| *** ykatabam has joined #openstack-meeting | 00:16 | |
| *** jamesmcarthur has joined #openstack-meeting | 00:45 | |
| *** longkb has joined #openstack-meeting | 00:46 | |
| *** jamesmcarthur has quit IRC | 00:58 | |
| *** jamesmcarthur has joined #openstack-meeting | 01:04 | |
| *** tetsuro has joined #openstack-meeting | 01:18 | |
| *** imacdonn has quit IRC | 01:23 | |
| *** felipemonteiro has joined #openstack-meeting | 01:23 | |
| *** imacdonn has joined #openstack-meeting | 01:23 | |
| *** bobh has quit IRC | 01:24 | |
| *** ykatabam has quit IRC | 01:26 | |
| *** jamesmcarthur has quit IRC | 01:43 | |
| *** bobh has joined #openstack-meeting | 01:43 | |
| *** mhen has quit IRC | 01:44 | |
| *** mhen has joined #openstack-meeting | 01:46 | |
| *** bobh has quit IRC | 01:47 | |
| *** bobh has joined #openstack-meeting | 01:51 | |
| *** jamesmcarthur has joined #openstack-meeting | 01:58 | |
| *** jamesmcarthur has quit IRC | 02:03 | |
| *** felipemonteiro has quit IRC | 02:17 | |
| *** hongbin has joined #openstack-meeting | 02:19 | |
| *** felipemonteiro has joined #openstack-meeting | 02:31 | |
| *** felipemonteiro has quit IRC | 02:31 | |
| *** hongbin has quit IRC | 02:35 | |
| *** hongbin has joined #openstack-meeting | 02:35 | |
| *** bobh has quit IRC | 02:37 | |
| *** lhinds has quit IRC | 02:37 | |
| *** haleyb has quit IRC | 02:37 | |
| *** bobh has joined #openstack-meeting | 02:41 | |
| *** hongbin_ has joined #openstack-meeting | 02:41 | |
| *** hongbin has quit IRC | 02:43 | |
| *** psachin has joined #openstack-meeting | 02:54 | |
| *** truongnh___ has joined #openstack-meeting | 03:10 | |
| truongnh___ | Could anyone help me to list what metrics are using in Live Migration of Masakari, please | 03:11 |
|---|---|---|
| *** bobh has quit IRC | 03:18 | |
| *** bobh has joined #openstack-meeting | 03:20 | |
| *** bobh has quit IRC | 03:25 | |
| *** felipemonteiro has joined #openstack-meeting | 03:32 | |
| *** ekcs has quit IRC | 03:37 | |
| *** ekcs has joined #openstack-meeting | 03:37 | |
| *** jamesmcarthur has joined #openstack-meeting | 03:41 | |
| *** ekcs has quit IRC | 03:45 | |
| *** jamesmcarthur has quit IRC | 03:46 | |
| *** truongnh___ has quit IRC | 03:48 | |
| *** tetsuro has quit IRC | 03:57 | |
| *** ykatabam has joined #openstack-meeting | 04:11 | |
| *** hongbin_ has quit IRC | 04:22 | |
| *** hongbin has joined #openstack-meeting | 04:24 | |
| *** hongbin has quit IRC | 04:38 | |
| *** ykatabam has quit IRC | 04:52 | |
| *** ykatabam has joined #openstack-meeting | 04:53 | |
| *** akhil_jain has left #openstack-meeting | 05:01 | |
| *** vishalmanchanda has quit IRC | 05:10 | |
| *** vishalmanchanda has joined #openstack-meeting | 05:10 | |
| *** apetrich has quit IRC | 05:23 | |
| *** felipemonteiro has quit IRC | 05:37 | |
| *** apetrich has joined #openstack-meeting | 05:39 | |
| *** Luzi has joined #openstack-meeting | 05:52 | |
| *** pcaruana has joined #openstack-meeting | 06:21 | |
| *** psachin has quit IRC | 06:29 | |
| *** janki has joined #openstack-meeting | 06:30 | |
| *** bzhao__ has joined #openstack-meeting | 06:42 | |
| *** jbadiapa has joined #openstack-meeting | 06:42 | |
| *** a-pugachev has joined #openstack-meeting | 06:44 | |
| *** chkumar|off is now known as chandankumar | 06:45 | |
| *** e0ne has joined #openstack-meeting | 06:45 | |
| *** yamamoto has quit IRC | 06:49 | |
| *** yamamoto has joined #openstack-meeting | 06:49 | |
| *** yamamoto has quit IRC | 06:53 | |
| *** yamamoto has joined #openstack-meeting | 06:53 | |
| *** yamamoto has quit IRC | 06:55 | |
| *** yamamoto has joined #openstack-meeting | 06:58 | |
| *** pcaruana has quit IRC | 06:58 | |
| *** rcernin has quit IRC | 06:59 | |
| *** yamamoto has quit IRC | 07:01 | |
| *** yamamoto_ has joined #openstack-meeting | 07:01 | |
| *** rcernin has joined #openstack-meeting | 07:05 | |
| *** ykatabam has quit IRC | 07:06 | |
| *** rcernin has quit IRC | 07:07 | |
| *** pcaruana has joined #openstack-meeting | 07:13 | |
| *** longkb has quit IRC | 07:29 | |
| *** longkb has joined #openstack-meeting | 07:29 | |
| *** ralonsoh has joined #openstack-meeting | 07:29 | |
| *** jamesmcarthur has joined #openstack-meeting | 07:42 | |
| *** jamesmcarthur has quit IRC | 07:46 | |
