| *** ChanServ has joined #openstack-mogan | 00:00 | |
| *** tepper.freenode.net sets mode: +o ChanServ | 00:00 | |
| zhenguo | morning mogan! | 00:48 |
|---|---|---|
| zhenguo | liusheng: ironic has changed to use 'baremetal' as default resource class, so we can't find filtered node | 00:52 |
| liusheng | zhenguo: ... that is the reason ! | 00:53 |
| zhenguo | liusheng: I think so | 00:53 |
| *** wanghao has quit IRC | 01:01 | |
| *** wanghao has joined #openstack-mogan | 01:02 | |
| wanghao | morning! | 01:02 |
| *** litao__ has joined #openstack-mogan | 01:13 | |
| openstackgerrit | wanghao proposed openstack/mogan master: Manage existing BMs: Part-1 https://review.openstack.org/479660 | 01:26 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Correct placement start order in devstack https://review.openstack.org/493521 | 01:28 |
| zhenguo | liusheng: we got this error GET http://mogan.readthedocs.io/projects/api-ref/en/latest/_static/api-site.css on readdocs for api | 01:39 |
| zhenguo | liusheng: GET failed | 01:39 |
| liusheng | zhenguo: maybe it because openstack doc frame lack of that css file | 01:41 |
| zhenguo | liusheng: but that's the extention of our api-ref, they should install with that | 01:41 |
| liusheng | zhenguo: don't know why.. | 01:42 |
| zhenguo | liusheng: can you fork ironic then test with it to see whether it's ok? | 01:42 |
| liusheng | zhenguo: ok | 01:42 |
| zhenguo | liusheng: thanks | 01:42 |
| liusheng | zhenguo: http://myironicdoc.readthedocs.io/en/latest/# | 01:48 |
| liusheng | zhenguo: same | 01:48 |
| zhenguo | liusheng :( | 01:48 |
| *** wanghao_ has joined #openstack-mogan | 01:51 | |
| *** wanghao has quit IRC | 01:52 | |
| zhenguo | liusheng: copying static files... WARNING: html_static_path entry '/home/docs/checkouts/readthedocs.org/readthedocs/templates/sphinx/_static' does not exist | 02:03 |
| zhenguo | liusheng: hi, I just find both doc and api-ref can't find the html_static_path | 02:20 |
| liusheng | zhenguo: :( | 02:21 |
| liusheng | zhenguo: maybe the openstack doc cannot be well compatible with readthedocs | 02:21 |
| zhenguo | liusheng: let me dig | 02:22 |
| Xinran | zhenguo, question : why we should start placement before mogan ? | 02:26 |
| zhenguo | Xinran: when mogan-engine started, it will sync resources to placement | 02:26 |
| zhenguo | Xirnan: so if placement not started, it wil fail, but seems not big issue if we can gracefully handle the exceptions | 02:27 |
| Xinran | zhenguo, that why we got the dismatch of resource name ? | 02:28 |
| zhenguo | Xinran: no | 02:29 |
| zhenguo | Xirnan: that's because ironic changed the default resource class name to 'baremetal' | 02:29 |
| zhenguo | Xinran: there's a patch up to change that in mogan :D | 02:29 |
| zhenguo | liusheng: https://review.openstack.org/#/c/493521/ | 02:29 |
| Xinran | zhenguo, ok I see, thanks :) | 02:30 |
| zhenguo | Xinran: np | 02:30 |
| zhenguo | liusheng: hi, the patch passed the gate | 02:30 |
| liusheng | zhenguo: cool, thanks | 02:30 |
| zhenguo | liusheng: hah | 02:30 |
| liusheng | zhenguo: maybe I can give a +2 +A to quick fix gate | 02:31 |
| liusheng | zhenguo: lol | 02:31 |
| zhenguo | liusheng: sure | 02:34 |
| zhenguo | liusheng: can you try to set html_static_path = ['static'] in api-ref/source/conf.py | 02:42 |
| zhenguo | liusheng: in fact you just need to remove the prefix '#' | 02:42 |
| liusheng | zhenguo: OK | 02:42 |
| zhenguo | liusheng: thanks | 02:42 |
| liusheng | zhenguo: let me try | 02:42 |
| * zhenguo brb | 02:43 | |
| openstackgerrit | Merged openstack/mogan master: Correct placement start order in devstack https://review.openstack.org/493521 | 03:07 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Make disabled flavors not available for common users https://review.openstack.org/492488 | 03:13 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Do not allow to remove flavor which is still in use. https://review.openstack.org/493412 | 03:13 |
| zhenguo | hi guys, please recheck you patches, our gate just got fixed | 03:14 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add node aggregate commands support https://review.openstack.org/492002 | 03:19 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add commands for aggregate node actions https://review.openstack.org/492370 | 03:19 |
| openstackgerrit | liusheng proposed openstack/mogan master: Trival-Fix: Correct the placement auth credentials in devstack https://review.openstack.org/493483 | 03:20 |
