openstackgerrit A change was merged to stackforge/rally: Adding sla_check to gate jobs  https://review.openstack.org/111347
23:13 openstackgerrit Boris Pavlovic proposed a change to stackforge/rally: Remove magic() method from context manager  https://review.openstack.org/104962

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