| *** lpetrut has joined #openstack-meeting | 07:50 | |
| *** tetsuro has joined #openstack-meeting | 08:11 | |
| *** tetsuro has quit IRC | 08:12 | |
| *** tetsuro_ has joined #openstack-meeting | 08:12 | |
| *** finucannot is now known as stephenfin | 08:20 | |
| *** ttsiouts has joined #openstack-meeting | 08:25 | |
| *** ttsiouts has quit IRC | 08:29 | |
| *** bobh has joined #openstack-meeting | 08:45 | |
| *** yamamoto_ has quit IRC | 08:46 | |
| *** yamamoto has joined #openstack-meeting | 08:46 | |
| *** bobh has quit IRC | 08:50 | |
| *** ykatabam has joined #openstack-meeting | 08:50 | |
| *** nguyenhai has quit IRC | 08:52 | |
| *** nguyenhai has joined #openstack-meeting | 08:53 | |
| *** yamamoto_ has joined #openstack-meeting | 08:57 | |
| *** yamamoto has quit IRC | 09:00 | |
| *** yamamoto_ has quit IRC | 09:01 | |
| *** yamamoto has joined #openstack-meeting | 09:01 | |
| *** lpetrut has quit IRC | 09:06 | |
| *** slaweq has joined #openstack-meeting | 09:08 | |
| *** ttsiouts has joined #openstack-meeting | 09:17 | |
| *** gary_perkins has joined #openstack-meeting | 09:20 | |
| *** ttsiouts has quit IRC | 09:23 | |
| *** electrofelix has joined #openstack-meeting | 09:27 | |
| *** sambetts|afk is now known as sambetts | 09:30 | |
| *** ociuhandu has joined #openstack-meeting | 09:39 | |
| *** ociuhandu has quit IRC | 09:40 | |
| *** ttsiouts has joined #openstack-meeting | 09:43 | |
| *** bnemec has joined #openstack-meeting | 09:45 | |
| *** yamamoto has quit IRC | 09:46 | |
| *** yamamoto has joined #openstack-meeting | 09:47 | |
| *** yamamoto has quit IRC | 09:52 | |
| *** bnemec has quit IRC | 09:53 | |
| *** psachin has joined #openstack-meeting | 09:56 | |
| *** psachin has quit IRC | 10:06 | |
| *** rfolco|rover has joined #openstack-meeting | 10:14 | |
| *** yamamoto has joined #openstack-meeting | 10:15 | |
| *** ykatabam has quit IRC | 10:20 | |
| *** aojea has joined #openstack-meeting | 10:55 | |
| *** bobh has joined #openstack-meeting | 10:57 | |
| *** bobh has quit IRC | 11:02 | |
| *** erlon has joined #openstack-meeting | 11:04 | |
| *** erlon has quit IRC | 11:11 | |
| *** erlon has joined #openstack-meeting | 11:14 | |
| *** priteau has joined #openstack-meeting | 11:19 | |
| *** markvoelker has joined #openstack-meeting | 11:22 | |
| *** pcaruana has quit IRC | 11:26 | |
| *** markvoelker has quit IRC | 11:26 | |
| *** bobh has joined #openstack-meeting | 11:27 | |
| *** pcaruana has joined #openstack-meeting | 11:48 | |
| *** longkb has quit IRC | 11:56 | |
| *** aojeagarcia has joined #openstack-meeting | 11:59 | |
| *** aojea has quit IRC | 12:01 | |
| *** aojea has joined #openstack-meeting | 12:02 | |
| *** aojeagarcia has quit IRC | 12:05 | |
| *** kopecmartin has quit IRC | 12:08 | |
| *** aojea has quit IRC | 12:09 | |
| *** janki has quit IRC | 12:10 | |
| *** kopecmartin has joined #openstack-meeting | 12:11 | |
| *** pcaruana has quit IRC | 12:12 | |
| *** pcaruana has joined #openstack-meeting | 12:12 | |
| *** efried_pto is now known as efried | 12:14 | |
| *** raildo has joined #openstack-meeting | 12:14 | |
| *** mjturek has joined #openstack-meeting | 12:15 | |
| *** bobh has quit IRC | 12:16 | |
| *** lhinds has joined #openstack-meeting | 12:18 | |
| *** smcginnis_vaca is now known as smcginnis | 12:26 | |
| *** jaypipes has joined #openstack-meeting | 12:27 | |
| *** tetsuro_ has quit IRC | 12:33 | |
| *** tpsilva has joined #openstack-meeting | 12:38 | |
| *** jesusaur has quit IRC | 12:46 | |
| *** kopecmartin has quit IRC | 12:47 | |
| *** yamamoto has quit IRC | 12:50 | |
| *** yamamoto has joined #openstack-meeting | 12:50 | |
| *** bobh has joined #openstack-meeting | 12:52 | |
| *** jroll has quit IRC | 12:53 | |
| *** jroll has joined #openstack-meeting | 12:54 | |
| *** yamamoto has quit IRC | 12:55 | |
| *** kopecmartin has joined #openstack-meeting | 12:55 | |
| *** edleafe has joined #openstack-meeting | 12:55 | |
| *** yamamoto has joined #openstack-meeting | 13:06 | |
| *** weshay_pto has quit IRC | 13:07 | |
| *** mriedem has joined #openstack-meeting | 13:11 | |
| *** cloudrancher has quit IRC | 13:12 | |
| *** tetsuro has joined #openstack-meeting | 13:12 | |
| *** cloudrancher has joined #openstack-meeting | 13:13 | |
