Perl 6 - the future is here, just unevenly distributed

IRC log for #openstack-rally, 2014-08-02

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

All times shown according to UTC.

Time Nick Message
20:59 ilbot3 joined #openstack-rally
20:59 Topic for #openstack-rally is now ☁ Rally IRC chat  ☁ Key persons to ask:  boris-42, rediskin, andreykurilin, k4n0  ☁ 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://wiki.openstack.org/wiki/Rally/Updates
20:59 openstackgerrit kumar rishabh proposed a change to stackforge/rally: Adding sla_check to gate jobs  https://review.openstack.org/111347
21:01 openstackgerrit OpenStack Proposal Bot proposed a change to stackforge/rally: Updated from global requirements  https://review.openstack.org/111530
21:20 openstackgerrit kumar rishabh proposed a change to stackforge/rally: Adding sla_check to gate jobs  https://review.openstack.org/111347
21:25 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  ☁ 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://wiki.openstack.org/wiki/Rally/Updates
21:29 boris-42 ping
21:29 boris-42 penguinRaider__ ^ now our chat is logged=)
21:38 lordd joined #openstack-rally
21:38 lordd joined #openstack-rally
21:44 lordd joined #openstack-rally
21:45 penguinRaider__ ping
21:45 penguinRaider__ topic?
21:47 penguinRaider__ boris-42, cool :-) this is really helpful but the bot seems dumb :-|
21:49 boris-42 penguinRaider__ yep it doesn't reply =)
21:49 boris-42 penguinRaider__ just listen us=)
21:52 boris-42 penguinRaider__ heh I will ask you one more thing
21:54 penguinRaider__ boris-42, NeutronNetworks.create_and_list_ports    | 0   | max_failure_percent | False   failed again :-|
21:55 boris-42 penguinRaider__ btw could you use tee command
21:55 boris-42 penguinRaider__ to write to file as well results
21:55 boris-42 penguinRaider__https://review.openstack.org/#/c/111347/4/tests_ci/rally-gate.sh
21:56 boris-42 penguinRaider__ file should be rally-plot/sla.txt
21:56 boris-42 penguinRaider__  Quota exceeded for resources: ['port']
21:58 boris-42 penguinRaider__ ok seems you are close to success
21:59 penguinRaider__ boris-42, okay :-) what does -1 yaml signify ?
21:59 penguinRaider__ boris-42, like network : -1?
21:59 boris-42 penguinRaider__ it set's neutron quotas for tenant to infinity
22:00 boris-42 penguinRaider__ network amount of networks
22:00 boris-42 penguinRaider__ port amount of ports and so on
22:00 penguinRaider__ boris-42, ah I see I thought almost same so port : -1 should work too?
22:00 boris-42 penguinRaider__ ?
22:01 boris-42 penguinRaider__ list on neutron quotas https://github.com/stackforge/rally/blob/master/rally/benchmark/context/quotas.py#L132-L163
22:01 boris-42 list of**
22:02 penguinRaider__ boris-42, Quota exceeded for resources: ['port'] I thought of adding port:-1 to the quota
22:02 boris-42 penguinRaider__ yep that will help
22:02 boris-42 penguinRaider__ that is actually what I said on review
22:03 penguinRaider__ boris-42,  okay :-)
22:07 openstackgerrit kumar rishabh proposed a change to stackforge/rally: Adding sla_check to gate jobs  https://review.openstack.org/111347
22:10 yingjun joined #openstack-rally
22:12 boris-42 tzabal ping
22:12 tzabal boris-42 pong
22:12 boris-42 tzabal did you work on required_context validator?
22:13 tzabal boris-42 ahm no because i think it works without that
22:13 boris-42 tzabal hmm
22:13 boris-42 tzabal how?
22:14 boris-42 tzabal what will happen if user forgot to specify blogbench context?
22:14 tzabal boris-42 the user doesnt need to specify the blogbench context in the task configuration file
22:15 tzabal boris-42 because it uses one that uses the benchmark scenario of blogbench
22:15 tzabal boris-42 https://review.openstack.org/#/c/97030/18/doc/samples/tasks/scenarios/vm/blogbench.json
22:15 tzabal boris-42 as we have in doc the json/yaml of other scenarios pre-made
22:16 tzabal boris-42 this will be like them; the user will just need to change image, flavor, fixed_network etc
22:16 tzabal boris-42 like in boot_runcommand_delete
22:16 tzabal boris-42 but now insidethe context, not in args
22:17 tzabal boris-42 so as you proposed about inheritance in context, i did a similar thing for benchmark scenarios
22:17 boris-42 tzabal one more time
22:17 boris-42 tzabal what will happen
22:17 boris-42 tzabal if I remove blogbench
