Earlier  
Later
New Search
Posted Nick Remark
#openstack-cyborg - 2017-08-02
16:42:55 jkilpatr zhipengh[m], that being said it's probably best to test for 'prod' setups along instead of being devstack only until everything works there only to see it all break again when we try to deploy for real
16:43:29 zhipeng yes I agree with that
#openstack-cyborg - 2017-08-03
12:24:24 dims zhipengh[m] : are you considering a tag/branch for pike for cyborg?
12:42:01 zhipengh[m] dims I'm definitely interested, although not familiar with the release engineering process
12:43:33 dims zhipengh[m] : most of the steps are documented here - https://docs.openstack.org/infra/manual/creators.html
12:45:44 dims zhipengh[m] : there's more stuff on this page - https://docs.openstack.org/infra/manual/drivers.html#tagging-a-release
12:46:18 dims see "Tagging a Release" for example
12:46:46 zhipengh[m] THX ! :)
12:49:19 dims zhipengh[m] : please join #openstack-release channel and we can talk more there if you have any questions
12:50:00 dims zhipengh[m] : since both cyborg and mogan are in the same boat, i've asked zhenguo to ping you as well
12:50:24 zhipengh[m] Haha ok
12:53:11 zhipengh[m] Looks similar to what we've been doing in OPNFV for release
12:54:10 zhipengh[m] dims since cyborg is not official, what would be the benefit to cut a release ?
12:55:22 dims zhipengh[m] : it's part of the process to become official, you can tell folks that there's a tag/branch for use with a specific openstack release if they wish to try it
12:55:47 zhipengh[m] Ah right
12:56:09 zhipengh[m] I remember Joe did this for tricircle
12:56:41 dims when it comes time to submit for governance, we can show that we have been putting out releases already and possibly picked up some operators/users
13:00:20 zhipengh[m] That is a great point
13:05:46 zhipeng okey I think the meetbot should capture dims's conversation on tag/branch
13:05:51 zhipeng #endmeeting
13:05:53 openstack Meeting ended Thu Aug 3 13:05:51 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
13:05:54 openstack Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-02-15.06.html
13:05:55 openstack Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-02-15.06.txt
13:05:56 openstack Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-02-15.06.log.html
13:07:46 zhipeng dims we have the longest openstack team meeting XD
13:11:17 dims LOL, i did not even know that a meeting was on :)
13:12:41 zhipengh[m] A little ambushy huh :P
13:21:42 dims LOL
#openstack-cyborg - 2017-08-09
15:01:14 crushil \o
15:01:22 zhipeng o/
15:01:30 zhipeng #startmeeting
15:01:37 openstack zhipeng: Error: A meeting name is required, e.g., '#startmeeting Marketing Committee'
15:01:47 zhipeng #startmeeting openstack-cyborg
15:01:49 openstack Meeting started Wed Aug 9 15:01:47 2017 UTC and is due to finish in 60 minutes. The chair is zhipeng. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:50 openstack Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:53 openstack The meeting name has been set to 'openstack_cyborg'
15:04:13 zhipeng anyone else is here for the meeting ?
15:05:15 zhuli I'am here
15:05:34 zhipeng hey zhuli :)
15:05:38 zhipeng thx for the new patches
15:05:42 zhipeng looks awesome
15:05:52 zhipeng #topic patch discussion
15:05:58 zhipeng #link https://review.openstack.org/#/q/status:open+project:openstack/cyborg
15:06:25 zhipeng zhuli could you go over the two patches you submitted ?
15:08:09 zhuli sure, one is for api which add some hook to cyborg
15:08:49 zhuli the other is WIP, aim to go through the api -> conductor -> db
15:10:07 zhipeng so about the second patch, what is the relationship between it and Justin's patch on conductor ?
15:10:21 zhuli after that, I'll add some unittest as well
15:13:12 zhipeng sorry got disconnected ...
15:13:28 zhipeng zhuli did you answer my previous question ?
15:13:39 zhuli hmm, justin's conductor patch is good for functionality but make all conductor related stuff together which I think should be refactored to be more modular
15:14:17 zhipeng modular on conductor itself ?
15:15:47 zhuli because we need a separate conductor service
15:16:13 zhuli modular on the whole cyborg project
15:16:48 zhipeng okey got it
15:17:04 zhipeng so the second patch would be a followup of Justin's patch , right ?
15:17:56 zhuli yes
15:18:07 zhipeng great :)
15:18:24 zhipeng crushil do you have any updates ?
15:18:33 crushil zhipeng, Nope
15:18:57 zhipeng your upcoming patch depends on the current unmerged Justin patch ?
15:19:42 crushil Yup
15:19:57 crushil Have some customer engagements going on right now and after that will get back working on the next driver patch
15:20:08 zhipeng okey
15:20:22 zhipeng if you got time plz review Justin's patch and help merge it
15:20:32 zhipeng I already gave my +2
15:20:36 crushil zhipeng, Will do
15:20:41 zhipeng thx :)
15:20:51 zhipeng #topic release branching/tagging
15:21:18 zhipeng so last week dims from Technical Committee made a suggestion that we should consider cut a pike release branch
15:21:53 zhipeng as part of the evidence/effort when we apply for the official project status
15:22:14 zhipeng do you guys have any thoughts on the criteria for the release cutting ?
15:22:34 zhipeng like up to what point we could consider having a release branch out
15:22:48 zhipeng since we are not official at the moment, we have more freedom on the timing
15:25:45 rushil I think we should cut a release in queen as we don't have anything to cut
15:27:03 zhuli second that
15:36:44 zhipeng rushil I don't think we should expect anything functional for pike
15:36:54 zhipeng but a basic framework still a release ?
15:38:37 rushil The problem is customers would assume that if it's in a "stable" branch, it's ready to be played with
15:39:30 rushil But if it still makes sense to you, we can still do it
16:00:20 ttk2[m] I have to agree on keeping any release Branch for Queens when we can at least provide basic functionality.
16:00:29 ttk2[m] Sorry always getting pulled into meetings lately.
16:01:09 ttk2[m] Crushil what do you need from my db commit? And I'll have to look at zhuli's patch to see what he needs conductor wise.
16:02:57 zhipeng ok then I think we could cut a branch when most of the basic stuff are there around PTG time frame, but we don't announce it
16:03:06 zhipeng so just as something to show for the TC
16:03:19 crushil ok
16:05:37 ttk2[m] +1
16:06:31 zhipeng #endmeeting
16:06:33 openstack Meeting ended Wed Aug 9 16:06:31 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
16:06:34 openstack Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-09-15.01.html
16:06:35 openstack Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-09-15.01.txt
16:06:37 openstack Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-08-09-15.01.log.html
#openstack-cyborg - 2017-08-11
11:55:25 jkilpatr zhuli, wow big commit. Nice!
11:55:51 jkilpatr you seem to understand oslo db better than I do, I couldn't find a good explanation for exactly what it does on top of sqlalchemy
11:57:40 jkilpatr zhuli, let me know when you think it's ready to play with, I'm going to be working on automated deployment / installer / testing I think all the major components have something in place to try using.
16:24:22 zhuli jkilpatr: I'll call you the moment I finish the dbsync stuff, then it's ready to play with.
#openstack-cyborg - 2017-08-16
03:11:23 zhuli jkilpatr: I have proposed this patch https://review.openstack.org/#/c/492134/, which pulls together the api,conductor,db processes to implement a basic workflow, I think you can start your work on it
03:12:01 zhuli jkilpatr: btw, please help review the patch :)
03:13:02 zhipengh[m] zhuli plz also send an email to the mailing list
03:13:18 zhuli ok

Earlier   Later