Earlier  
Posted Nick Remark
#openstack-cyborg - 2020-04-16
03:02:32 openstack The meeting name has been set to 'openstack_cyborg'
03:02:38 brinzhang welcome to songwenping_
03:03:01 chenke from beijing university?
03:03:04 Sundar Welcome, songwenping_ . Would you like to introduce yourself?
03:03:14 Yumeng Sundar: Good. How's it going with your new job?
03:03:14 chenke welcome songwenping
03:03:25 songwenping_ Hi all
03:03:28 Yumeng welcome songwenping
03:03:45 s_shogo welcome
03:03:55 songwenping_ I am from Inspur.
03:04:15 Sundar Good. Brin has company. :)
03:04:27 songwenping_ do the nova-cyborg-interaction work.
03:04:42 brinzhang songwenping_ mainly work on nova-cyborg interaction, and his take some patch in this meeting
03:04:54 brinzhang Sundar, Yumeng, start meeting?
03:05:04 Sundar Interesting. What aspect of Nova interaction?
03:05:34 xinranwang welcome
03:05:37 Yumeng brinzhang,songwenping_ : Are you planning to use cyborg in production, right?
03:05:40 brinzhang GPU, and others
03:05:58 brinzhang Yumeng, considering
03:05:59 Sundar brinzhang: The meeting has started. We'll get to the patches after hearing a bit more about the Nova interaction.
03:06:22 Sundar Is it about more instance ops like shelve and suspend?
03:07:32 Sundar songwenping_, brinzhang: ^
03:08:04 brinzhang Sundar: now follow you patch https://review.opendev.org/#/c/716185/6/api-guide/source/accelerator-support.rst
03:08:12 songwenping_ not do these operators
03:08:28 xinranwang Sundar: I'd like to know what operation could be add in next release, shelve/unshelve, like what else? And is there any gap here to complete. Because I have started to investigate it and hope add more operations in Nova in next release.
03:09:11 brinzhang Sundar, we also blocked these operations now, maybe we can open that after enough develop
03:09:34 shaohe_feng hi all
03:09:55 Yumeng xinranwang,brinzhang,Sundar and all: here is the etherpad for V release. https://etherpad.opendev.org/p/cyborg-victoria-goals we can add topics now.
03:09:58 brinzhang Sundar: we are still in testing
03:10:40 Sundar xinranwang: The API guide that brinzhang posted above lists the unsupported operations. Many opes there are important for operators. Sean Mooney from Nova is also working on some of these ops for V release, I think. You may want to check with him.
03:11:00 Sundar brinzhang: Good to know. Thanks.
03:11:15 brinzhang Yumeng: sean-k-mooney works on evacuate now
03:11:20 Sundar Yumeng: Thanks for posting the V release etherpad.
03:11:35 brinzhang https://review.opendev.org/#/c/715326/
03:12:04 chenke Hi Sundar, do you mean sean will implement some virtual machine operations in the next version?
03:12:22 xinranwang sean-k-mooney: Is there already some patches from sean mooney? I'd like to have a look.
03:12:28 Sundar chenke: Yes, he says he intends to do that
03:12:28 xinranwang Sundar:
03:12:46 Sundar xinranwang: Please see https://review.opendev.org/#/c/715326/
03:13:15 xinranwang Sundar: thanks.
03:13:15 chenke Okay, this is good news, for both cyborg and nova》
03:13:51 Sundar Yes.Shall we get started on our patches? https://review.opendev.org/#/q/status:open+project:openstack/cyborg+branch:master
03:14:03 Sundar I am reviewing https://review.opendev.org/#/c/696089/
03:14:33 chenke Sure.
03:14:55 brinzhang Sundar https://review.opendev.org/#/c/696089/, how about this patch? need your check
03:15:25 brinzhang except some nits inline, others looks good to me
03:15:36 Sundar brinzhang: Yes, that's what I am looking at. Will complete after this meeting.
03:16:13 Sundar BTW, Release candidate 1 for Ussuri is due Apr 23 https://releases.openstack.org/ussuri/schedule.html
03:16:36 Sundar So, 8 days left. What are the highest priority patches to merge before that?
03:17:11 Yumeng brinzhang: Thanks for posting 696089!
03:17:22 Sundar Programming support https://review.opendev.org/698190 ?