| *** jesusaur has joined #openstack-meeting | 13:15 | |
| *** jamesmcarthur has joined #openstack-meeting | 13:17 | |
| *** a-pugachev has quit IRC | 13:26 | |
| *** eharney has joined #openstack-meeting | 13:27 | |
| *** rubasov has joined #openstack-meeting | 13:28 | |
| *** munimeha1 has joined #openstack-meeting | 13:33 | |
| *** awaugama has joined #openstack-meeting | 13:36 | |
| *** elbragstad is now known as lbragstad | 13:39 | |
| *** dustins has joined #openstack-meeting | 13:41 | |
| *** felipemonteiro has joined #openstack-meeting | 13:48 | |
| *** liuyulong has joined #openstack-meeting | 13:56 | |
| *** liuyulong is now known as liuyulong|away | 13:57 | |
| *** yamamoto has quit IRC | 13:58 | |
| *** yamamoto has joined #openstack-meeting | 13:59 | |
| *** lpetrut has joined #openstack-meeting | 14:03 | |
| *** yamamoto has quit IRC | 14:04 | |
| *** tidwellr has joined #openstack-meeting | 14:05 | |
| *** Luzi has quit IRC | 14:05 | |
| *** SteelyDan is now known as dansmith | 14:11 | |
| *** haleyb has joined #openstack-meeting | 14:12 | |
| *** a-pugachev has joined #openstack-meeting | 14:18 | |
| *** dustins has quit IRC | 14:20 | |
| *** dustins has joined #openstack-meeting | 14:20 | |
| *** erlon has quit IRC | 14:21 | |
| *** yamamoto has joined #openstack-meeting | 14:41 | |
| *** jamesmcarthur has quit IRC | 14:41 | |
| *** hongbin has joined #openstack-meeting | 14:43 | |
| *** tetsuro has quit IRC | 14:44 | |
| *** jamesmcarthur has joined #openstack-meeting | 14:46 | |
| *** caboucha has joined #openstack-meeting | 14:47 | |
| *** jamesmcarthur has quit IRC | 14:55 | |
| *** jamesmcarthur has joined #openstack-meeting | 14:55 | |
| *** felipemonteiro has quit IRC | 15:02 | |
| *** david-lyle has joined #openstack-meeting | 15:07 | |
| *** zaneb has quit IRC | 15:10 | |
| *** pcaruana has quit IRC | 15:12 | |
| *** e0ne has quit IRC | 15:15 | |
| *** zaneb has joined #openstack-meeting | 15:17 | |
| *** a-pugachev has quit IRC | 15:18 | |
| *** a-pugachev has joined #openstack-meeting | 15:18 | |
| *** e0ne has joined #openstack-meeting | 15:28 | |
| *** e0ne has quit IRC | 15:32 | |
| *** manjeets has joined #openstack-meeting | 15:40 | |
| *** finix has joined #openstack-meeting | 15:45 | |
| *** tovin07 has joined #openstack-meeting | 15:47 | |
| *** jamesmcarthur has quit IRC | 15:47 | |
| *** bnemec has joined #openstack-meeting | 15:51 | |
| *** ircuser-1 has joined #openstack-meeting | 15:53 | |
| *** finix has quit IRC | 15:54 | |
| *** bnemec has quit IRC | 15:59 | |
| *** mlavalle has joined #openstack-meeting | 16:00 | |
| mlavalle | #startmeeting neutron_performance | 16:00 |
| openstack | mlavalle: Error: Can't start another meeting, one is in progress. Use #endmeeting first. | 16:00 |
| *** gyee has joined #openstack-meeting | 16:00 | |
| *** njohnston has joined #openstack-meeting | 16:01 | |
| mlavalle | hi, will the Congress meeting be over soon? | 16:01 |
| njohnston | o/ | 16:01 |
| rubasov | o/ | 16:01 |
| clarkb | mlavalle: anyone can stop a meeting after one hour. If you issue #endmeeting it should stop the existing meeting | 16:01 |
| rubasov | mlavalle: there was not much said before | 16:02 |
| mlavalle | clarkb: yhanks. I know. I was just trying to be polite | 16:02 |
| rubasov | mlavalle: so I think they accidentally left it open | 16:02 |
| mlavalle | #endmeeting | 16:02 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/" | 16:02 | |
| openstack | Meeting ended Mon Oct 22 16:02:35 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:02 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/congressteammeeting/2018/congressteammeeting.2018-10-19-04.01.html | 16:02 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/congressteammeeting/2018/congressteammeeting.2018-10-19-04.01.txt | 16:02 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/congressteammeeting/2018/congressteammeeting.2018-10-19-04.01.log.html | 16:02 |
| mlavalle | #startmeeting neutron_performance | 16:02 |
| openstack | Meeting started Mon Oct 22 16:02:45 2018 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:02 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:02 |
| *** openstack changes topic to " (Meeting topic: neutron_performance)" | 16:02 | |
