Thursday, 2019-08-22

*** edleafe has quit IRC00:55
*** zhipeng has quit IRC00:55
*** ChanServ has quit IRC00:55
*** yaawang has quit IRC00:56
*** stephenfin has quit IRC00:56
*** openstackgerrit has quit IRC00:56
*** Li_Liu has quit IRC00:56
*** irclogbot_3 has quit IRC00:56
*** jayahn has quit IRC00:56
*** yikun has quit IRC00:56
*** persia has quit IRC00:56
*** ildikov has quit IRC00:56
*** dustinc has quit IRC00:56
*** masayukig has quit IRC00:56
*** gmann has quit IRC00:56
*** donnyd has quit IRC00:56
*** efried has quit IRC00:56
*** Li_Liu has joined #openstack-cyborg01:14
*** openstackgerrit has joined #openstack-cyborg01:14
*** edleafe has joined #openstack-cyborg01:14
*** stephenfin has joined #openstack-cyborg01:14
*** yaawang has joined #openstack-cyborg01:14
*** zhipeng has joined #openstack-cyborg01:14
*** ChanServ has joined #openstack-cyborg01:14
*** sets mode: +o ChanServ01:14
*** efried has joined #openstack-cyborg01:14
*** zhipeng has quit IRC01:15
*** persia has joined #openstack-cyborg01:22
*** gmann has joined #openstack-cyborg01:22
*** irclogbot_1 has joined #openstack-cyborg01:24
*** gmann has quit IRC01:27
*** gmann has joined #openstack-cyborg01:28
*** s_shogo has joined #openstack-cyborg01:34
*** zhipeng has joined #openstack-cyborg02:03
*** Yumeng has joined #openstack-cyborg02:07
*** chenke has joined #openstack-cyborg02:08
*** xinranwang has joined #openstack-cyborg02:19
*** Coco_gao_ has joined #openstack-cyborg02:19
Coco_gao_Any one here?02:20
xinranwangI'm here02:20
Coco_gao_Are we waiting for Sundar for today's agenda?02:21
openstackgerritXinran WANG proposed openstack/cyborg master: fake driver bug fix
xinranwangI think so02:23
Li_LiuHi guys02:24
Coco_gao_Hi Li02:39
zhipenglong lost leader :P02:48
Coco_gao_Sundar will join in 5 mins02:56
Li_LiuI am back in China now02:59
*** Sundar has joined #openstack-cyborg03:02
SundarHi all03:02
Sundar#startmeeting openstack-cyborg03:02
openstackMeeting started Thu Aug 22 03:02:58 2019 UTC and is due to finish in 60 minutes.  The chair is Sundar. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.03:03
*** openstack changes topic to " (Meeting topic: openstack-cyborg)"03:03
openstackThe meeting name has been set to 'openstack_cyborg'03:03
Sundar#topic Roll call03:03
*** openstack changes topic to "Roll call (Meeting topic: openstack-cyborg)"03:03
*** shaohe_feng has joined #openstack-cyborg03:03
Coco_gao_#info Coco_gao03:03
Yumeng#info Yumeng03:03
s_shogo#info s_shogo03:03
chenke#info chenke03:04
Li_Liu#info Li_Liu03:04
SundarWelcome, all03:04
Sundar#topic Python 303:05
*** openstack changes topic to "Python 3 (Meeting topic: openstack-cyborg)"03:05
SundarThere has been some feedback from TC to fix Python 3 tests and make them voting jobs03:05
SundarWe should minimally try to get the Python 3 stuff fixe03:06
Sundars_shogo: Thanks for your patches. Is it possible to prioritize it? We can even test with the patches03:06
xinranwang#info xinranwang03:07
s_shogoI'm waiting for merge of P1-P9 patches, after that , I'll delete WIP in py3 patch.03:07
SundarWe can test and update the Py3 patch with the P5-P9 patches. I agree we need to speed up the merging of those patches. But we don't need to wait for that to update Py3 fixes, right?03:08
s_shogook, I'll delete the WIP and request review  today.03:09
*** wangzhh has joined #openstack-cyborg03:09
s_shogoI intend to commit the patch with "non-voting" ,are there any problems?03:10
SundarWe all voted in a past IRC meeting that it should be non-voting for now. So I agree.03:10
SundarDo you need help with testing with the patches? You could do: 'git review -d 670470' and then 'git review -x 667524'03:11
SundarThat will pull the P5-P9 patches and your patch in one branch03:11
s_shogoThanks, I'll do that.03:12
SundarGreat. Thanks a lot, s_shogo :)03:12
Sundar#topic Merging patches03:12
*** openstack changes topic to "Merging patches (Meeting topic: openstack-cyborg)"03:12
SundarAs s_shogo pointed out, a lot of things are blocked on merging the P5-P9 patches. Can you all please prioritize that?03:12
