08:30:03 <anac1> #startmeeting Yardstick work meeting 08:30:03 <collabot> Meeting started Mon Nov 23 08:30:03 2015 UTC. The chair is anac1. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:30:03 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic. 08:30:03 <collabot> The meeting name has been set to 'yardstick_work_meeting' 08:30:09 <anac1> #info Ana Cunha 08:30:34 <QiLiang> #info QiLiang 08:30:42 <kubi> #info kubi 08:31:26 <anac1> #topic kvm 08:31:31 <rexlee1> #info Rex 08:31:56 <anac1> qiliang: any further discussions with kvm? 08:31:57 <QiLiang> kvm'guy will call me tomorrow 08:32:07 <anac1> ok, thanks 08:32:34 <anac1> #info follow up on kvm topic next meeting 08:32:41 <anac1> #topic ha 08:33:11 <anac1> any news on ha? the new patch replaces the previous? 08:34:04 <QiLiang> yingming did some refactor 08:34:11 <QiLiang> not done yet 08:34:25 <anac1> ok, does it replace the previous patch? 08:34:55 <anac1> #info refactoring work on-going 08:35:23 <anac1> #topic openstack kilo 08:35:49 <anac1> #link https://gerrit.opnfv.org/gerrit/#/c/3507/ 08:36:15 <anac1> #info patch for making yardstick work on kilo 08:36:41 <anac1> have you tried kilo in Huawei lab? 08:37:01 <anac1> with yardstick, i mean 08:37:16 <patrick1> #info Patrick 08:37:19 <QiLiang> tried but failed 08:37:20 <patrick1> sorry for late 08:37:44 <anac1> patcick1: no problem 08:37:45 <QiLiang> heat stack create failed 08:38:06 <anac1> check jnon's patch, we need to test it 08:38:23 <QiLiang> ok thanks. 08:38:43 <anac1> #topic yardstick dashboard 08:39:09 <anac1> we agreed to follow up today, anyone has managed to try influxdB? 08:39:59 <anac1> jnon did, i think, but he is sick today 08:40:30 <anac1> do we need more time to investigate ? 08:40:51 <QiLiang> ok 08:41:02 <anac1> #info need more time to investigate, follow up next meeting 08:41:12 <anac1> #topic ipvsix 08:41:37 <anac1> anyone participated in the ipvsix meeting last week? 08:42:01 <QiLiang> i've short meeting with binhu last friday 08:42:29 <QiLiang> they have a ping6 test case 08:42:33 <anac1> #link https://wiki.opnfv.org/ipv6_opnfv_project/minutes_20151120 08:42:53 <anac1> yes, do you know what changes they mean in yardstick? 08:43:33 <anac1> if you know, they great ! 08:43:39 <anac1> then great ! 08:43:40 <QiLiang> installer or yardstick need some setup to support ipv6 08:44:00 <QiLiang> because ipv6 have some bug in odl 08:44:15 <anac1> ok, is the suggestion to try with compass? 08:44:51 <anac1> fuel will not take any new requests for b-release 08:44:59 <QiLiang> yes firstly try ipv6 test use compass installer 08:45:20 <anac1> ok 08:46:06 <anac1> we follow up the status next week ? 08:46:33 <QiLiang> ok 08:46:40 <anac1> #info a trial with compass and ipv6 08:46:55 <anac1> #info follow up status next week 08:47:12 <anac1> #topic other 08:47:36 <anac1> #link https://wiki.opnfv.org/releases/brahmaputra 08:48:17 <anac1> #info no b-release project meeting was held last week 08:48:58 <anac1> anything else ? 08:49:13 <QiLiang> i'm ok 08:49:28 <YiminWang> ping anac1. sorry for late. I'm refacting the ha code. It needs about a more week and will replace the previous. 08:49:58 <anac1> ok, thanks. did you get comments from ha on the test cases? 08:51:36 <anac1> on the test results to testing dashboard, we need to decide what test cases we will publish there 08:51:51 <YiminWang> yes, I comucated with patrick 08:52:22 <anac1> thanks, YiminWang 08:52:44 <anac1> #action all add comments on https://etherpad.opnfv.org/p/yardstick_visualization 08:53:00 <QiLiang> only 5 graphs per test project can dashboard display? 08:53:12 <YiminWang> thanks for your focus 08:53:49 <anac1> 5 is the current assumption 08:54:20 <anac1> that's why we need yardstick dashboard, as we have more than 5 08:55:42 <anac1> #info the test&performance meeting, usually on thursdays, will add a second slot on wednesdays 8h30 UTC 08:55:50 <QiLiang> that's a littler difficult to select :) 08:56:03 <anac1> yes, i think very difficult to select 08:56:36 <anac1> we maybe take 1 compute, 1 storage, 1 network ? 08:57:16 <QiLiang> agree 08:58:16 <kubi> so we need to decide which tools result to report for each scenario ? 08:59:11 <anac1> yes, it's time we decide what result we want and which tool to report 09:00:00 <kubi> we may need some tools which can give a score 09:00:20 <anac1> a score, meaning average ? 09:01:08 <rexlee1> i think we may select one typical case from each category if we can only choose 5 09:01:30 <kubi> for example ,for compute scenario ,we can use unixbench to give a score 09:01:54 <anac1> kubi, ok, i agree, let's add that to jira? 09:02:10 <kubi> ok 09:02:19 <anac1> great 09:02:37 <kubi> i 'd like to try it 09:02:46 <anac1> rexlee1: yes, we need to select one typical for each 09:02:55 <anac1> kubi: great, thanks ! 09:04:09 <anac1> that's all for today, thanks everyone 09:04:13 <anac1> #endmeeting