08:30:15 <kubi001> #startmeeting Yardstick Work Meeting 08:30:15 <collabot`> Meeting started Thu Nov 3 08:30:15 2016 UTC. The chair is kubi001. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:30:15 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic. 08:30:15 <collabot`> The meeting name has been set to 'yardstick_work_meeting' 08:30:40 <kubi001> #topic roll call 08:30:48 <kubi001> #info kubi 08:30:54 <JingLu> #info Jing 08:31:00 <Mingjiang> #info Mingjiang 08:31:12 <chenjiankun> #info Jack 08:31:23 <Kanglin> #info Kanglin 08:31:56 <chenjiankun> #info chenjiankun 08:34:11 <kubi001> #topic action follow up 08:35:29 * kubi001 kubi will create Epics in JIRA with D-planning. 08:36:54 <Mingjiang> https://wiki.opnfv.org/display/yardstick/Yardstick+Danube+Release+Planning 08:36:59 <Mingjiang> #link https://wiki.opnfv.org/display/yardstick/Yardstick+Danube+Release+Planning 08:37:06 <kubi001> Mingjiang: thanks 08:37:28 <Mingjiang> the D-released jira is listed at the front 08:37:40 <kubi001> #info AP1 Done https://wiki.opnfv.org/display/yardstick/Yardstick+Danube+Release+Planning 08:39:00 <kubi001> so please feel free to claim the JIRA task 08:39:51 <kubi001> new epics will be added after discussion 08:40:25 <kubi001> new stories will be added as you like 08:40:47 <kubi001> stories should be belong to one Epics 08:41:22 <kubi001> any question about the Danube Jira? 08:43:05 <kubi001> #topic HA testing 08:43:26 <kubi001> Kanglin: did you atttend HA weekly meeting yesterday? 08:43:41 <zhihui> It is better to manage JIRA task via sprints. We can plan which tasks will be implement in one sprint. 08:44:12 <kubi001> zhihui: sounds good. 08:44:20 <zhihui> sorry for interruption. 08:44:52 <kubi001> it seem that there is need some setting of JIRA 08:44:58 <Mingjiang> zhihui: o 08:45:06 <Kanglin> kubi001: yes. But they discussed about Openstack Sumit and HA Api, finally there's no time left for HA test cases. 08:45:09 <Mingjiang> i think it's a good suggestion 08:45:22 <kubi001> zhihui: I will to knwo how to do it. I knew Functest has the sprints 08:45:56 <Kanglin> So HA test cases discussion is delayed to next HA meeting 08:46:23 <zhihui> Yes, we can borrow it from functest. :) 08:46:31 <kubi001> Kanglin: ok, thanks for feedback 08:46:44 <kubi001> zhihui: good idea 08:46:47 <Mingjiang> kubi: zhihui: i've created a agile board for yardstick in my jira accout called "agile of yardstick", now it has a backlog of danube 08:47:23 <Mingjiang> #link https://jira.opnfv.org/secure/RapidBoard.jspa?rapidView=145&view=planning&selectedIssue=YARDSTICK-19&versions=visible&selectedVersion=10632 08:48:14 <JingLu> looks fantastic 08:48:28 <zhihui> It looks great. 08:48:39 <Mingjiang> still there would be more work to do if we're going to use sprints.. 08:49:05 <kubi001> Mingjiang: cool 08:49:09 <Mingjiang> i think we can discuss more in the next meeting 08:49:27 <zhihui> agree. 08:49:44 <kubi001> Mingjiang: Ok, I will to try it after the meeting and let's discuss it at next meeting 08:50:06 <Mingjiang> kubi001: that's great! 08:50:56 <kubi001> #info HA testing case will be discuss at next HA meeting 08:51:19 <kubi001> #action kubi will learn how to create the sprint and discuss at next meeting 08:52:17 <kubi001> #topic others 08:52:32 <kubi001> zhihui: any update about DPDK testing? 08:52:42 <zhihui> yes 08:52:57 <kubi001> zhihui: I talked with Tapio at this openstack summit 08:53:11 <zhihui> I think the scenario ovs does not works for DPDK for now. 08:53:12 <kubi001> he also run the DPDK test case at local 08:53:29 <zhihui> on which scenario? 08:54:05 <kubi001> I'm not sure, he just said he run the DPDK test case, and It seems that he run it succesfully 08:54:32 <kubi001> and he also mentioned it is not necessary to set the hugepage 08:54:47 <zhihui> Can he share some experience about it ? 08:54:56 <zhihui> It will helpful. 08:55:23 <kubi001> zhihui: yes, I will send email to him and CC to you 08:55:39 <zhihui> kubi, thanks. 08:57:04 <zhihui> I am curious why it is not necessary to set hugepage? 08:57:12 <kubi001> zhihui: thanks for continuous work for DPDK 08:58:50 <zhihui> the scenario ovs bounds the port of compute to DPDK interface. It is different with the test env which I develop dpdk test case. 08:59:14 <zhihui> And with this scenario, the vm is not reachable by floating ip. 08:59:52 <kubi001> zhihui: that would be a problem if the scenario can't support the floating ip 09:01:21 <kubi001> let's check it after the meeting 09:01:22 <JingLu> I think in some fuel OVS scenarios, the floating IP actually works. 09:01:25 <zhihui> I don't know, I posted my problem on ask.opnfv.org. But not reply. 09:02:02 <zhihui> I deployed this scenario on my local env last month. 09:02:49 <zhihui> https://ask.opnfv.org/question/1189/deploy-scenario-os-nosdn-ovs-ha-by-fuel-but-instance-can-not-reachable-by-ip-interface-dpdk0-cannot-allocate-memory/ 09:03:57 <zhihui> I am not sure if it is a hardware configuration issues. 09:04:44 <JingLu> there is a success job Link: https://build.opnfv.org/ci/view/yardstick/job/yardstick-fuel-baremetal-daily-colorado/467/console 09:06:25 <zhihui> Yes, it is successful on lf-pod2, but failed on the other pods, e.g. Ericsson pods 09:07:11 <kubi001> zhihui: so , it means that there is some hareware dependency ? 09:08:03 <zhihui> Link: https://build.opnfv.org/ci/view/yardstick/job/yardstick-fuel-baremetal-daily-colorado/461/console 09:08:56 <zhihui> maybe. I am not sure. 09:10:31 <kubi001> zhihui: Is it possible to run your test case at lf-pod2? 09:12:11 <zhihui> It need test. 09:13:06 <kubi001> zhihui: Ok, you could add the DPDK test case into the suite which is run at lf-pod2 if you need. 09:13:23 <zhihui> Ok, I will consider. 09:13:29 <kubi001> zhihui: thanks 09:13:40 <kubi001> anything else? 09:15:14 <kubi001> thanks all for today 09:15:19 <kubi001> #endmeeting