| openstack | The meeting name has been set to 'neutron_performance' | 16:02 |
| slaweq | hi | 16:02 |
| njohnston | hello | 16:03 |
| mlavalle | njohnston, slaweq, rubasov: thanks for attending | 16:03 |
| rubasov | hello again | 16:03 |
| clarkb | mlavalle: ah well they stopped ~12 hours ago so I don't know anyone is still around to do it for you :) | 16:04 |
| mlavalle | #topic Discussion | 16:05 |
| *** openstack changes topic to "Discussion (Meeting topic: neutron_performance)" | 16:05 | |
| mlavalle | Eearlier today I sent each one of you (and others) a message with html files | 16:05 |
| mlavalle | did everyone see it? | 16:05 |
| slaweq | I'm just looking at them now | 16:05 |
| rubasov | mlavalle: yes, I'm looking too | 16:06 |
| rubasov | mlavalle: could you please give access to the add-router-interface.html too? | 16:06 |
| rubasov | mlavalle: for some reason I can't open it | 16:06 |
| slaweq | yes, I don't have access to this one also | 16:06 |
| *** david-lyle has quit IRC | 16:07 | |
| mlavalle | sure I'll resend it soon | 16:07 |
| mlavalle | in the meantime, let me make a few general statements about them | 16:08 |
| mlavalle | if you want to deploy a test environment with devstack enabling osprofiler, you just need to add two lines to your local.conf file: | 16:09 |
| mlavalle | enable_plugin osprofiler https://git.openstack.org/openstack/osprofiler | 16:09 |
| mlavalle | OSPROFILER_COLLECTOR=redis | 16:10 |
| mlavalle | it doesn't have to be redis | 16:10 |
| njohnston | if you use redis, do you have to already have that up and running, or does devstack handle setting it up for you? | 16:10 |
| mlavalle | there are other options. But I don't have any preference and redis seems to be the most used by the odprofiler team | 16:10 |
| mlavalle | njohnston: devstack sets it up for you | 16:11 |
| njohnston | nice | 16:11 |
| mlavalle | It most be part of what the osprofiler plugin for devstack does for you | 16:11 |
| mlavalle | when stack.sh is done, you will see the following section added to neutron.conf: http://paste.openstack.org/show/732600/ | 16:13 |
| mlavalle | Please note that with this setup, you enable tracing in general and sqlalchemy tracing | 16:13 |
| *** dustins has quit IRC | 16:14 | |
| mlavalle | and as you can see, the 'SECRET_KEY' is the default | 16:14 |
| rubasov | side note: and similar config should also be added to all other openstack component config files when devstack is done | 16:14 |
| slaweq | have it got big performance impact on running neutron server? if not maybe we could enable it in gate jobs? | 16:14 |
| mlavalle | rubasov: you got it | 16:14 |
| rubasov | slaweq: the docs say it shouldn't | 16:15 |
| mlavalle | slaweq: I cannot see any diffrence in performance | 16:15 |
| mlavalle | whatsoever | 16:15 |
| njohnston | +1 for adding it to the CI | 16:15 |
| slaweq | ok, and how to get such report from the API call? is it generated for each call then? | 16:16 |
| rubasov | slaweq: it's integrated with osc | 16:16 |
| rubasov | it's like this: openstack ... --os-profile SECRET_KEY | 16:17 |
| mlavalle | this is an example of a command execution along with the generation of the corresponding html file: http://paste.openstack.org/show/732601/ | 16:17 |
| rubasov | then a trace is generated for that call | 16:17 |
| rubasov | and osc gives back an id of the trace | 16:17 |
| mlavalle | Note that "Trace ID: b723c5c7-f87e-49e3-a51e-f1071a00c433" is added to the output of the command | 16:17 |
| mlavalle | and you use that in the osprofiler command to generate the html file | 16:18 |
| rubasov | eg: osprofiler trace show --html ... --connection-string redis://localhost:6379 | 16:18 |
| mlavalle | makes sene? | 16:19 |
| slaweq | yep | 16:19 |
| slaweq | but this means that we can't easily integrate it with tempest jobs | 16:20 |
| slaweq | right? | 16:20 |
| mlavalle | well, let me explain | 16:20 |
| mlavalle | apparently it is integrated with Rally.... | 16:20 |
| slaweq | ok, sorry for disturbing then. Go on mlavalle :) | 16:21 |
| rubasov | yes as mlavalle said, the test runner must know about osprofiler, and rally seems to know that | 16:21 |
| mlavalle | these early sessions of this meeting are more brainstroming than anythig else | 16:21 |
