13:00:22 #startmeeting senlin 13:00:23 Meeting started Tue Sep 26 13:00:22 2017 UTC and is due to finish in 60 minutes. The chair is ruijie_. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:24 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:26 The meeting name has been set to 'senlin' 13:00:50 evening :) 13:01:09 hi, ruijie 13:01:21 hi XueFeng 13:01:30 HELLO 13:01:40 hi Qiming 13:01:56 hello, Qiming 13:02:55 let's get started :) 13:03:04 the tempest plugin 13:03:06 https://review.openstack.org/#/c/506550/ 13:03:07 patch 506550 - governance - Add senlin-tempest-plugin to project list 13:03:16 this patch is under reviewed 13:04:01 another one is abuot the lb policy: https://review.openstack.org/#/c/504548/ 13:04:01 patch 504548 - senlin - Add code to fix the problem that support existed l... (MERGED) 13:04:11 sorry I merged it myself .. 13:04:43 it's fine, it looks okay to me 13:04:51 do you still have some concerns? 13:05:14 yes, since that improvement could be proposed later 13:05:51 another thing is the schedule of the meetup 13:05:54 :) 13:07:42 so, the 2nd weekend of Oct is a little bit early, I guess 13:12:20 about that patch, https://review.openstack.org/#/c/504548/2/senlin/policies/lb_policy.py, can we just ignore the error when attaching? 13:12:20 patch 504548 - senlin - Add code to fix the problem that support existed l... (MERGED) 13:12:27 Hi, ruijie_, how about the main works this cycle? 13:12:37 I replied on the mailinglist, I won't be able to join you if it is scheduled on Oct 14/15 13:12:45 I will be traveling to Wuhan 13:13:30 yes Qiming, I saw the e-mail, the schedule is flexable, so that we all can join it 13:14:14 XieFeng, actually, the meetup is aim to discuss features 13:14:46 ok 13:15:21 may need to talk to each other, we have options 13:15:44 if we cannot get most of up together, then we can use zoom or something to have a virtual meeting 13:15:51 up/us 13:16:16 one week later would be good for me 13:16:18 Better to together 13:16:44 yes Qiming, XueFeng, I did't get all the replies 13:19:26 okay, Qiming, XueFeng, I am going to ping everyone to check the schedule and opinions :) 13:19:34 thanks 13:19:52 that's my job :) 13:20:24 about the patch we just talked about, I received e-mail from Tingting Yu 13:20:45 https://review.openstack.org/#/c/504548/ 13:20:45 patch 504548 - senlin - Add code to fix the problem that support existed l... (MERGED) 13:21:21 the patch itself is fine, about the TODO in commonts, that is not easy to roll back 13:22:03 yes 13:22:31 some nodes successfully joined the LB pool, others failed 13:23:03 yes, need to erase the lb info in node.data either 13:23:45 can we just ignore the error, since we support cluster_recover in lb policy 13:30:05 we may need a design on how to handle such kinds of partial failures 13:32:26 em.. that may be fine we have 10 nodes in the cluster and in the attaching process, 9 success, 1 fail 13:33:02 it may be and may be not okay 13:33:10 really depends on the usage scenario 13:36:07 but we can recover the failed nodes later 13:37:53 we do not know whether the lb service went down, or just network problems, but we can be optimistic and provide methods to recover it? 13:41:58 will think about it latter :) Qiming, XueFeng: that's all from my side 13:42:23 yes, it makes sense 13:42:59 however, I am not quite sure cluster_recover operation can cover these failure scenarios 13:43:08 we can talk about it duing meetup 13:43:35 yes Qiming, then we can improve the cluster_recover operation :) 13:43:40 sure 13:46:33 Qiming, XueFeng, do we have other topics to share 13:47:22 no 13:49:16 thanks for joining, Gentlemen :) 13:49:36 we can end it a little bit early :) 13:49:50 #endmeeting