| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-cyborg - 2020-03-11 | |||
| 05:54:03 | openstackgerrit | Brin Zhang proposed openstack/cyborg master: Add 'description' field to the device profile object https://review.opendev.org/703293 | |
| 06:09:51 | openstackgerrit | Brin Zhang proposed openstack/cyborg master: Add 'description' field to the device profile object https://review.opendev.org/703293 | |
| 07:21:45 | openstackgerrit | Sundar Nadathur proposed openstack/cyborg master: Update contributing.rst per Ussuri goal and guidelines. https://review.opendev.org/712246 | |
| 08:22:52 | openstackgerrit | chenker proposed openstack/cyborg master: Bugfix: expose placement service unavailable before diff cyborg db https://review.opendev.org/711912 | |
| 08:44:23 | openstackgerrit | chenker proposed openstack/cyborg master: Bugfix: Set the appropriate status code for ServiceUnavailable https://review.opendev.org/712283 | |
| 11:38:28 | openstackgerrit | Brin Zhang proposed openstack/cyborg master: Improve the accuracy of device profile API responses in docs https://review.opendev.org/712014 | |
| 11:42:49 | openstackgerrit | Brin Zhang proposed openstack/cyborg master: Improve the accuracy of device profile API responses in docs https://review.opendev.org/712014 | |
| 21:07:00 | openstackgerrit | Sundar Nadathur proposed openstack/cyborg master: Check during ARQ bind that the target instance has no other ARQs. https://review.opendev.org/712231 | |
| 21:36:13 | openstackgerrit | Sundar Nadathur proposed openstack/cyborg master: Check during ARQ bind that the target instance has no other ARQs. https://review.opendev.org/712231 | |
| #openstack-cyborg - 2020-03-12 | |||
| 02:11:26 | Sundar | Hello all | |
| 02:11:34 | brinzhang | \O | |
| 02:12:11 | Sundar | Hi brinzhang. Anybody else here? | |
| 02:13:26 | brinzhang | I think they are waiting 3:00UTC, right? | |
| 02:13:49 | brinzhang | Few people joined irc | |
| 02:14:35 | Sundar | Oh yes, with daylight savings, my meeting time has shifted again | |
| 02:16:05 | brinzhang | ah, nothing, wait 45min later | |
| 02:20:48 | chenke | brinzhang https://review.opendev.org/#/c/710911/ need your reply. | |
| 02:21:55 | brinzhang | chenke: will check | |
| 02:22:04 | chenke | ok. thanks. | |
| 02:25:59 | Sundar | chenke, brinzhnag: I addressed your comments in https://review.opendev.org/#/c/712231/ | |
| 02:26:08 | Sundar | brinzhang: ^ | |
| 02:26:35 | chenke | Ye. I reply you. seems UT passed. | |
| 02:29:48 | brinzhang | Sundar: hi | |
| 02:31:34 | brinzhang | Sundar: thanks add that test case of bound exception, that what I want ^^ | |
| 02:32:30 | openstackgerrit | chenker proposed openstack/cyborg master: Bugfix: Set the appropriate status code for ServiceUnavailable https://review.opendev.org/712283 | |
| 02:33:24 | Sundar | brinzhang: That's what I was mentioning yesterday. If I apply the UT to self.Patch_json, it would fail. I solved it by factoring out _check_if_bound, and applying UT to it separately. | |
| 02:34:02 | Sundar | We still need to solve the problem of handling the HTTP error for exception test cases for post/patch etc. | |
| 02:34:33 | chenke | Your latest approach is appropriate, UT should only test this function. | |
| 02:34:59 | Sundar | chenke, brinzhang: You may want to read the new commit message. I added lots more info in there. | |
| 02:35:01 | brinzhang | Sundar: I will check again your latest patch at afternoon. | |
| 02:35:29 | Sundar | Sure, np | |
| 02:36:40 | brinzhang | "We still need to solve the problem of handling the HTTP error for exception test cases for post/patch etc.", I hasn't concern this issue, in future, I think I will meet | |
| 02:37:37 | brinzhang | If anyone can get this clearly, can push a commit that we can check together | |
| 02:37:50 | chenke | Sundar cool. The latest commit message is clear. | |
| 02:38:20 | openstackgerrit | YumengBao proposed openstack/cyborg master: Fix GPU's improper cpid_type allowed value and attach_handle_type value https://review.opendev.org/709433 | |
| 02:38:49 | openstackgerrit | YumengBao proposed openstack/cyborg master: Fix GPU's improper cpid_type allowed value and attach_handle_type value https://review.opendev.org/709433 | |
| 02:39:02 | brinzhang | Sundar, chenke, xinranwang, Yumeng, I want to start use launchpad, do you think all? | |
| 02:40:19 | Sundar | brinzhang: launchpad for blueprints? | |
| 02:40:39 | brinzhang | I think it's easier to manager bug,blueprint, as I know cinder,nova,neutron all use it | |
| 02:41:35 | Sundar | I have no objection either way. But we would need to migrate all stories in Storyboard to launchpad. | |
| 02:42:22 | brinzhang | Any tools can do? | |
| 02:49:23 | Sundar | brb in 10 min | |
| 02:51:29 | chenke | Didn't it have been discussed in the email before? According to the opinions of community leaders, cyborg has already used storyborad, so there is no need to change it back. I think storyborad is good. | |
| 02:57:51 | Sundar | back | |