| tovin07 | slaweq, for tempest job, you can see this patch https://review.openstack.org/#/c/523935/ | 16:22 |
| mlavalle | #link https://docs.openstack.org/rally/latest/quick_start/tutorial/step_10_profiling_openstack_internals.html | 16:22 |
| tovin07 | #link Add profiler support into Tempest https://review.openstack.org/#/c/523935/ | 16:22 |
| slaweq | tovin07: thx, I will check that | 16:23 |
| mlavalle | tovin07: ahhh, that's so cool. Is this patch something that will be merged soon? or we take the patch and depploy it? | 16:23 |
| rubasov | that's nice | 16:23 |
| rubasov | I'm working on the other half - enabling osprofiler in a tempest gate check job: https://review.openstack.org/611835 | 16:24 |
| mlavalle | that's a good step forward rubaso | 16:26 |
| *** yamahata_ has joined #openstack-meeting | 16:26 | |
| mlavalle | rubasov: | 16:26 |
| mlavalle | I spent ometime over the weekend looking at results of our rally jobs | 16:26 |
| mlavalle | and it doesn't seem that we are generating the traces | 16:26 |
| rubasov | last friday I thought ps3 was working already, then today I realized I misread the logs | 16:26 |
| rubasov | but I'm progressing slowly as the gate gives me a chance once every few hours | 16:27 |
| mlavalle | but if we were, according to the rally docs we would generate a osprofiler id for each request | 16:27 |
| rubasov | mlavalle: yes, it's only enabling osprofiler in devstack, not tracing any call yet | 16:27 |
| tovin07 | mlavalle, it depends on tempest team | 16:28 |
| mlavalle | and then we could retrieve the file generated and generate htnl analsis | 16:28 |
| mlavalle | tovin07: thanks :-) | 16:28 |
| tovin07 | I think the patch is good to me for now | 16:28 |
| rubasov | mlavalle: exactly | 16:29 |
| mlavalle | so I think one action item would be to create a rally job with osprofiler enabled | 16:29 |
| tovin07 | if you guys feel plain html output from osprofiler is boring | 16:29 |
| *** ttsiouts has quit IRC | 16:30 | |
| tovin07 | you can use jeager as backend instead of redis | 16:30 |
| slaweq | mlavalle: I can take care of this rally job | 16:30 |
| *** erlon has joined #openstack-meeting | 16:30 | |
| mlavalle | I think it should be a periodic job | 16:30 |
| tovin07 | like this one for a simple glance command (sorry, it's not related to neutron :D ) | 16:30 |
| tovin07 | http://137.116.141.93:16686/trace/92dbc0064374abfc | 16:30 |
| mlavalle | nice tovin07 ! | 16:31 |
| mlavalle | what do other think of a periodic job? | 16:31 |
| slaweq | mlavalle: periodic rally job with enabled osprofiler? | 16:32 |
| mlavalle | yeah | 16:32 |
| mlavalle | I just don't want to have too many jobs in the check queue | 16:32 |
| tovin07 | for rally job, we can look at this patch, hope that it can help https://review.openstack.org/#/c/467252/ | 16:32 |
| slaweq | would it be a problem to just enable it for neutron-rally-task job which we have now in check queue? | 16:32 |
| njohnston | let's put it in experimental to start, so we can trigger it when we want it | 16:32 |
| njohnston | slaweq++ | 16:33 |
| mlavalle | ok, experimental sounds good | 16:33 |
| rubasov | mlavalle: at first it likely can't vote anyway, so it makes sense to make it periodical instead of check | 16:33 |
| mlavalle | and no, I don see a problem using neutron-rally-task as a base | 16:34 |
| mlavalle | so.... | 16:34 |
| mlavalle | agree? | 16:34 |
| mlavalle | with the experimental job | 16:34 |
| slaweq | mlavalle: but I was asking if we need new job for that | 16:34 |
| mlavalle | slaweq: let's not make that decision today | 16:34 |
| mlavalle | I want to see how much effect it has first | 16:34 |
| rubasov | maybe stupid question: how do I trigger an experimental job? | 16:34 |
| slaweq | ok, it can be experimental job at the beginning | 16:35 |
| slaweq | rubasov: You write a comment "check experimental" in gerrit | 16:35 |
| rubasov | slaweq: thanks | 16:35 |
| slaweq | ok, I have one more question about this rally experimental job then | 16:35 |
| mlavalle | #action slaweq to create a experimental rally job with osprofiler enabled | 16:36 |
| slaweq | from commit message in https://review.openstack.org/#/c/467252/ and from rally docs it looks that this integration "just" put profiler id to each api call | 16:36 |