03:18:18 shaohe_feng link: https://etherpad.opendev.org/p/cyborg-victoria-goals
03:18:32 shaohe_feng I have add this program API in the etherpad
03:18:42 shaohe_feng this is really needed in product env to adopt cyborg
03:18:43 Yumeng Sundar: from my side. I think Zuul check bandit failure can be merged before Ussuri is due. https://review.opendev.org/#/q/topic:fix-bandit-check-failures+(status:open+OR+status:merged)
03:18:56 Sundar Yumeng: agreed
03:19:23 Sundar But it is failing the bandit check :)
03:19:32 brinzhang Yumeng, I found the bandit are not be voted, so you should enable it in zuul task
03:19:59 brinzhang and that should be works fine, that we can consider merge it.
03:20:31 brinzhang a questions, what is bandit to do? I am know few of it
03:20:32 Yumeng Sundar, brinzhang: yes. After we fixed all failure, we can set that to voting. and at that time, bandit check will not fail again.
03:21:09 brinzhang IMO, firstly, you should enable this task in zuul task
03:21:29 Sundar brinzhang: Bandit is a tool to detect security flaws. https://pypi.org/project/bandit/
03:21:32 brinzhang than fix the failure, that we can ensure that correctly
03:22:20 brinzhang Sundar: ack, thanks, will see it later
03:22:23 Sundar brinzhang: bandit is now a non-voting job. We need to fix the issues there and then make it voting. Otherwise, other open patches will be affected.
03:22:48 xinranwang https://review.opendev.org/#/c/696089/ The rafactor arq api is also important, I think.
03:22:48 Yumeng bandit is for code security check, which can check operations like if shell operations in cyborg code is safe
03:22:49 chenke Ye. I think that's the right way
03:23:07 Sundar So, the programming API https://review.opendev.org/698190 is not for Ussuri?
03:23:10 brinzhang Sundar: I mean, we should enable it as the base patch, and then fix the issue, right?
03:24:00 shaohe_feng Bandit will scan the python code and gen AST to check security issues
03:24:49 s_shogo Sundar , shaohe_feng : As I mentioned in the program patch , as reply for the Sundar's comment (https://review.opendev.org/698190) , I would like to take discussion the interface of PATCH API or not. IMO, that effects the schedule of merge.
03:25:15 shaohe_feng we had better let programming API in cyborg as soon as possible
03:25:34 shaohe_feng so we should try do our best to make it ready
03:26:04 shaohe_feng this is a very important feature
03:26:16 shaohe_feng s_shogo thank you for post this feature.
03:26:19 Sundar brinzhang: IMHO, it is better to have a patch series that fixes the open bandit issues and then make it a voting job. We need to merge other open patches for Ussuri. They will be affected if we make it a voting job too soon. Agreed?
03:26:24 brinzhang shaohe_feng:I also prefer to use update method
03:26:45 shaohe_feng update for firmware update?
03:27:04 shaohe_feng do you means a new update API?
03:27:52 brinzhang Sundar: we just enable the vote in zuul, but cannot merged, that cannot impact any open patches, I think
03:27:59 Yumeng Sundar,brinzhang : Agreed. We should fix bandit failures first, then update bandit job as voting in zuul check. otherwise, if we change voting first, other patches will get Zuul -1.
03:28:19 shaohe_feng let's welcome a new fresher haibin-huang
03:28:38 brinzhang if we just fix that bandit failures, we cannot ensure it work fine after we enabled bandit vote in zuul
03:28:40 Yumeng brinzhang: I will post all patches for bandit today.
03:28:44 shaohe_feng she will try to help verify the program API in product evn.
03:29:17 chenke s/she/he
03:29:28 haibin-huang23 I am very glad to join cyborg family.
03:29:29 Sundar Welcome haibin-huang . Would you like to introduce yourself?
03:29:38 shaohe_feng sorry, he :]
03:29:48 Yumeng welcome haibin-huang! :)
03:29:53 shaohe_feng haibin-huang you can help to review this new API
03:29:58 brinzhang welcome haibin-huang
03:30:03 chenke welcome haibin again. since beijing hackson.
03:30:17 s_shogo welcome haibin-huang
03:30:18 xinranwang welcome haibin-huang23
03:30:26 songwenping welcome haibin-huang too
03:30:41 shaohe_feng #link https://review.opendev.org/#/c/698190/
03:30:59 brinzhang Yumeng: ok, sound good, we shuold ensure the bandit run works fine, than star merge the open fix bandit patches

Earlier   Later