Wednesday, 2019-06-19

« Tuesday, 2019-06-18 Index Thursday, 2019-06-20 »
*** ikuo_o has joined #openstack-cyborg00:27
*** yuki_t has joined #openstack-cyborg02:21
*** Coco_gao has joined #openstack-cyborg02:47
*** Biwei has joined #openstack-cyborg02:58
Coco_gaoHi all03:00
BiweiHi03:00
*** Sundar has joined #openstack-cyborg03:01
Sundar#startmeeting openstack-cyborg03:01
openstackMeeting started Wed Jun 19 03:01:46 2019 UTC and is due to finish in 60 minutes.  The chair is Sundar. Information about MeetBot at http://wiki.debian.org/MeetBot.03:01
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.03:01
*** openstack changes topic to " (Meeting topic: openstack-cyborg)"03:01
openstackThe meeting name has been set to 'openstack_cyborg'03:01
Sundar#topic Roll call03:01
*** openstack changes topic to "Roll call (Meeting topic: openstack-cyborg)"03:01
Sundar#info Sundar03:02
Biwei#info Biwei03:02
Coco_gao#info Coco_gao03:02
SundarHi Biwei, Coco_gao03:02
BiweiHi Sundar03:02
*** wangzhh has joined #openstack-cyborg03:03
SundarLet's wait a minute for folks to join03:03
SundarHi wangzhh03:03
Coco_gaoI am sorry that I may miss two week's meeting03:03
wangzhhHi Sundar. :)03:03
SundarCoco_gao: You will miss next 2 weeks?03:04
Coco_gaoNope03:04
Coco_gaoI missed03:04
SundarAh, got it. NP.03:04
Coco_gaolast two week's meeting.03:04
SundarAgenda: https://wiki.openstack.org/wiki/Meetings/CyborgTeamMeeting#Agenda03:05
SundarPlease take a look.03:05
ikuo_o#info ikuo_o03:05
SundarAnything to be added to the agenda?03:06
Sundar#topic Device profile names03:06
*** openstack changes topic to "Device profile names (Meeting topic: openstack-cyborg)"03:06
SundarOne question that has come up, esp. in the review of https://review.opendev.org/#/c/626057/, is whether device profiles are unique, or should be unique03:07
SundarMy answer is in the response to the comment in Line 466 of https://review.opendev.org/#/c/626057/10/cyborg/db/sqlalchemy/api.py03:07
SundarPlease take a look first. Then, we'll discuss.03:08
SundarCan you all see it?03:09
wangzhhYep.03:09
wangzhhAs my comment in line.  The name should be unique.03:11
Coco_gaoIn the db, it was unique03:11
Coco_gaoI see Li posted that the name should not be used to identify.03:12
Sundarwangzhh: https://review.opendev.org/gitweb?p=openstack/cyborg.git;f=cyborg/db/sqlalchemy/models.py;hb=refs/changes/57/626057/10#l18303:12
SundarIt is unique in the db03:13
SundarCoco_gao: given my response, do you agree that names must be unique and should be used to identify DPs? I am ok with using UUIDs also, if there is a need03:13
wangzhhCool. I think it's better to identify DP with UUID.03:15
Coco_gaoFor other projects, we usually use UUIDs not the names.03:15
Coco_gao+1 better to use UUID03:16
SundarYes. But, if we use UUIDs in all APIs, the operator would have to embed DP uuids in flavors, instead of names.03:17
wangzhhWe just use UUIDs in internal API. For the rest api, name is more friendly.03:18
ikuo_oCan we find UUIDs from name by using APIs?03:18
wangzhhJust like nova. We can support nova show VM_UUID/VN_NAME.03:19
Sundarikuo_o: Yes, at the cost of a db access03:19
ikuo_othanks.03:19
Sundarwangzhh: Internal APIs -- e.g. from API layer to object layer -- can use UUIDs. The problem then is that the API layer needs to do a db lookup to get the uuid from the name, right?03:20
wangzhhYes.03:20
Coco_gaoSo the name should also be unique?03:20
Coco_gaoone uuid corrosponds to one name?03:21
SundarCoco_gao: The name has to be unique anyway. If not, when the operator puts a dp name in a flavor, it would be ambiguous.03:21
SundarYes, one uuid == one name03:21
SundarI think an API to rename a device profile would have to use UUIDs. We don't have that today.03:22
SundarWe can do that in the future once Cyborg is is use03:23
Sundar*in use03:23
Coco_gaoAgree03:23
wangzhh+1 Great.03:23
Coco_gaoUUIDs is not urgent before Cyborg is in use03:23
Coco_gaoIf we have time, better to add that, but that's not urgent so far.03:24
SundarOne caveat: we have /v2/device_profiles/{name} as a valid URL03:25
SundarSo, if we want a specific UUID, it would have to be a query like: /v2/device_profiles?uuid={uuid}03:25
SundarBut Nova does not use the former: it only does /v2/device_profiles?nam={name}03:26
SundarSo, we have some flexibility to change the former03:26
SundarAny thoughts?03:29
Coco_gaoYou mean we should focus on the name API then see if we need UUID API?03:31
SundarYes03:31
ikuo_oI'm sorry I have to leave now... if any question to NTT, yuki_t will answer.03:31
Coco_gaoseems good for me03:31
SundarTake care, ikuo_o.03:31
wangzhhEmmm, fine.03:32
SundarThanks, Coco_gao.03:32
Sundarwangzhh: Thanks03:32
Sundar#topic Format for controlpath_id and attach_handles03:32