| liusheng | zhenguo: seems this still needed, but a trival fix, :D https://review.openstack.org/#/c/493483/ | 03:20 |
| zhenguo | liusheng: sure | 03:21 |
| openstackgerrit | Tao Li proposed openstack/mogan master: Manage existing BMs: Part-2 https://review.openstack.org/481544 | 03:21 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add support for node list command https://review.openstack.org/491644 | 03:23 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add node aggregate commands support https://review.openstack.org/492002 | 03:29 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add commands for aggregate node actions https://review.openstack.org/492370 | 03:29 |
| openstackgerrit | Merged openstack/mogan-specs master: Node aggregates https://review.openstack.org/470927 | 03:30 |
| openstackgerrit | Merged openstack/mogan-specs master: Change adopt to manage in managing-existing-node spec https://review.openstack.org/490310 | 03:30 |
| openstackgerrit | Merged openstack/python-moganclient master: Remove log translations https://review.openstack.org/482825 | 03:32 |
| openstackgerrit | Merged openstack/python-moganclient master: Update the docs link to readthedocs https://review.openstack.org/493416 | 03:33 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Improve flavor set command to support updating flavor https://review.openstack.org/486865 | 03:45 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Remove some useless code of flavor unset command https://review.openstack.org/492785 | 03:47 |
| liusheng | zhenguo: it cannnot fix the doc issue | 03:54 |
| liusheng | zhenguo: http://my-mogan-api-ref.readthedocs.io/en/latest/v1/index.html#service-urls | 03:54 |
| openstackgerrit | Merged openstack/mogan master: Make disabled flavors not available for common users https://review.openstack.org/492488 | 04:32 |
| openstackgerrit | Merged openstack/mogan master: Trival-Fix: Correct the placement auth credentials in devstack https://review.openstack.org/493483 | 04:41 |
| openstackgerrit | Tao Li proposed openstack/mogan master: Manage existing BMs: Part-2 https://review.openstack.org/481544 | 06:02 |
| openstackgerrit | Tao Li proposed openstack/mogan master: Manage existing BMs: Part-2 https://review.openstack.org/481544 | 06:17 |
| openstackgerrit | Tao Li proposed openstack/mogan master: Manage existing BMs: Part-2 https://review.openstack.org/481544 | 06:27 |
| zhenguo | liusheng: is there a way to show flavor access info with CLI? | 07:22 |
| zhenguo | litao__: does nova instance move to error state if power action failed? | 07:24 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Add checks for aggregate affinity_zone https://review.openstack.org/493495 | 07:29 |
| litao__ | zhenguo: let me check | 07:30 |
| zhenguo | litao__: ok, thanks | 07:30 |
| openstackgerrit | Merged openstack/python-moganclient master: Add support for node list command https://review.openstack.org/491644 | 07:31 |
| liusheng | zhenguo: not yet | 07:32 |
| zhenguo | liusheng: how nova handle that? | 07:32 |
| zhenguo | liusheng: return with flavor or a separated CLI | 07:33 |
| liusheng | zhenguo: seems Nova has sapareted commands to list flavor acess | 07:36 |
| liusheng | zhenguo: but OSC use flavor show | 07:36 |
| zhenguo | liusheng: ok | 07:36 |
| zhenguo | liusheng: you mean OSC can show flavor with access info? | 07:36 |
| liusheng | zhenguo: yes | 07:36 |
| zhenguo | liusheng: I forget the reason why we don't return it with flavor | 07:37 |
| liusheng | zhenguo: not sure, did we rejected to return it with flavor show ? | 07:40 |
| zhenguo | liusheng: I remember there was a patch up for that by you | 07:41 |
| liusheng | zhenguo: let me check | 07:41 |
| zhenguo | liusheng: ok | 07:41 |
| liusheng | zhenguo: cannot find any patches about this :( | 07:44 |
| zhenguo | liusheng: hah, let me find | 07:44 |
| zhenguo | liusheng: https://review.openstack.org/#/c/463308/ | 07:46 |
| liusheng | zhenguo: oh, it is in mogan side, i searched in client side | 07:46 |
| zhenguo | liusheng: hah | 07:47 |
| zhenguo | liusheng: if client should return it, better to change on mogan side | 07:47 |
| zhenguo | liusheng: but we should not return it to comon useres | 07:47 |
| zhenguo | *users | 07:47 |
| liusheng | zhenguo: so deprecate the flavor access APIs ? | 07:48 |
| liusheng | zhenguo: we have sapareted flavor access APIs | 07:48 |
