| openstackgerrit | wanghao proposed openstack/mogan master: Refactor exception raise by using mogan exception https://review.openstack.org/475971 | 00:49 |
|---|---|---|
| zhenguo | morning mogan! | 00:51 |
| zhenguo | liusheng: I cleaned up the 'small' flavor in tempest with this https://review.openstack.org/#/c/482430/ | 00:52 |
| wanghao | morning! | 00:52 |
| *** litao__ has joined #openstack-mogan | 00:53 | |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Remove log translations https://review.openstack.org/482825 | 00:56 |
| *** liujiong has joined #openstack-mogan | 01:26 | |
| openstackgerrit | Merged openstack/mogan master: Remove old flavor test mock https://review.openstack.org/482526 | 01:28 |
| openstackgerrit | liusheng proposed openstack/mogan master: Update the quickstart doc about spawning server https://review.openstack.org/483151 | 01:37 |
| liusheng | litao__: please see the commit message of the moganclient patch | 01:51 |
| openstackgerrit | liusheng proposed openstack/mogan master: Update the quickstart doc about spawning server https://review.openstack.org/483151 | 01:56 |
| zhenguo | hi all, the weekly meeting will happen soon, please move to #openstack-meeting | 01:59 |
| shaohe_feng | morning zhenguo | 01:59 |
| zhenguo | shaohe_feng: o/ | 02:00 |
| *** zhuli_ has quit IRC | 02:01 | |
| openstackgerrit | Xinran WANG proposed openstack/mogan master: Add filters to server list API https://review.openstack.org/473323 | 02:04 |
| litao__ | liusheng: OK | 03:01 |
| liusheng | zhenguo: hi zhenguo, I am debuging the detach/attach tempest test, do you konw if the attach vif in Ironic is a async operation ? | 03:02 |
| zhenguo | liusheng: seems not | 03:02 |
| zhenguo | liusheng: I remember it's a call instead of cast | 03:02 |
| zhenguo | liusheng: let me check | 03:02 |
| liusheng | zhenguo: according to the situation, after attaching call return, seems the vif didn't be attached immediatelly | 03:03 |
| zhenguo | liusheng: https://github.com/openstack/ironic/blob/master/ironic/conductor/rpcapi.py#L928 | 03:04 |
| zhenguo | liusheng: maybe some bug with attaching interface? | 03:05 |
| liusheng | zhenguo: that is strange, but if we wait seconds after attaching then detach, the test is ok | 03:05 |
| zhenguo | liusheng: seems it is not always fail | 03:07 |
| liusheng | zhenguo: yes, sometimes, the detaching step of the test returns the interface notfound of the server | 03:08 |
| liusheng | zhenguo: just tested with ironic vif-attach and vif-detach command at same time, it is ok | 03:30 |
| zhenguo | liusheng: wrt https://review.openstack.org/#/c/482430/ , how about adding specific flavor in another patch, the purpose of the patch is change to use IRONIC_RESOURCE_CLASS, and keep the original way of testing | 03:30 |
| zhenguo | liusheng: there's a TODO for you to add specific flavors :D | 03:31 |
| liusheng | zhenguo: yes, I am ok, and have replied you, you can leave the TODO, if you don't fix it, please don't remove it. hah | 03:31 |
| zhenguo | liusheng: sure | 03:32 |
| zhenguo | liusheng: do you think we should keep the FlavorNotFound exception? | 03:35 |
| liusheng | zhenguo: I think we should, we need to consider run tempest if there isn't a default flavor | 03:37 |
| zhenguo | liusheng: ok, will add it back | 03:37 |
| liusheng | zhenguo: the exception an show explicit info of this situation | 03:37 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Use IRONIC_USE_RESOURCE_CLASSES in devstack plugin https://review.openstack.org/482430 | 03:39 |
| *** wanghao_ has joined #openstack-mogan | 04:20 | |
| *** wanghao has quit IRC | 04:21 | |
| zhenguo | shaohe_feng: please have a look at this again https://review.openstack.org/#/c/482430/ | 05:01 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Add indexes to tables https://review.openstack.org/483203 | 05:48 |