*** openstack changes topic to "Format for controlpath_id and attach_handles (Meeting topic: openstack-cyborg)"03:32
SundarPleasee line 79 https://review.opendev.org/#/c/626057/10/cyborg/accelerator/drivers/fpga/intel/driver.py03:33
SundarThe intended usage of controlpathid_info and attach_handle_info, IMHO, are like {domain: 0, bus: ... }03:34
SundarBut apparently it got encoded as a string '0000:05...'03:34
SundarIMHO, that is not good because it is an implcit syntax. That may possibly vary among systems?03:34
SundarIt is better to make it an explicit dictionary or JSOn string03:35
SundarThoughts?03:36
Coco_gaoAgree to use JSON string and change the related patches.03:37
SundarGood, thanks03:38
wangzhhAll look good to me. Just effort.03:38
Sundarwangzhh: Yes. Hopefully, only the driver report needs to be changed?03:39
SundarIt is a string either way03:39
Coco_gaoDose the json string have the specific keys?03:40
wangzhhYes. I think so.03:40
SundarCool03:40
Sundar#topic Need to close specs03:40
*** openstack changes topic to "Need to close specs (Meeting topic: openstack-cyborg)"03:40
Coco_gaoonly domain and bus?03:40
SundarTrain release schedule: https://releases.openstack.org/train/schedule.html03:40
Coco_gaoor it varies?03:40
SundarCoco_gao: No, all 4.03:40
SundarI had "..." for other 203:40
SundarWe are past Milestone 1 of Train03:41
Coco_gaoMaybe we add a ovo for that?03:41
SundarI am a little concerned that the core specs are not still not reviewed.03:41
openstackgerritShaoHe Feng proposed openstack/cyborg master: bug fix: deploy every cyborg components correctly  https://review.opendev.org/65929203:41
SundarNova-Cyborg spec got two +2s and singificant +1s. So, that is on track03:41
Coco_gaothe workflow is not +1?03:42
Coco_gaoOK, I will review the specs as soon as possible.03:43
Coco_gaoand update related pathes.03:43
SundarWell, one reviewer posed a question. So, it was decided not to merge the spec till I answered the question. I did that today and updated the spec.03:43
SundarThanks, Coco_gao03:43
Coco_gaoI see03:43
Coco_gaoThank you for you efforts03:44
SundarIn the list https://review.opendev.org/#/q/status:open+project:openstack/cyborg-specs+owner:Sundar, please review just the top 303:44
Coco_gaoIt always not easy to merge nova code.03:44
SundarOr Nova specs :)03:44
SundarI suggest we use the Zoom meeting time tomorrow to review the Cyborg specs.03:45
Coco_gaoOK.03:45
Coco_gaoI will join then03:45
SundarGreat03:45
wangzhhOK. Got it.03:46
Coco_gaowhat's the time?03:46
SundarCool. Hope Li_Liu and others can join. Can one of you please ping them?03:46
SundarUTC 0130 (China @9:30 am Thu; US West Coast @6:30 pm Wed)03:46
Coco_gaoI will ping Li to see if he has time.03:47
wangzhhWe can ping others in the wechat group. @ everyone.03:47
Coco_gaoyes03:48
SundarSure03:48
Sundar#topic AoB03:48
*** openstack changes topic to "AoB (Meeting topic: openstack-cyborg)"03:48
wangzhhBTW, NTT guys...03:48
Coco_gaoYes, will NTT jion?03:48
Sundaryuki_t: Can you join and ask ikuo_o too?03:49
wangzhhThey can't access wechat, may miss the message.03:49
Coco_gaoSundar maybe you can send the meeting invitation?03:49
SundarI will email ikuo_o. I can resend the Zoom meeting invitation to all of you.03:49
Coco_gaothat's good03:50
wangzhhOK03:50
Coco_gaoBye, Sundar, have a good night~03:50
SundarSent now03:50
SundarThanks, all. Have a good day.03:51
Sundar#endmeeting03:51
*** openstack changes topic to "Pending patches (Meeting topic: openstack-cyborg)"03:51
openstackMeeting ended Wed Jun 19 03:51:13 2019 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)03:51
openstackMinutes:        http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-06-19-03.01.html03:51
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-06-19-03.01.txt03:51
openstackLog:            http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-06-19-03.01.log.html03:51
wangzhhBye~03:51
*** Biwei has quit IRC03:51
*** Sundar has quit IRC03:51
*** yuki_t has quit IRC04:07
*** links has joined #openstack-cyborg05:03
*** irclogbot_3 has quit IRC05:30
*** irclogbot_1 has joined #openstack-cyborg05:31
*** ikuo_o has quit IRC05:35
*** Coco_gao has quit IRC05:56
*** wangzhh has quit IRC05:58
*** yikun has joined #openstack-cyborg06:08
*** helenafm has joined #openstack-cyborg07:38
*** links has quit IRC11:37
*** links has joined #openstack-cyborg11:43
*** links has quit IRC12:17
*** links has joined #openstack-cyborg12:24
*** links has quit IRC12:24
*** Sundar has joined #openstack-cyborg15:24
*** helenafm has quit IRC15:52
*** Sundar has quit IRC22:44
« Tuesday, 2019-06-18 Index Thursday, 2019-06-20 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!