| zhenguo | liusheng: do you know why we don't deprecate it before? | 07:48 |
| liusheng | zhenguo: dont know :( | 07:49 |
| zhenguo | liusheng: hah, maybe we should | 07:49 |
| zhenguo | liusheng: I can't find anything harmful | 07:49 |
| zhenguo | liusheng: and only admins can do that | 07:50 |
| liusheng | zhenguo: yes | 07:50 |
| zhenguo | liusheng: ok, let's deprecate it | 07:50 |
| liusheng | zhenguo: ok | 07:51 |
| zhenguo | Xinran: attach interface should also support specifying port_id | 07:53 |
| zhenguo | Xinran: only network supported presently | 07:53 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add node aggregate commands support https://review.openstack.org/492002 | 07:53 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Add commands for aggregate node actions https://review.openstack.org/492370 | 07:53 |
| litao__ | zhenguo: The power on/off failed don't set the server to error state, just rollback task_state. But reboot server failed will set server to error status when the server' state is not running. | 07:54 |
| zhenguo | litao__: is there any error recorded if the server status don't move to error state? | 07:55 |
| litao__ | zhenguo: Yes , It will record the error to db | 07:56 |
| zhenguo | litao__: how? with fault_info? | 07:56 |
| litao__ | zhenguo: yes, use fault_info | 07:58 |
| zhenguo | litao__: but I remember only when the instance in error state, it will return with get instance | 07:58 |
| zhenguo | litao__: does nova use cast or call for power action? | 07:59 |
| litao__ | zhenguo: cast for powering off and call for powering on , cast for rebooting | 08:00 |
| zhenguo | litao__: so power on can return error with API | 08:01 |
| zhenguo | litao__: power off just set the fault_info to db and rebooting move the instance to error state? | 08:01 |
| litao__ | yes, but reboot failed and the state is running will not set server to error state. | 08:03 |
| litao__ | zhenguo: reboot failed and the state is shutdown will set server to error state | 08:04 |
| zhenguo | litao__: you mean cast method will set the the instance to error state, right? | 08:04 |
| Xinran | zhenguo, you mean ironic port ? I saw your comments : https://review.openstack.org/#/c/459921/6/api-ref/source/v1/samples/server_networks/server-attach-interface-req.json | 08:05 |
| zhenguo | Xinran: neutron port instead of ironic port | 08:05 |
| zhenguo | Xinran: we need to specify to attach the interface to a specific neutron port | 08:05 |
| litao__ | zhenguo: only for rebooting failed | 08:06 |
| zhenguo | Xinran: instead of letting neutron to create a port in the specified network | 08:06 |
| zhenguo | litao: how about powering off? | 08:06 |
| zhenguo | litao__: oh, only when the instance status is shutdown and reboot failed then nova will move it to error | 08:07 |
| Xinran | zhenguo, ok got it | 08:08 |
| litao__ | zhenguo: YES | 08:09 |
| litao__ | Power on: call : not set server to error, just record error to db | 08:09 |
| litao__ | Power off: cast: not set server to error, just record error to db | 08:09 |
| litao__ | Reboot: cast: record error to db when server is running, set server to error when server is shutdown | 08:09 |
| zhenguo | litao__: so how users know the power action failed if just record error to db | 08:09 |
| openstackgerrit | Merged openstack/python-moganclient master: Add node aggregate commands support https://review.openstack.org/492002 | 08:09 |
| litao__ | zhenguo: It should use notification to tell users | 08:13 |
| zhenguo | litao__: seems yes | 08:13 |
| zhenguo | litao__: so we can follow nova's way? | 08:13 |
| litao__ | zhenguo: sure, we can | 08:14 |
| zhenguo | litao__: ok, then you can rebase your patch | 08:14 |
| litao__ | zhenguo: ok | 08:14 |
| zhenguo | litao__: thanks | 08:14 |
| litao__ | zhenguo,liusheng ,wanghao,shaohe_feng : pls review my patch in your spare time. | 08:27 |
| litao__ | https://review.openstack.org/#/c/481544/ | 08:27 |
| zhenguo | litao__: sure | 08:34 |
| liusheng | zhenguo, litao__ actually I think it is a bit werid about the api request: POST /v1/manageable_servers/ it looks like creating a manageable server, wdyt ? | 08:35 |
| zhenguo | liusheng: yes, | 08:35 |
| zhenguo | liusheng: but not find other good way | 08:36 |
| liusheng | zhenguo, litao__ how about POST /v1/manageable_servers/{manageable_server_id}/adopt/ | 08:36 |
| zhenguo | liusheng: adopt is a verb | 08:37 |