Coco_gao_About P5 patch,   a lot of patch is denpending on it.03:14
SundarGreat. Xinran's patches on Nova notification and Placement are important too. Once they also merge, we can enable tempest CI, and that is what we need to kick off Nova integ.03:14
Yumenghi Sundar: I noticed you gave a -2 on P5: Basic changes for API layer.    are there any problems?03:14
SundarCoco_gao: Agreed. I have tested the P5-P9 with fake and FPGA drivers.03:15
Coco_gao_yumeng also need to modify the database03:15
xinranwangSundar:  do you want to discuss vendor_id, vendor_name stuff, which we discussed yesterday?03:15
SundarYumeng: It is a procedural -2 to prevent the patch series from merging piece by piece. It will be cleaner if we all review all the patches as a whole and then the whole thing merges around the same time.03:15
SundarSame reasoning as with the -2 for in Nova03:16
xinranwangPlease review placement report patch, it has less dependency on P5-P6 patch. :)03:17
YumengSundar: got it . thanks03:17
Coco_gao_So can we make the priority list about the patches?03:17
SundarCoco_gao: I would suggest: P5-P9, Nova notification, Placement report. What do you all think?03:18
SundarNot saying other patches are less important :)03:18
SundarJust tryin to unblock as much as we can and get the ball back to Nova developers' court.03:18
Coco_gao_I think some of the patches had dependency issues so I am asking about the priority.03:19
xinranwangI think we can review placement from now, cause it has less dependency on your patches.03:19
xinranwangnotification after p5-p903:19
Coco_gao_Both dependency and our Train release goal are matter.03:20
Sundarxinranwang: Not disagreeing, but Placement needs to be tested with P5-P9 patches, right?03:20
Coco_gao_P5-P9, Nova notification, Placement report.  What about python3?03:20
xinranwangwe can run placement patch without nova-cyborg integration03:20
SundarOk, we could do partial testing with CYborg driver -> agent -> conductor (db) -> Placement. Without using Nova. Is that what you are saying?03:21
xinranwangYes, exactl.03:21
Coco_gao_OK, got that.03:21
xinranwangAlso, placement provides cmd line like openstack resource provider list, we can check the result directly by using this.03:22
SundarOk, then order with dependencies is: [ P5-P9, Placement report], then nova notification. Sounds good?03:23
xinranwangyes,  great03:24
Coco_gao_I see there are several patches which had conflict problems.03:24
Coco_gao_Will that be solved if we merge P5-P903:24
SundarCoco_gao_: In, I see 4 merge conflicts and they are all old patches only for reference. Rest should be good.03:25
Coco_gao_Seems good to me because I know where to start my reviews.03:25
xinranwangSundar wangzhh: I noticed that GPU driver report vendor id and fill the vendor field with ID, is there any possibility that we fill vendor field by vendor name?03:26
SundarIMHO, we need the vendor name for the device somehwere, so that we can create the traits with the vendor name, as listed in;f=specs/train/approved/cyborg-nova-placement.rst;hb=refs/changes/45/603545/7#l17903:27
Coco_gao_We can get the vendor name from vendor ID right?03:27
Coco_gao_Since the vendor ID is unique03:27
SundarYes, but the driver report does not include the vendor name as a field today03:28
wangzhhOf course, but I suggest record vendorId in db. And translate it to name if needed.03:28
SundarPerhaps it will be easiest to add vendor name as a separate field in the driver_objects.device, to minimize changes to GPU driver and other drivers03:28
Coco_gao_Agree that we should keep vendor ID03:29
xinranwangeach driver should know about the mapping relation between id and name, but cyborg-agent/conductor don't03:29
SundarThe Cyborg driver for a vendor could even hardcode the vendor name, if it will handle only one vendor.03:30
xinranwangSo IMO, driver should report vendor name anyway.03:30