| *** zhuli_ has joined #openstack-mogan | 06:14 | |
| zhenguo | liusheng: any new clues about the attach-detach-test-case? | 06:44 |
| liusheng | zhenguo: no :( | 06:56 |
| liusheng | zhenguo: still debuging | 06:57 |
| zhenguo | liusheng: hah | 06:57 |
| *** wanghao has joined #openstack-mogan | 07:05 | |
| *** wanghao_ has quit IRC | 07:08 | |
| zhenguo | liusheng, shaohe_feng, wanghao, litao__, zhangyang: https://etherpad.openstack.org/p/mogan-cfp-sydney | 07:21 |
| zhenguo | please paste your ideas there, thanks | 07:22 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Add indexes to tables https://review.openstack.org/483203 | 07:43 |
| *** wanghao has quit IRC | 07:48 | |
| *** wanghao has joined #openstack-mogan | 07:49 | |
| wanghao | zhenguo: sure | 07:51 |
| zhenguo | wanghao: thanks | 08:10 |
| litao__ | liusheng: If we needn't use '_', should we remove '_' in i18n.py file? | 08:20 |
| liusheng | litao__: why we needn't "_" ? | 08:20 |
| liusheng | litao__: we have use _() in client code | 08:21 |
| litao__ | liusheng: But in your patch, you remove the '_' in you log message | 08:26 |
| liusheng | litao__: don't need to translate log, but _() is not only used in log translation | 08:27 |
| liusheng | litao__: is also used in exception message | 08:28 |
| liusheng | litao__: or help messages | 08:28 |
| litao__ | liusheng: got it | 08:32 |
| liusheng | litao__: thanks | 08:32 |
| zhenguo | liusheng: please remmeber to upadte the placement patch | 08:37 |
| liusheng | zhenguo: yes, I am modifying | 08:38 |
| zhenguo | liusheng: ok, wrt attach-detach, maybe just skip it or add a sleep and improve it later | 08:38 |
| liusheng | zhenguo: hmm, maybe add a sleep and a TODO, and when I debuging the test, seems another problem occurd | 08:39 |
| zhenguo | liusheng: what? | 08:39 |
| liusheng | zhenguo: if I create a server and then delete server, then, I run the tempest, it will fail and say the server not fonud, but the notfound server is the previous deleted server | 08:40 |
| liusheng | zhenguo: it is strange | 08:40 |
| zhenguo | liusheng: in local env? | 08:41 |
| liusheng | zhenguo: yes | 08:41 |
| zhenguo | liusheng: ... | 08:41 |
| liusheng | zhenguo: seems because some unknown reason, the "server = self._resource or self._get_resource(server_uuid)" in controller.py cache the previous server | 08:41 |
| liusheng | zhenguo: just guess so | 08:42 |
| liusheng | zhenguo: s/controller.py/servers.py | 08:42 |
| zhenguo | liusheng: the uuid is same? | 08:42 |
| liusheng | zhenguo: yes, it uuid is the previous deleted server | 08:44 |
| liusheng | zhenguo: but if I restart mogan-api and mogan-engine after delete the server, it is ok | 08:44 |
| liusheng | zhenguo: so I guess may it can cache the previous server's uuid | 08:44 |
| zhenguo | liusheng: seems yes | 08:46 |
| zhenguo | liusheng: not sure what's the reason we use the cache | 08:49 |
| zhenguo | liusheng: what happen if you manually delete then create and delete? | 08:52 |
| liusheng | zhenguo: it seems only occurd in tempest | 08:52 |
| liusheng | zhenguo: but I don't met that previously | 08:53 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the compute_node and compute_port objects and db interfaces https://review.openstack.org/478406 | 08:55 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of listing availability zone api and clean some legacy code https://review.openstack.org/478403 | 08:55 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the methods about updating node resources to Mogan db https://review.openstack.org/478357 | 08:55 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of node listing api of Mogan https://review.openstack.org/478361 | 08:55 |
| openstackgerrit | liusheng proposed openstack/mogan master: Refactor the scheduler to use placement service https://review.openstack.org/477426 | 08:55 |