| 02:59:31 | brinzhang | chenke: I removed -1 from that patch, but I think Sundar's concern is right, how about change the fpga_program_v2 to fpga_program in this patch? | |
| 03:00:37 | openstack | Meeting started Thu Mar 12 03:00:37 2020 UTC and is due to finish in 60 minutes. The chair is Sundar. Information about MeetBot at http://wiki.debian.org/MeetBot. | |
| 03:00:37 | chenke | Ok. thanks. I replied, in my opinion, that's not what this patch should do. I think the theme should be consistent. | |
| 03:00:37 | Sundar | #startmeeting openstack-cyborg | |
| 03:00:38 | openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | |
| 03:00:41 | openstack | The meeting name has been set to 'openstack_cyborg' | |
| 03:00:57 | Sundar | o/ Sundar | |
| 03:01:03 | chenke | #info chenke | |
| 03:01:19 | Yumeng | #info Yumeng | |
| 03:01:19 | xinranwang | Hi all | |
| 03:01:20 | brinzhang | #info brinzhang | |
| 03:01:21 | s_shogo | #info s_shogo | |
| 03:01:23 | s_shogo | Hi all | |
| 03:01:25 | xinranwang | #info xinranwang | |
| 03:02:14 | Sundar | First off, I am happy t o propose s_shogo and brinzhang as core reviewers: http://lists.openstack.org/pipermail/openstack-discuss/2020-March/013230.html | |
| 03:03:17 | Sundar | It will take effect next Friday (March 20) barring any objection. | |
| 03:03:41 | brinzhang | Sundar: It's my honor, thanks, thanks all | |
| 03:04:00 | Sundar | brinzhang: welcome, you have earned it. | |
| 03:04:03 | chenke | Congratulations Brin, Shogo. | |
| 03:04:08 | Yumeng | congrats! brinzhang,and s_shogo | |
| 03:04:14 | s_shogo | Thanks all! | |
| 03:05:17 | xinranwang | congratulations! | |
| 03:06:16 | brinzhang | Thanks Sundar, chenke, Yumeng, xinranwang, s_shogo ^^ | |
| 03:06:28 | Sundar | Secondly, we should take stock of our progress in the U release. | |
| 03:07:34 | Sundar | Here's the U release planning etherpad: https://etherpad.openstack.org/p/cyborg-ptg-ussuri | |
| 03:08:04 | Sundar | Nova integration | |
| 03:08:31 | Sundar | Nova integration is making good progress, still hope to finish it by March. | |
| 03:09:09 | Sundar | Cyborg client: OpenStackSDK part is done. Hereafter it is incremental improvement, right? | |
| 03:09:31 | Sundar | s_shogo, all: ^ | |
| 03:11:37 | s_shogo | Yes, I'll implement filter function and so on >> client | |
| 03:12:19 | Sundar | Would the client be mostly done by Ussuri? Or do you think it'll go to V release? | |
| 03:13:02 | s_shogo | The SDK including cyborg part was released yesterday. Thus that could be released in U release,IMO. | |
| 03:13:44 | s_shogo | ( SDK's release cycle is independent from OpenStack's one) | |
| 03:13:58 | Sundar | The reason I am asking is, the Cyborg client has been released midcycle. If most functionality is in, we could consider an end-of-cycle release instead of waiting for V. | |
| 03:15:04 | xinranwang | client is based on v2 API, I think we don't have v2 deployable API now, right ? | |
| 03:15:17 | s_shogo | OK I got it. Release in U could be chosen | |
| 03:15:41 | s_shogo | Yes >> xinranwang | |
| 03:16:13 | brinzhang | The SDK cannot match Cyborg API now? | |
| 03:17:10 | s_shogo | brinzhang No, the Latest SDK supports many Cyborg API, but that is not perfect like xinranwang mentioned. | |
| 03:17:11 | brinzhang | I know the v2 microversion in progress, if so, the SDK/cyborgclient maybe cannot works good, rirht? | |
| 03:18:01 | brinzhang | s_shogo: Ok, later, while I am free, I will test in my evn ^^ | |
| 03:18:09 | xinranwang | s_shogo: got it. | |
| 03:18:18 | s_shogo | brinzhang thanks:) | |
| 03:18:24 | xinranwang | brinzhang: yes, we should support microversion in client side too | |
| 03:18:57 | Sundar | brinzhang: The microversion support in the API server side means that the client must specific headers with the microversion number. That is probably the next big infrastructure in the client. Right, s_shogo? | |
| 03:19:08 | brinzhang | xinranwang: necessary. in clinet we must support microversion to request the Cyborg api | |
| 03:19:11 | Sundar | *must send | |
| 03:19:42 | xinranwang | brinzhang: yes, exactly | |
| 03:20:03 | brinzhang | s_shogo: I installed cyborgclient by "pip install python-cyborgclient", is that ok by this way? | |
| 03:20:37 | s_shogo | Yes >>Sundar , xinranwang and me start to investigate a method of supporting them with sdk's functions. | |
| 03:20:56 | Sundar | Great. Thanks, s_shogo | |
| 03:21:02 | Sundar | xinranwang: Support for deployables in the client is an incremental feature. That brings me to the 2nd item above. /v2/devices API for inventory is done. But /v2/deployables for programming is open. | |
| 03:21:42 | brinzhang | s_shogo: we should add --os-cyborg-api-version <cyborg-api-ver> in your client implemation | |
| 03:21:58 | Sundar | So, our top 3 priorities for U are still the same as above: Nova integration | |
| 03:22:15 | Sundar | The last one includes microversion support. | |
| 03:22:29 | xinranwang | Sundar: yes, device API is done except upgrade firmware, and deployable v2 is missing | |