Sundarxinranwang: Agreed. Proposal: Add vendor name to  driver_objects.device . Any objections?03:30
wangzhhSundar, It is  hardcoded in driver now.03:30
xinranwangno objection.03:31
xinranwangBtw, i find driver name in deployable tables, but no assignment.03:31
shaohe_fengYes, I translate the vendor ID to name for FPGA in my DEMO03:32
Coco_gao_Agree. The object.device will not add vendor name? Only driver_objects.device will?03:32
shaohe_fengit can add a vendor name03:32
SundarCoco_gao_: Good point. It will be easier to add it to device object too, for use in ext-arq binding for creating traits03:33
shaohe_fengbut maybe more driver infos, such as driver version.03:33
SundarDo we need to add that to the db then? Oherwise, how how do we get it in the device object by querying?03:34
wangzhhSo just add name in object, not in db, right?03:34
Coco_gao_My question is do you think we should keep vendor name in object or database?03:34
shaohe_fengfor a long term.  maybe we can translate the ID to name at present.03:34
Coco_gao_vendor ID is already in DB?03:34
xinranwangSundar:  what is driver name you added in deployable table stand for?03:35
wangzhhcoco  yep03:35
SundarIf you don;t add the vendor name to db, the translation from id to name may have to be outside the driver too, which is not good03:35
xinranwangCoco_gao_: yes, in device tables03:35
YumengCoco_gao_:yes.  already have vendor_id  in db device table03:36
Coco_gao_About the ext-arq binding, do we need to qurey DB for vendor name?03:36
Sundarxinranwang: It is the driver that is used for programming and device updates. At discovery time, we know which deployable has which driver. So, we can use that info to locate the driver needed for programming.03:36
xinranwangthere is no assignment of this field03:37
SundarCoco_gao_: For ext_arq binding, we are queryig db for the deployable now. We can extend tthat to query the device and its vendor name too03:37
xinranwangmaybe we can extract vendor name from this field?03:37
shaohe_fenghere is what I do, just for a temporary solution03:37
shaohe_feng       driver_name = db_deployable.driver_name03:38
Sundarxinranwang: The intention is that, when the agent discovers a device from a driver, it sends the driver name to the conductor and has it persisted.03:39
Sundarshaohe_feng: That is what we are doing today. We could use the driver name and vendor name as same-- at least for Train. Is that what I am hearing from many of you?03:40
xinranwangSundar:  so what will it look like? intel_fpga_driver?  or 0x8086 like shaohe's demo does.03:41
shaohe_fengfor fpga driver, I maintain a map, both vendor ID and name  can work.03:42
shaohe_fengSundar: yes,  we need a driver name03:42
Sundarxinranwang: just 'intel', for example. That is what we use to load a driver module. The intel_fpga_driver is the name in setup.cfg.03:43
xinranwangso we should fill driver_name with 'intel' for example?03:44
xinranwangif so, we can extract vendor name for driver_name..03:44
SundarYes, xinranwang. To summarize, I think we are agreeing to update the driver_name field in the deployable and use that as 'vendor name' to create the traits as per;f=specs/train/approved/cyborg-nova-placement.rst;hb=refs/changes/45/603545/7#l17903:44
SundarSounds good?03:45
SundarThen no need to add vendor_name to any object03:45
SundarGreat. If others have objecttions, please LMK. Let's move on in the interest of time03:46
Sundar#topic python cyborg client03:47
*** openstack changes topic to "python cyborg client (Meeting topic: openstack-cyborg)"03:47
s_shogoI'm working that with P5-P9 patch in advance.03:47
SundarThanks, s_shogo. I am wondering if we have the time to update the cyborg client to v2 API?03:48
Coco_gao_So zhenghao's patch should use Vendor ID and then update deployable vendor name?03:48
SundarCoco_gao_: Yes03:48
Coco_gao_update to deployable table with vendor name03:48
Coco_gao_Sounds good.03:48
SundarIt already has a driver name. At least today, driver name and vendor name are the same.03:49
