| *** rh-jelabarre has quit IRC | 00:18 | |
| *** rh-jelabarre has joined #openstack-rally | 00:23 | |
| *** rh-jelabarre has quit IRC | 03:38 | |
| *** agopi_ has joined #openstack-rally | 03:54 | |
| *** agopi has quit IRC | 03:54 | |
| *** agopi_ is now known as agop | 05:30 | |
| *** agop is now known as agopi | 05:30 | |
| *** rcernin has quit IRC | 06:43 | |
| *** e0ne has joined #openstack-rally | 07:10 | |
| *** e0ne has quit IRC | 07:12 | |
| *** e0ne has joined #openstack-rally | 07:13 | |
| *** e0ne has quit IRC | 07:14 | |
| *** e0ne has joined #openstack-rally | 07:15 | |
| *** e0ne has quit IRC | 07:17 | |
| *** agopi_ has joined #openstack-rally | 07:56 | |
| *** tosky has joined #openstack-rally | 07:58 | |
| *** agopi has quit IRC | 07:59 | |
| *** agopi_ is now known as agopi | 08:04 | |
| *** agopi_ has joined #openstack-rally | 08:11 | |
| *** agopi has quit IRC | 08:14 | |
| *** agopi_ is now known as agopi | 08:27 | |
| *** e0ne has joined #openstack-rally | 10:12 | |
| *** rh-jelabarre has joined #openstack-rally | 12:39 | |
| *** e0ne has quit IRC | 12:47 | |
| *** dave-mccowan has joined #openstack-rally | 12:53 | |
| *** dave-mccowan has quit IRC | 13:01 | |
| *** zul has quit IRC | 13:26 | |
| *** zul has joined #openstack-rally | 13:26 | |
| *** aojea_ has joined #openstack-rally | 13:45 | |
| *** slaweq has joined #openstack-rally | 13:48 | |
| slaweq | andreykurilin: hi | 13:48 |
|---|---|---|
| slaweq | andreykurilin: do You remember my attempts to integrate neutron rally job with osprofiler? | 13:49 |
| andreykurilin | slaweq: hey! yes :) | 13:49 |
| slaweq | I recently switched to jaeger backend: https://review.openstack.org/#/c/615350/ | 13:49 |
| slaweq | and it is better as there is no timeout | 13:49 |
| slaweq | but osprofiler details for each test and iteration are empty: http://logs.openstack.org/50/615350/21/check/neutron-rally-task/fb2480e/results/report.html.gz#/NeutronNetworks.create_and_delete_networks/output | 13:50 |
| slaweq | do You know maybe why it can be like that? or should I ask people from osprofiler? | 13:50 |
| andreykurilin | checking.. | 13:51 |
| andreykurilin | slaweq: hm... It wound be nice if we can print any stats from jaeger directly without interacting with rally&osprofiler to ensure that the data is saved there.. | 14:03 |
| slaweq | andreykurilin: I'm affraid that there is nothing like that stored in job logs | 14:05 |
| slaweq | I can later try to deploy local devstack with osprofiler and check where it will be stored | 14:06 |
| slaweq | andreykurilin: I will back to You if I will have something | 14:06 |
| slaweq | andreykurilin: I have also second question - about rally and python 3 | 14:06 |
| *** mvkr has quit IRC | 14:07 | |
| slaweq | I just tried to switch it here: https://review.openstack.org/#/c/624358/ | 14:07 |
| slaweq | and it failed for me with some "strange" error | 14:07 |
| *** dave-mccowan has joined #openstack-rally | 14:08 | |
| andreykurilin | slaweq: based on https://github.com/openstack/osprofiler/commit/0b878cfd24aef2ff4604c697bf39339c84accc78 jaeger should be deployed on docker. I expect that jaeger should have something like 'dump' command that we can add to ci for debugging purpose | 14:08 |
| slaweq | andreykurilin: should I expect that rally isn't ready to be run with python 3? I don't think so but | 14:08 |
| andreykurilin | checking | 14:09 |
| andreykurilin | slaweq: Rally was one of the first project who started checking python 3. When 'python 3' community wide goal started, rally was used for a lot of projects to start gating python3 because tempest was not ready. Guess, I mixed something while creatig new 'Zuul V3' jobs | 14:11 |
| slaweq | andreykurilin: I suppose that it should works but I just wanted to be sure :) | 14:12 |
| *** dave-mccowan has quit IRC | 14:14 | |
| *** aojea_ has quit IRC | 14:14 | |
| andreykurilin | slaweq: here is an issue - https://github.com/openstack/rally-openstack/blob/master/tests/ci/playbooks/roles/snapshot-available-os-resources/tasks/main.yaml#L4 wring call for `python` binary. | 14:15 |
| andreykurilin | slaweq: will propose a fix in several minutes | 14:18 |
| slaweq | andreykurilin: You are fast, thx a lot :) | 14:18 |
| andreykurilin | slaweq: dims proposed a fix for the similar issue long time ago when zuul v2 was our ci system. it was easy to filter commits and find what I missed while porting to zuul v3 :) | 14:21 |
| slaweq | andreykurilin: easy for You, I would spent probably a week debugging it :D | 14:22 |
| slaweq | thx once again | 14:22 |
| andreykurilin | slaweq: thank you for reporting the issue :) | 14:22 |
| *** openstackgerrit has joined #openstack-rally | 14:26 | |