| *** ttsiouts has joined #openstack-meeting | 16:36 | |
| mlavalle | yes | 16:36 |
| slaweq | how You think we should then genearate reports from each call in job? | 16:36 |
| mlavalle | that's my understanding | 16:36 |
| slaweq | should we add some script which will do it in post-run? | 16:36 |
| mlavalle | my understanding is that we should be able to download the generated database and do the analysis locally | 16:37 |
| mlavalle | but it is an assumption | 16:37 |
| slaweq | ok, I will investigate that | 16:37 |
| slaweq | and ask infra if will be necessary | 16:38 |
| mlavalle | I would be looking at the person enabling the job to help us answer this | 16:38 |
| mlavalle | i.e. slaweq | 16:38 |
| mlavalle | :-) | 16:38 |
| slaweq | :) | 16:38 |
| rubasov | mlavalle: you mean the whole tracepoint db, right? | 16:38 |
| mlavalle | rubasov: yeah | 16:38 |
| mlavalle | this is the process I forsee: | 16:39 |
| mlavalle | 1) We run the rally job | 16:39 |
| rubasov | we may want to log the trace ids somewhere | 16:39 |
| slaweq | rubasov: trace ids will be stored in rally result page for each iteration of each test | 16:39 |
| mlavalle | 2) We identify problematci requests from the rally report | 16:39 |
| *** ttsiouts has quit IRC | 16:39 | |
| rubasov | slaweq: cool | 16:39 |
| mlavalle | 3) with the trace ids associated with the problematic requests, we analyze from the traces db | 16:40 |
| mlavalle | makes sense? | 16:40 |
| mlavalle | again, that is what I am assuming based on what I know so far | 16:40 |
| slaweq | mlavalle: I think so, let's test that in practise :) | 16:40 |
| rubasov | it does to me | 16:41 |
| *** kopecmartin is now known as kopecmartin|off | 16:41 | |
| mlavalle | The other think is that someone should look at the requests that we are currently testing in our Rally job.... | 16:41 |
| mlavalle | the problematic ones that slaweq identified in his original analysis are not all there | 16:42 |
| mlavalle | #link http://paste.openstack.org/show/731683/ | 16:42 |
| mlavalle | #link http://paste.openstack.org/show/731684/ | 16:42 |
| mlavalle | any volunteer to do that? | 16:43 |
| rubasov | mlavalle: I can take that | 16:43 |
| mlavalle | rubasov: thanks. I think the action item is to see what is missing and start adding some of them | 16:44 |
| rubasov | mlavalle: ack | 16:44 |
| mlavalle | Let's focus on the top 10 offenders in slaweq's analysis | 16:44 |
| mlavalle | #action rubasov to add top request offnders to rally job | 16:45 |
| slaweq | maybe we can (for now) skip first one and focus on 2-10th :) | 16:45 |
| mlavalle | slaweq: absolutely. good point | 16:45 |
| rubasov | slaweq: of course | 16:45 |
| mlavalle | LOL | 16:45 |
| *** zaneb has quit IRC | 16:45 | |
| slaweq | :) | 16:45 |
| njohnston | :-) | 16:45 |
| mlavalle | The other think that draw my attention looking at the html file is that we are duplicating or triplicating the information for DB related traces | 16:46 |
| *** zaneb has joined #openstack-meeting | 16:46 | |
| mlavalle | that is due to the fact that we enable the tracing both in the neutron-lib and the neutron DB api modules | 16:46 |
| mlavalle | so if you look at the each DB query, it shows up several times | 16:47 |
| mlavalle | I can take an action item to experiment and simplify that | 16:47 |
| *** ekcs has joined #openstack-meeting | 16:47 | |
| mlavalle | agree? | 16:47 |
| njohnston | sounds good | 16:48 |
| rubasov | mlavalle: yes | 16:48 |
| *** ekcs has joined #openstack-meeting | 16:48 | |
| *** ekcs has quit IRC | 16:48 | |
| njohnston | let me know if you'd like any help with that | 16:48 |
| mlavalle | #action mlavalle to simplify the DB traces | 16:48 |
| mlavalle | njohnston: will do | 16:48 |
| *** ekcs has joined #openstack-meeting | 16:48 | |
| *** efried has quit IRC | 16:49 | |
| mlavalle | I have one final point related to the traces | 16:49 |
| *** efried has joined #openstack-meeting | 16:49 | |
| mlavalle | if you look for example at the subnet creation ones, the largest amount of time is spent with the dhcp server rpc | 16:50 |
| mlavalle | about two thirds | 16:50 |
| mlavalle | and even in other requests where the relative weight is not that much (for example router create), you will see the dhcp rcp time is significant | 16:50 |