Sundars_shogo: May be we need somebody to help you, for the client?03:49
s_shogoSundar : ok, I think so. I'll report the prospection for the v2 API Work in every IRC  meeting.03:49
Sundars_shogo: Thanks a lot. Appreciate it. :)  Please do ask if you need help.03:50
*** links has joined #openstack-cyborg03:50
SundarBTW, TC has proposed:
s_shogoSundar: Ok, Thanks.03:50
SundarThey are proposing an intermediate release of the client. IMHO, it is useless because it is currently v1 based, has no Python 3 fixes and so nothing new from Stein release03:51
SundarWe may as well club the client update with Train release. What do you al think?03:51
SundarCoco_gao_, wangzhh, xinranwang, shaohe_feng, Li_Liu, all: ^03:53
Coco_gao_Agree if we have time. I don't know whether we can make that happen, are you optimistic about that?03:54
xinranwangwhat's club mean....03:54
Sundarxinranwang: I mean combine client update with rest of Train release.03:55
SundarCoco_gao_: Yes, i share that concern. Once s_shogo lets us know if it can happen by Sep, we can decide.03:55
xinranwangYes, nice to have.03:56
SundarYes, It will probably be lesser priority than Nova integ, Python 3, RBAC/security etc. Anybody thinks otherwise?03:57
Sundar#topic Other important Train tasks03:57
*** openstack changes topic to "Other important Train tasks (Meeting topic: openstack-cyborg)"03:57
shaohe_fengshould the async bind in this release?03:57
Sundarwangzhh: Could you give us an update on RBAC, whenever you have the time?03:58
Sundarshaohe_feng: Yes. It is needed for Nova integ03:58
xinranwangI notice current code does not create deployable with driver_name, does it? Maybe need one patch on it?03:58
Sundarxinranwang: Yes, it is a gap03:58
shaohe_fengThen we will add 2 more state for bind.03:58
shaohe_feng ARQ_BIND_STARTED and ARQ_DELETING03:59
SundarBIND_STARTED is already being added in my patches03:59
SundarBut I agree03:59
shaohe_fengthere will be 7 states03:59
shaohe_fengUse “->” for transform , “X->” for can’t transform.04:00
shaohe_fengTING X->  ARQ_BOUND04:00
shaohe_fengany comments on it?04:00
SundarYes. Between bind and unbind, making bind async is more important. We don;t do much in unbind today04:00
xinranwangSundar:  I can help on it. I will submit a patch by today to fix that.04:00
Sundarxinranwan: Excellent. Thanks :)04:00
shaohe_feng^ Sundar Coco_gao_ xinranwang wangzhh Yumeng04:00
xinranwangSundar: np04:01
Sundarshaohe_feng: I agree04:01
Coco_gao_We'd better merge the old patch P5-P9 first, and then discuss what are the states?04:01
Coco_gao_Maybe a new patch for states update?04:01
SundarI think shaohe_fend will propose a new patch for async un/bind04:02
Coco_gao_That's good.04:02
Sundarwangzhh: Can you please let us know about RBAC?04:02
xinranwangbtw, please review this when you got time :)04:02
shaohe_fengYes. After discussion, I should also implement the code. So I'm afraid there is no time left for me04:02
Coco_gao_maybe he is offline04:02
wangzhhTest in my local env04:03
shaohe_fengwe should discuss it ASAP04:03
Coco_gao_you can arrange a meeting to discuss that.04:03
wangzhhprobably  submit patch tomorrow04:03
shaohe_fengany comments on the states transform above?04:03
Sundarshaohe_feng: I agree with your states. I think they will suffice. Let's start with them and have people code review it?04:04
shaohe_feng^ Sundar Coco_gao_ xinranwang wangzhh Yumeng04:04
SundarI will also prioritize it for next week's IRC call04:04
Coco_gao_As good as my expectation, thank you.04:04
SundarHopefully, P5-P9 patches would have merged by then :)04:05
Yumengshaohe_feng: no objections ^^04:05
Coco_gao_Let's set a deadline for P5-P9 reviews?04:05
Sundar#topic AoB04:05
*** openstack changes topic to "AoB (Meeting topic: openstack-cyborg)"04:05
SundarCoco_gao_: That would be welcome!04:05
wangzhhshaohe no objections04:05
Coco_gao_So far so good.04:06
Yumenghey, btw, I would like to introduce you my colleague chenke, he will join us and contribute in cyborg.04:06