| zhenguo | liusheng: seems the _get_resouce will be used by policy authorize | 08:58 |
| zhenguo | liusheng: then it will cache the resource, so seems there's no problems | 08:58 |
| liusheng | zhenguo: yes | 08:58 |
| liusheng | zhenguo: no, in my situation, it will cache the server of last request | 08:58 |
| liusheng | zhenguo: not very sure, will git later | 08:59 |
| zhenguo | liusheng: only for tempest? | 08:59 |
| liusheng | zhenguo: seems ye | 08:59 |
| zhenguo | liusheng: the tempest will create then delete the server, right? | 09:14 |
| zhenguo | liusheng: but because you create and delete a server before, so it will have a cache now | 09:14 |
| liusheng | zhenguo: yes, but Notfound error occured when detach the interface, before deleting | 09:15 |
| zhenguo | liusheng: detach interface? | 09:15 |
| liusheng | zhenguo: and the Notfond server isn't the server created by this tempest test | 09:15 |
| liusheng | zhenguo: create a server -- > attach interface --> detach interface --> delete server | 09:16 |
| zhenguo | liusheng: it's the tempest workflow right? | 09:16 |
| liusheng | zhenguo: yes | 09:16 |
| zhenguo | liusheng: without atatch/detach it will be ok? | 09:17 |
| liusheng | zhenguo: seems, yes | 09:17 |
| zhenguo | liusheng: seems I got the reason | 09:18 |
| zhenguo | liusheng: authorize_wsgi have a need_target parameter | 09:19 |
| zhenguo | liusheng: only if it's True we will do _get_resouce to update the cache | 09:19 |
| zhenguo | liusheng: so the parameter on policy check for attach_interface should change to True | 09:19 |
| liusheng | zhenguo: hmm, it is possible | 09:20 |
| zhenguo | liusheng: you can try | 09:20 |
| liusheng | zhenguo: I confuse me a long time :( | 09:20 |
| liusheng | zhenguo: ok, let me update the placement patch firstly | 09:21 |
| zhenguo | liusheng: ok, I will update the detach patch. hah | 09:21 |
| liusheng | zhenguo: thanks | 09:21 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the compute_node and compute_port objects and db interfaces https://review.openstack.org/478406 | 09:25 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of listing availability zone api and clean some legacy code https://review.openstack.org/478403 | 09:25 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the methods about updating node resources to Mogan db https://review.openstack.org/478357 | 09:25 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of node listing api of Mogan https://review.openstack.org/478361 | 09:25 |
| openstackgerrit | liusheng proposed openstack/mogan master: Refactor the scheduler to use placement service https://review.openstack.org/477426 | 09:25 |
| liusheng | zhenguo: ^^ I have updated the placement patch as we dicussed | 09:26 |
| zhenguo | liusheng: thanks, seems it almost ready to land, lol | 09:26 |
| liusheng | zhenguo: hah, you can test and review again | 09:27 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Detach interface for server https://review.openstack.org/461767 | 09:29 |
| zhenguo | liusheng: ok, will test when my env is ready | 09:29 |
| liusheng | zhenguo: thanks, let me run again with your newest patchset | 09:30 |
| zhenguo | liusheng: hah | 09:30 |
| *** wanghao has quit IRC | 09:38 | |
| *** wanghao has joined #openstack-mogan | 09:38 | |
| zhenguo | liusheng: seems I always need to run unstack.sh twice to kill the mogan services | 09:38 |
| *** wanghao has quit IRC | 09:38 | |
| liusheng | zhenguo: why? I didn't found that | 09:39 |
| zhenguo | liusheng: not sure, hah | 09:39 |
| *** wanghao has joined #openstack-mogan | 09:39 | |
| * zhenguo brb | 09:39 | |
| *** wanghao has quit IRC | 09:39 | |
| *** wanghao has joined #openstack-mogan | 09:40 | |
| *** wanghao has quit IRC | 09:40 | |
| *** wanghao has joined #openstack-mogan | 09:41 | |
| *** wanghao has quit IRC | 09:41 | |