| mlavalle | even the largest individually | 16:51 |
| slaweq | yes, I also noticed that now | 16:51 |
| mlavalle | I am not drawing any conclusions just yet about that | 16:51 |
| mlavalle | becuase I tested without concurrent load | 16:51 |
| mlavalle | I just wanted to draw your attention to the fact | 16:52 |
| slaweq | I also found in router-create call that there is a lot of "SELECT 1" calls there | 16:52 |
| *** e0ne has joined #openstack-meeting | 16:52 | |
| slaweq | do You know what it is maybe? | 16:52 |
| mlavalle | nope | 16:52 |
| rubasov | my guess was some sqlalchemy connection check | 16:52 |
| rubasov | but that's just a guess | 16:52 |
| njohnston | yes, I believe rubasov is correct | 16:52 |
| slaweq | there is e.g. such call which took 50 ms which is quite significant comapring to other calls | 16:52 |
| mlavalle | interesting | 16:53 |
| mlavalle | ok | 16:53 |
| mlavalle | I think we've had a very productive meeting today :-) | 16:53 |
| mlavalle | anything else we should discuss? | 16:53 |
| tovin07 | about Berlin summit | 16:54 |
| rubasov | one more note about traces I noticed | 16:54 |
| tovin07 | will anyone join? | 16:54 |
| mlavalle | tovin07: I'll be there | 16:54 |
| mlavalle | and rubasov as well | 16:54 |
| rubasov | if you trace nova too the SECRET_KEY can leak into periodic jobs | 16:54 |
| rubasov | just ignore that if the trace is suspiciously long | 16:54 |
| mlavalle | rubasov: nice point! thanks | 16:55 |
| tovin07 | mlavalle, see you there :D | 16:55 |
| rubasov | tovin07: yep, as mlavalle said | 16:55 |
| mlavalle | tovin07: maybe we can pick your brains in Berlin, if we have questions :_) | 16:55 |
| tovin07 | yep, hope that i can help | 16:56 |
| mlavalle | :-) | 16:56 |
| mlavalle | tovin07: this meeting is bi-weekly. you are more than welcome to contribute | 16:56 |
| mlavalle | ok team, thanks for your attendance | 16:57 |
| rubasov | thank you guys | 16:57 |
| mlavalle | #endmeeting | 16:57 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/" | 16:57 | |
| openstack | Meeting ended Mon Oct 22 16:57:12 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:57 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/neutron_performance/2018/neutron_performance.2018-10-22-16.02.html | 16:57 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/neutron_performance/2018/neutron_performance.2018-10-22-16.02.txt | 16:57 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/neutron_performance/2018/neutron_performance.2018-10-22-16.02.log.html | 16:57 |
| slaweq | thx, o/ | 16:57 |
| rubasov | bye | 16:57 |
| njohnston | o/ | 16:58 |
| *** mlavalle has left #openstack-meeting | 16:58 | |
| *** dklyle has joined #openstack-meeting | 17:02 | |
| *** a-pugachev has quit IRC | 17:03 | |
| *** erlon has quit IRC | 17:03 | |
| *** tovin07 has quit IRC | 17:04 | |
| *** eharney has quit IRC | 17:04 | |
| *** ekcs has quit IRC | 17:05 | |
| *** e0ne has quit IRC | 17:06 | |
| *** dustins has joined #openstack-meeting | 17:09 | |
| *** felipemonteiro has joined #openstack-meeting | 17:11 | |
| *** rubasov has left #openstack-meeting | 17:12 | |
| *** armax has joined #openstack-meeting | 17:12 | |
| *** raildo has quit IRC | 17:14 | |
| *** felipemonteiro has quit IRC | 17:15 | |
| *** eharney has joined #openstack-meeting | 17:15 | |
| *** jamesmcarthur has joined #openstack-meeting | 17:18 | |
| *** sambetts is now known as sambetts|afk | 17:18 | |
| *** bnemec has joined #openstack-meeting | 17:19 | |
| *** jamesmcarthur has quit IRC | 17:22 | |
| *** pcaruana has joined #openstack-meeting | 17:27 | |
| *** yamahata_ has quit IRC | 17:28 | |
| *** e0ne has joined #openstack-meeting | 17:30 | |
| *** irclogbot_3 has joined #openstack-meeting | 17:34 | |
| *** raildo has joined #openstack-meeting | 17:37 | |
| *** ralonsoh has quit IRC | 17:42 | |
| *** raildo has quit IRC | 17:43 | |
| *** shilpasd_ has joined #openstack-meeting | 17:44 | |
| *** jamesmcarthur has joined #openstack-meeting | 17:44 | |
| *** electrofelix has quit IRC | 17:44 | |
| *** diablo_rojo has joined #openstack-meeting | 17:47 | |
| *** jamesmcarthur_ has joined #openstack-meeting | 17:47 | |