SundarWhat deadline do we want to set?04:06
shaohe_fengso are you satisfy the state transform ?04:06
Coco_gao_end of this week?04:06
Coco_gao_including the weekend.04:06
SundarWelcome, chenke :)04:06
shaohe_fengwelcome cheke04:06
xinranwangwelcome chenke04:06
Coco_gao_I had already saw your patch, welcome chenke.04:07
SundarCoco_gao_: Sounds good to me. Others good with that?04:07
chenkeHello everyone, I am new here, I hope to contribute to cyborg.04:07
shaohe_fengsorry s/cheke/ chenke04:07
Sundarchenke, could you please tell us your backgrounnd or skill set?04:07
chenkePreviously, I mainly contributed to the watcher project.04:07
SundarOk, sounds good04:08
chenke I have been working on watcher for two years. Now is a core for watcher.04:08
Coco_gao_Sounds really good04:09
chenke I am happy to contribute to cyborg.04:09
SundarIn terms of other Train tasks, we continue to work on enabling tempest. The is needed to enable upstream tempest CI04:09
Coco_gao_Glad you can join us.04:09
Sundarchenke, BTW, we have a Storyboard:!/project/openstack/cyborg . Please feel free to see if you find anything interesting that you can take up. You can reach us on IRC or me by email (
chenkeI will do it. Thanks.04:11
SundarGreat. wangzhh: Thanks for your response on RBAC too.04:11
SundarAnything else for today?04:11
Yumengno from my side04:12
Coco_gao_Good night Sundar04:12
SundarThanks for a productive discussion, everybody. Have a good day (or night)!04:12
*** openstack changes topic to "Pending patches (Meeting topic: openstack-cyborg)"04:12
openstackMeeting ended Thu Aug 22 04:12:52 2019 UTC.  Information about MeetBot at . (v 0.1.4)04:12
openstackMinutes (text):
*** Sundar has quit IRC04:14
*** s_shogo has quit IRC04:58
*** chenke has quit IRC05:50
*** xinranwang has quit IRC06:48
*** wangzhh has quit IRC07:49
openstackgerritShogo Saito proposed openstack/cyborg master: python3 patch (including zuul config)
*** Coco_gao_ has quit IRC08:20
openstackgerritXinran WANG proposed openstack/cyborg master: Notify Nova when binding finished
*** xinranwang has joined #openstack-cyborg09:44
*** s_shogo has joined #openstack-cyborg09:54
s_shogosphinx 's warning policy seems to be changed (Aug 19), thus cyborg's tox docs test fails...10:04
s_shogoTo avoid this issue, we have to comment out a part of doc/source/
openstackgerritXinran WANG proposed openstack/cyborg master: Fill "driver_name" field in Deployable object
openstackgerritXinran WANG proposed openstack/cyborg master: Placement report
openstackgerritShogo Saito proposed openstack/cyborg master: Fix docs gate issue
openstackgerritchenker proposed openstack/cyborg master: Solve tox -e docs error
openstackgerritShogo Saito proposed openstack/cyborg master: python3 patch (including zuul config)
*** s_shogo has quit IRC11:17
*** shaohe_feng has quit IRC11:55
openstackgerritDonny Davis proposed openstack/cyborg master: Cleanup index / add in driver support matrix
openstackgerritDonny Davis proposed openstack/cyborg master: Cleanup index / add in driver support matrix
*** Yumeng has quit IRC12:18
*** donnyd has joined #openstack-cyborg12:32
openstackgerritMerged openstack/cyborg master: Fix docs gate issue
*** links has quit IRC12:46
*** xinranwang has quit IRC12:55
openstackgerritDonny Davis proposed openstack/cyborg master: Cleanup index / add in driver support matrix
*** links has joined #openstack-cyborg13:13
*** links has quit IRC13:46
*** altlogbot_1 has joined #openstack-cyborg13:46
*** altlogbot_1 has quit IRC13:48
*** altlogbot_3 has joined #openstack-cyborg13:51
*** altlogbot_3 has quit IRC16:14
*** altlogbot_3 has joined #openstack-cyborg16:14
*** irclogbot_1 has quit IRC18:14
*** irclogbot_0 has joined #openstack-cyborg18:14
*** altlogbot_3 has quit IRC18:29
*** edleafe has quit IRC18:29
*** edleafe has joined #openstack-cyborg18:35
*** persia has left #openstack-cyborg23:56

Generated by 2.15.3 by Marius Gedminas - find it at!