| *** rook has quit IRC | 00:07 | |
| *** rook has joined #openstack-performance | 00:07 | |
| *** rfolco has joined #openstack-performance | 01:24 | |
| *** rfolco has quit IRC | 01:25 | |
| *** notmyname has quit IRC | 04:36 | |
| *** notmyname has joined #openstack-performance | 04:48 | |
| *** pcaruana has joined #openstack-performance | 06:37 | |
| *** lezbar has quit IRC | 07:26 | |
| *** lezbar has joined #openstack-performance | 07:31 | |
| openstackgerrit | Ilya Shakhat proposed openstack/performance-docs: Extend test plan and results for OpenStack networking https://review.openstack.org/341515 | 09:00 |
|---|---|---|
| *** lezbar has quit IRC | 10:00 | |
| *** lezbar has joined #openstack-performance | 10:15 | |
| rook_ | Cassowary: increase faster? | 11:43 |
| rook_ | so does that mean increase in time or decrease? | 11:43 |
| *** lezbar has quit IRC | 12:27 | |
| *** rfolco has joined #openstack-performance | 12:33 | |
| *** bogdando has quit IRC | 13:01 | |
| *** lezbar has joined #openstack-performance | 13:59 | |
| Cassowary | rook_: as I increase CPU in each step, the time that takes to do the resize increases. This is true for both cases (1-2,2-3,3-4 and 1-2,1-3,1-4,1-5). But in the latter case the rate at which the resize time increase is lower than the former. | 14:03 |
| Cassowary | rook_: This means, the rate at which the resize time increases is higher when you increase VM CPU size by +1 at each step, compared to when you increase the CPU of a VM with 1CPU, +1/+2/+3/+4…..+n. (Note that in the latter case, at each step the VM reduces its size back to 1CPU). | 14:12 |
| *** lezbar has quit IRC | 14:22 | |
| *** catintheroof has joined #openstack-performance | 14:24 | |
| *** lezbar__ has joined #openstack-performance | 14:24 | |
| *** zz_dimtruck is now known as dimtruck | 14:35 | |
| *** linuxgeek_ has joined #openstack-performance | 15:05 | |
| *** linuxaddicts has quit IRC | 15:09 | |
| *** harlowja_at_home has joined #openstack-performance | 15:22 | |
| *** pcaruana has quit IRC | 15:26 | |
| *** harlowja_at_home has quit IRC | 16:54 | |
| *** linuxgeek_ has quit IRC | 17:00 | |
| *** linuxgeek_ has joined #openstack-performance | 17:08 | |
| *** linuxgeek_ has quit IRC | 17:22 | |
| *** pcaruana has joined #openstack-performance | 17:23 | |
| *** harlowja has joined #openstack-performance | 17:43 | |
| *** doffm has joined #openstack-performance | 19:44 | |
| doffm | Is osprofiler already integrated with Rally? Are there any configuraion options I will have to provide Rally to get traces enabled for all Rally calls? | 19:45 |
| *** msimonin has joined #openstack-performance | 19:45 | |
| DinaBelova | doffm not yet, this is still planned to be done | 19:49 |
| DinaBelova | Cassowary btw, do you have any comparison table with n-m stats against the time it takes to resize? | 19:50 |
| DinaBelova | as this sounds very suspicious :) | 19:50 |
| doffm | DinaBelova: What would you reccomend? Turn osprofiler on for all calls, or is there a 'hack' for rally to add the request options? | 19:54 |
| doffm | *recommend, sorry. | 19:54 |
| DinaBelova | doffm, osprofiler work is triggered basically via adding specific headers to the API call - so, this should be done inside rally itself via adding specific HTTP headers to the specific projects calls - like instance boot, etc. Frankly speaking sadly I did not have the opportunity to investigate HOW this can be done easilly. | 19:56 |
| DinaBelova | as even if osprofiler is enabled in the config files, you need to ask profiling explicitly | 19:56 |
| DinaBelova | via CLI or adding these HTTP headers inside python clients | 19:57 |
| DinaBelova | this sounds as a hack into rally, and I'm not sure how to make this happen with less influence on rally code | 19:57 |
| DinaBelova | so I would say there might be a a way to hack into rally code and pass these headers to the OpenStack clients somehow - python-openstackclient supports osprofiler headers passing via CLI, but it needs to be chekced if this can be done via python-client itself (I was out of osprofiler for a while and there were lots of changes done without me, need to get bacl soon) | 20:00 |
| DinaBelova | *back, sorry | 20:00 |
| doffm | Ok, thanks. That gives me a good idea of where to start. | 20:05 |
| *** msimonin1 has joined #openstack-performance | 20:13 | |
| *** msimonin has quit IRC | 20:13 | |
| *** arnoldje has joined #openstack-performance | 20:21 | |
| *** msimonin1 has quit IRC | 20:31 | |
| *** msimonin has joined #openstack-performance | 20:32 | |
| *** msimonin1 has joined #openstack-performance | 20:50 | |
| *** msimonin has quit IRC | 20:50 | |
| *** msimonin1 has quit IRC | 21:05 | |
| *** msimonin has joined #openstack-performance | 21:05 | |
| *** msimonin has quit IRC | 21:06 | |
| *** msimonin1 has joined #openstack-performance | 21:07 | |
| *** msimonin has joined #openstack-performance | 21:08 | |
| *** msimonin1 has quit IRC | 21:08 | |
| *** dimtruck is now known as zz_dimtruck | 21:23 | |
| *** zz_dimtruck is now known as dimtruck | 21:43 | |
| *** msimonin has quit IRC | 21:46 | |
| *** msimonin has joined #openstack-performance | 21:46 | |
| *** arnoldje has quit IRC | 22:00 | |
| *** msimonin1 has joined #openstack-performance | 22:14 | |
| *** msimonin has quit IRC | 22:14 | |
| *** msimonin1 has quit IRC | 22:18 | |
| *** msimonin has joined #openstack-performance | 22:18 | |
| *** msimonin1 has joined #openstack-performance | 22:19 | |
| *** msimonin1 has quit IRC | 22:19 | |
| *** msimonin has quit IRC | 22:19 | |
| *** msimonin2 has joined #openstack-performance | 22:19 | |
| *** bsilverman has joined #openstack-performance | 23:20 | |
| *** arnoldje has joined #openstack-performance | 23:45 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!