22:17 boris-42 tzabal from this https://review.openstack.org/#/c/97030/18/doc/samples/tasks/scenarios/vm/blogbench.json
22:17 boris-42 tzabal file
22:17 boris-42 tzabal and run task
22:17 tzabal boris-42 i haven't tried that
22:18 tzabal boris-42 but i assume that it will fail
22:18 boris-42 tzabal so that's why you need require_context
22:18 tzabal boris-42 ok
22:19 boris-42 tzabal validation.required_context("blogbench")
22:19 boris-42 tzabal on top of scenario
22:19 boris-42 tzabal will check that user specified blogbench context
22:19 boris-42 tzabal if not it will fail before running any of benchmark in task
22:20 boris-42 tzabal saying you know, you didn't specify required benchmark
22:20 tzabal boris-42 ok, i will add it
22:20 boris-42 tzabal can you do that in separated patch?
22:20 tzabal boris-42 :)
22:20 tzabal boris-42 yes sure
22:21 boris-42 tzabal it's required by other patch
22:21 boris-42 tzabal from other guy
22:21 tzabal boris-42 ah ok then
22:21 boris-42 tzabal so we will get it in fast
22:21 boris-42 tzabal and you'll continue work on your stuff
22:21 boris-42 tzabal here is required service https://github.com/stackforge/rally/blob/master/rally/benchmark/validation.py#L359
22:21 tzabal boris-42 ok, so i will not need to add in the dependency hell of my 3 patches in gerrit :P
22:22 boris-42 tzabal =)
22:22 boris-42 tzabal btw chat is know logged
22:22 boris-42 tzabal http://irclog.perlgeek.de/openstack-rally
22:22 tzabal boris-42 good
22:22 boris-42 tzabal https://review.openstack.org/#/c/98172/12
22:22 boris-42 tzabal hehe=)
22:23 tzabal boris-42 this is what i am talking about
22:23 boris-42 tzabal just get rid of it
22:23 boris-42 tzabal during git rebase -i
22:23 boris-42 tzabal you can remove any of commits from branch
22:23 boris-42 tzabal by removing line during edit
22:24 tzabal boris-42 and what will happen in gerrit?
22:24 boris-42 tzabal so there will be still record
22:24 boris-42 3 records*
22:24 boris-42 tzabal 2 will be in updated and in proper chain
22:24 boris-42 tzabal 3 won't be affected
22:24 boris-42 tzabal 3 one you can just remove
22:24 boris-42 tzabal (abandon)
22:25 boris-42 tzabal gerrit is very smart
22:25 tzabal boris-42 hm and it is written in java
22:25 tzabal boris-42 joking, java is good
22:25 boris-42 tzabal why you even need to know that?
22:26 boris-42 tzabal when it just works
22:26 tzabal boris-42 i was just reading for gerrit
22:26 boris-42 heh
22:26 tzabal boris-42 otherwise i dont really care, i would care only if i had to install it somewhere :P
22:35 tzabal boris-42 required_context will take as parameter a list of strings that represents the name of the required contexts?
22:36 tzabal boris-42 names*
22:41 boris-42 tzabal  I will prefer *args
22:42 tzabal boris-42 ok
22:42 boris-42 tzabal https://github.com/stackforge/rally/blob/master/rally/benchmark/scenarios/nova/servers.py#L83
22:42 boris-42 tzabal like here we have for requier_services
22:46 openstackgerrit Rafi Khardalian proposed a change to stackforge/rally: Add support for Nova live migrations  https://review.openstack.org/111444
22:47 openstackgerrit Rafi Khardalian proposed a change to stackforge/rally: Add support for Nova live migrations  https://review.openstack.org/111444
22:58 boris-42 rmk can you provide unit tests ^
22:58 rmk I dont know how to test these
22:58 boris-42 rmk like others..
22:58 boris-42 rmk with mocking
22:58 boris-42 rmk take a look how other benchmarks are tested
22:59 boris-42 rmk this one for example https://github.com/stackforge/rally/blob/master/tests/benchmark/scenarios/nova/test_servers.py#L183
23:05 rmk Going to ask an honest question, how exactly are these tests useful?  Everything is mocked, we're testing that Python works.
23:07 rmk This seems like coverage for the sake of coverage.
23:13 boris-42 rmk nope
23:13 boris-42 rmk it shows that python works
23:13 boris-42 rmk believe me it's very importnat
23:13 boris-42 rmk in case of you are not writing code on language that is complied
23:13 boris-42 rmk like C
23:14 boris-42 rmk so at least we will know the there is no mistakes in algorithm and code
23:17 boris-42 rmk going to sleep =) see you and hope to see unit tests for your benchmark
23:44 yingjun joined #openstack-rally

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