| openstackgerrit | Andrey Kurilin proposed openstack/rally-openstack master: [ci] Use proper python binary depending on USE_PYTHON3 var https://review.openstack.org/624409 | 14:26 |
| slaweq | andreykurilin: thx, I just made my patch as depends on Yours ^^ to check if it will be fine now :) | 14:29 |
| openstackgerrit | Andrey Kurilin proposed openstack/rally-openstack master: [ci] Use proper python binary depending on USE_PYTHON3 var https://review.openstack.org/624409 | 14:30 |
| andreykurilin | slaweq: please recheck it :) there is a new patch-set :) | 14:30 |
| slaweq | sure, done | 14:32 |
| *** mvkr has joined #openstack-rally | 14:44 | |
| *** e0ne has joined #openstack-rally | 14:45 | |
| andreykurilin | slaweq: it looks like we cannot use jaeger at CI | 14:48 |
| andreykurilin | slaweq: osprofiler is able just write traces there, but unable to fetch them back. | 14:49 |
| andreykurilin | https://github.com/openstack/osprofiler/blob/master/osprofiler/drivers/jaeger.py#L108-L118 | 14:49 |
| slaweq | andreykurilin: yep :/ | 14:50 |
| andreykurilin | but I think we can copy-paste the logic of enabling jaeger (using docker) and make it possible to use elasticsearch, for example. First of all, es driver for osprofiler can fetch data. The second thing, it is possible to write some debugging script to dump data from elastic directly | 14:53 |
| andreykurilin | https://github.com/openstack/osprofiler/commit/0b878cfd24aef2ff4604c697bf39339c84accc78 | 14:53 |
| andreykurilin | let me try to do it quickly | 14:55 |
| andreykurilin | slaweq, funny fact: osprofiler was under Rally program long time ago and than moved to Oslo :) | 14:56 |
| *** itlinux has quit IRC | 15:06 | |
| andreykurilin | slaweq: https://review.openstack.org/624421 | 15:14 |
| openstackgerrit | Chuck Short proposed openstack/rally-openstack master: Add additional loadbalancer scenarios for octavia https://review.openstack.org/620997 | 15:19 |
| *** agopi has quit IRC | 15:24 | |
| slaweq | andreykurilin: thx, I just commented it :) | 15:25 |
| andreykurilin | slaweq: oh, I missed that you are enabling mongo driver. ok. we have a chance to check all of them :) | 15:29 |
| *** agopi has joined #openstack-rally | 15:29 | |
| *** agopi has quit IRC | 15:35 | |
| *** aojea_ has joined #openstack-rally | 15:36 | |
| slaweq | andreykurilin: yes, but mongo didn't help | 15:56 |
| andreykurilin | slaweq: it was slow? | 15:56 |
| slaweq | yep | 15:56 |
| openstackgerrit | Andrey Kurilin proposed openstack/rally-openstack master: [ci] Use proper python binary depending on USE_PYTHON3 var https://review.openstack.org/624409 | 15:59 |
| *** itlinux has joined #openstack-rally | 16:22 | |
| *** e0ne has quit IRC | 16:36 | |
| *** rallydev-bot has quit IRC | 16:53 | |
| andreykurilin | Since our gitter-irc proxy had issues. reposting missing messages manually | 16:54 |
| andreykurilin | Ilya Shakhat @shakhat 18:49 | 16:54 |
| andreykurilin | @andreykurilin hi! looks like it's hard to make osprofiler work... | 16:54 |
| andreykurilin | how does Rally get report from osprofiler? via CLI or via internal API? | 16:54 |
| *** rallydev-bot has joined #openstack-rally | 16:55 | |
| rallydev-bot | [From Gitter] andreykurilin : @shakhat hey. maybe it is not so easy, but I like this case because several projects are trying to solve a single issue :) | 16:55 |
| rallydev-bot | [From Gitter] andreykurilin : @shakhat here is a code of making osprofiler's code - https://github.com/openstack/rally-openstack/blob/master/rally_openstack/embedcharts/osprofilerchart.py#L43-L73 | 16:56 |
| rallydev-bot | [From Gitter] andreykurilin : *osprofiler's reports | 16:56 |
| *** itlinux_ has joined #openstack-rally | 16:59 | |
| *** aojea_ has quit IRC | 17:02 | |
| *** aojea_ has joined #openstack-rally | 17:03 | |
| *** itlinux has quit IRC | 17:03 | |
| *** aojea_ has quit IRC | 17:07 | |
| *** zul has quit IRC | 17:20 | |
| *** mvkr has quit IRC | 18:04 | |
| *** e0ne has joined #openstack-rally | 18:10 | |
| *** aojea has joined #openstack-rally | 18:56 | |
| openstackgerrit | Chuck Short proposed openstack/rally-openstack master: Add additional loadbalancer scenarios for octavia https://review.openstack.org/620997 | 19:32 |
| *** zul has joined #openstack-rally | 20:04 | |
| *** mvkr has joined #openstack-rally | 21:18 | |
| *** aojea has quit IRC | 21:19 | |
| *** e0ne has quit IRC | 21:47 | |
| *** itlinux_ has quit IRC | 21:57 | |
| *** rcernin has joined #openstack-rally | 21:59 | |
| *** itlinux has joined #openstack-rally | 22:26 | |
| *** slaweq has quit IRC | 22:26 | |
| *** itlinux has quit IRC | 22:45 | |
| *** slaweq has joined #openstack-rally | 23:15 | |
| *** slaweq has quit IRC | 23:19 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!