| zhenguo | liusheng: and we already decided to get rid of the notion adopt | 08:37 |
| liusheng | zhenguo: yes, we can define a custom interface adopt than post | 08:37 |
| litao__ | liusheng: I think it is a bit complexity | 08:37 |
| liusheng | zhenguo: or manage | 08:37 |
| zhenguo | liusheng: that's also weird | 08:38 |
| liusheng | litao__ zhenguo or POST /v1/manageable_servers/{manageable_server_id} with a request body includeing: {'action': manage} | 08:39 |
| litao__ | The resource is manageable_server like server | 08:39 |
| litao__ | liusheng: How do i get the manageable_server_id? | 08:39 |
| liusheng | litao__: yes, but post /v1/manageable_servers/ looks like you will add a manageable server | 08:39 |
| liusheng | litao__: don't we will provide get_all() | 08:40 |
| litao__ | liusheng: yes, i will create a manageable server | 08:40 |
| zhenguo | liusheng: then it should not be POST but PUT | 08:40 |
| liusheng | litao__: by wanghao's patch | 08:40 |
| liusheng | litao__:no, you don't create a manageable server | 08:40 |
| liusheng | litao__: you are managing an existing manageable server | 08:41 |
| litao__ | liusheng: Actually I will manage an existing bare metal node | 08:42 |
| liusheng | zhenguo: not very sure about POST or PUT | 08:42 |
| liusheng | litao__: yes | 08:42 |
| litao__ | liusheng: The manageable server is a resource like server | 08:42 |
| liusheng | litao__: it is "manageable server" before managing, then you manage it, it is a normal mogan server | 08:43 |
| zhenguo | liusheng: POST should be with a specific resource id | 08:43 |
| litao__ | liusheng: So I think the spec sould be modified for list intereface. | 08:43 |
| zhenguo | liusheng: should not be I mean :D | 08:44 |
| liusheng | zhenguo: hah | 08:44 |
| litao__ | zhenguo: what is resource id, you mean bare metal node uuid? | 08:45 |
| liusheng | litao__: do you will put your interface "post" under the same controller class with wanghao's ? | 08:45 |
| liusheng | litao__: I found one is ManageableServersController, another is ManageableServerController | 08:45 |
| zhenguo | litao__: I mean not specific for this manageable servers, but the generic using | 08:46 |
| litao__ | liusheng: The spec design them in one Controller | 08:48 |
| litao__ | zhenguo: more details | 08:49 |
| zhenguo | liusheng: I mean liusheng's suggestion should be be POST if we use URI like /resources/resource_id | 08:50 |
| zhenguo | litao__^^ | 08:50 |
| liusheng | zhenguo: hah | 08:50 |
| zhenguo | liusheng, litao__: POST should be /resources, and PUT/DELETE/GET should be with /resources/resource_id | 08:51 |
| liusheng | zhenguo: maybe my suggestion is not good, but post /v1/manageable_servers/ is really like we adding a manageable server resource | 08:51 |
| liusheng | zhenguo: got it | 08:51 |
| zhenguo | liusheng: yes, aggree, that' cinder's way | 08:52 |
| liusheng | zhenguo litao__ maybe we can keep current way if we don't have a better choice, we can improve in future | 08:55 |
| zhenguo | liusheng: ok, | 08:56 |
| litao__ | liusheng: OK, I agree | 08:56 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Add checks for aggregate affinity_zone https://review.openstack.org/493495 | 09:03 |
| openstackgerrit | Xinran WANG proposed openstack/mogan master: Specify image when do rebuilding https://review.openstack.org/490421 | 09:11 |
| *** wanghao_ has quit IRC | 10:15 | |
| *** wanghao has joined #openstack-mogan | 10:15 | |
| *** wanghao has quit IRC | 10:16 | |
| *** wanghao has joined #openstack-mogan | 10:16 | |
| *** wanghao has quit IRC | 10:16 | |
| *** wanghao has joined #openstack-mogan | 10:17 | |
| *** wanghao has quit IRC | 10:17 | |
| *** wanghao has joined #openstack-mogan | 10:17 | |
| *** wanghao has quit IRC | 10:18 | |
| *** wanghao has joined #openstack-mogan | 10:18 | |
| *** wanghao has quit IRC | 10:19 | |
| *** wanghao has joined #openstack-mogan | 10:19 | |
| *** wanghao has quit IRC | 10:20 | |
| *** wanghao has joined #openstack-mogan | 10:20 | |
| *** wanghao has quit IRC | 10:20 | |
| *** wanghao has joined #openstack-mogan | 10:21 | |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Add checks for aggregate affinity_zone https://review.openstack.org/493495 | 11:11 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: WIP: Add socat console support https://review.openstack.org/493836 | 11:22 |
| *** litao__ has quit IRC | 11:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!