| *** jamesmcarthur has quit IRC | 17:48 | |
| *** shilpasd_ has quit IRC | 17:48 | |
| *** yamahata_ has joined #openstack-meeting | 17:50 | |
| *** irclogbot_3 has quit IRC | 17:56 | |
| *** pcaruana has quit IRC | 17:57 | |
| *** ttsiouts has joined #openstack-meeting | 17:58 | |
| *** raildo has joined #openstack-meeting | 18:01 | |
| *** irclogbot_3 has joined #openstack-meeting | 18:06 | |
| *** raildo has quit IRC | 18:11 | |
| *** apetrich has quit IRC | 18:17 | |
| *** e0ne has quit IRC | 18:24 | |
| *** cloudrancher has quit IRC | 18:29 | |
| *** cloudrancher has joined #openstack-meeting | 18:30 | |
| *** apetrich has joined #openstack-meeting | 18:30 | |
| *** yamahata_ has quit IRC | 18:51 | |
| *** flaper87 has quit IRC | 18:51 | |
| *** flaper87 has joined #openstack-meeting | 18:51 | |
| *** jamesmcarthur_ has quit IRC | 18:54 | |
| *** lpetrut has quit IRC | 18:56 | |
| *** jamesmcarthur has joined #openstack-meeting | 18:56 | |
| *** ttsiouts has quit IRC | 19:01 | |
| *** yamahata has joined #openstack-meeting | 19:02 | |
| *** kopecmartin|off has quit IRC | 19:03 | |
| *** jamesmcarthur has quit IRC | 19:06 | |
| *** jamesmcarthur has joined #openstack-meeting | 19:06 | |
| *** jamesmcarthur has quit IRC | 19:11 | |
| *** irclogbot_3 has quit IRC | 19:14 | |
| *** irclogbot_3 has joined #openstack-meeting | 19:15 | |
| *** ttsiouts has joined #openstack-meeting | 19:18 | |
| *** jamesmcarthur has joined #openstack-meeting | 19:32 | |
| *** jamesmcarthur has quit IRC | 19:36 | |
| *** awaugama has quit IRC | 19:42 | |
| *** felipemonteiro has joined #openstack-meeting | 19:44 | |
| *** jamesmcarthur has joined #openstack-meeting | 19:47 | |
| *** felipemonteiro has quit IRC | 19:47 | |
| *** jamesmcarthur has quit IRC | 19:51 | |
| *** cloudrancher has quit IRC | 19:53 | |
| *** cloudrancher has joined #openstack-meeting | 19:54 | |
| *** ttsiouts has quit IRC | 20:04 | |
| *** ttsiouts has joined #openstack-meeting | 20:05 | |
| *** ttsiouts has quit IRC | 20:09 | |
| *** irclogbot_3 has quit IRC | 20:27 | |
| *** felipemonteiro has joined #openstack-meeting | 20:28 | |
| *** ttsiouts has joined #openstack-meeting | 20:39 | |
| *** ttsiouts has quit IRC | 20:43 | |
| *** zaneb has quit IRC | 20:45 | |
| *** caboucha has quit IRC | 20:52 | |
| *** ttsiouts has joined #openstack-meeting | 21:01 | |
| *** cloudrancher has quit IRC | 21:05 | |
| *** cloudrancher has joined #openstack-meeting | 21:06 | |
| *** kopecmartin has joined #openstack-meeting | 21:07 | |
| *** bobh has quit IRC | 21:08 | |
| *** bobh has joined #openstack-meeting | 21:09 | |
| *** bobh has quit IRC | 21:13 | |
| *** armax has quit IRC | 21:14 | |
| *** zaneb has joined #openstack-meeting | 21:21 | |
| *** slaweq has quit IRC | 21:24 | |
| *** zaneb has quit IRC | 21:27 | |
| *** dustins has quit IRC | 21:30 | |
| *** stevebaker has quit IRC | 21:31 | |
| *** dmellado has quit IRC | 21:32 | |
| *** gouthamr has quit IRC | 21:32 | |
| *** slaweq has joined #openstack-meeting | 21:41 | |
| *** slaweq has quit IRC | 21:45 | |
| *** priteau has quit IRC | 21:49 | |
| *** armax has joined #openstack-meeting | 21:51 | |
| *** mriedem is now known as mriedem_away | 22:07 | |
| *** bnemec has quit IRC | 22:18 | |
| *** felipemonteiro has quit IRC | 22:44 | |
| *** ykatabam has joined #openstack-meeting | 22:45 | |
| *** rcernin has joined #openstack-meeting | 22:46 | |
| *** cloudrancher has quit IRC | 22:51 | |
| *** cloudrancher has joined #openstack-meeting | 22:53 | |
| *** gyee has quit IRC | 23:00 | |
| *** tpsilva has quit IRC | 23:01 | |
| *** eharney has quit IRC | 23:08 | |
| *** hongbin has quit IRC | 23:17 | |
| *** zaneb has joined #openstack-meeting | 23:23 | |
| *** zaneb has quit IRC | 23:23 | |
| *** zaneb has joined #openstack-meeting | 23:24 | |
| *** mriedem_away has quit IRC | 23:27 | |
| *** rcernin_ has joined #openstack-meeting | 23:28 | |
| *** rcernin has quit IRC | 23:30 | |
| *** liuyulong|away has quit IRC | 23:31 | |
| *** dklyle has quit IRC | 23:32 | |
| *** david-lyle has joined #openstack-meeting | 23:32 | |
| *** david-lyle has quit IRC | 23:35 | |
| *** dklyle has joined #openstack-meeting | 23:35 | |
| *** ssbarnea has quit IRC | 23:40 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!