| *** wanghao has joined #openstack-mogan | 09:41 | |
| *** wanghao has quit IRC | 09:42 | |
| *** wanghao has joined #openstack-mogan | 09:42 | |
| *** wanghao has quit IRC | 09:42 | |
| *** wanghao has joined #openstack-mogan | 09:43 | |
| *** liusheng has quit IRC | 09:43 | |
| *** wanghao has quit IRC | 09:43 | |
| *** liusheng has joined #openstack-mogan | 09:44 | |
| *** liusheng has quit IRC | 09:44 | |
| *** liusheng has joined #openstack-mogan | 09:45 | |
| liusheng | zhenguo: seems your updates or the detack/attach patch solve my problem | 09:45 |
| *** wanghao_ has joined #openstack-mogan | 09:48 | |
| zhenguo | liusheng: hah | 09:52 |
| zhenguo | liusheng: need to get rid of the False parameter | 09:52 |
| liusheng | zhenguo: I should told you this problem early, hah | 09:54 |
| liusheng | zhenguo: it confuse me a long time | 09:54 |
| zhenguo | liusheng: hah, I also don't quite familiar with that either, | 09:54 |
| zhenguo | liusheng: it's done by shaohe_feng | 09:54 |
| liusheng | zhenguo: oh, but I couldn't think aout that | 09:55 |
| zhenguo | liusheng: hah | 09:55 |
| zhenguo | liusheng: please help to look at this https://review.openstack.org/#/c/483203/ when you got time | 09:55 |
| zhenguo | liusheng: after dinner, my env will be ready for testing the placement patches. | 09:56 |
| * zhenguo away | 09:56 | |
| *** liujiong has quit IRC | 10:22 | |
| openstackgerrit | Merged openstack/mogan master: Use IRONIC_USE_RESOURCE_CLASSES in devstack plugin https://review.openstack.org/482430 | 10:26 |
| zhenguo | liusheng: seems the detach patch need to be rebased as the project-config change | 10:57 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Detach interface for server https://review.openstack.org/461767 | 10:57 |
| openstackgerrit | liusheng proposed openstack/mogan master: Support running api under uWSGI https://review.openstack.org/482057 | 11:15 |
| zhenguo | liusheng: not sure if the server_ids and nodes order will be same between in scheduler and in engine manager | 11:36 |
| zhenguo | liusheng: we set allocations based on that, and then on engine manager we also use the list to assgin node to server | 11:37 |
| liusheng | zhenguo: I confirmed that, it is in same order | 11:40 |
| zhenguo | liusheng: always in the same order? | 11:41 |
| zhenguo | liusheng: if so, it's ok | 11:41 |
| liusheng | zhenguo: zhenguo yes | 11:42 |
| zhenguo | liusheng: I found you implement a delete _allocations_fo_resource_providers | 11:42 |
| liusheng | zhenguo: yes, I think placement should provide the api. hah | 11:42 |
| zhenguo | liusheng: liusheng: hah | 11:43 |
| zhenguo | liusheng: the patch LGTM, only one thing left, please check my comments | 11:49 |
| zhenguo | liusheng: when reverting, seems we should also not delete the allocations | 11:49 |
| openstackgerrit | liusheng proposed openstack/mogan master: Refactor the scheduler to use placement service https://review.openstack.org/477426 | 11:51 |
| liusheng | zhenguo: done | 11:52 |
| zhenguo | liusheng: thanks | 11:52 |
| zhenguo | liusheng: the detach patch just passed the gate, please help to +2 again | 11:53 |
| liusheng | zhenguo: done, hah | 11:53 |
| zhenguo | liusheng: hah | 11:53 |
| * zhenguo away | 11:54 | |
| *** litao__ has quit IRC | 11:59 | |
| openstackgerrit | Merged openstack/mogan master: Update the quickstart doc about spawning server https://review.openstack.org/483151 | 12:03 |
| openstackgerrit | Merged openstack/mogan master: Detach interface for server https://review.openstack.org/461767 | 12:29 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/mogan master: Updated from global requirements https://review.openstack.org/483334 | 12:40 |
| *** liusheng has quit IRC | 17:25 | |
| *** liusheng has joined #openstack-mogan | 17:26 | |
| *** harlowja has quit IRC | 17:27 | |
| *** shaohe_feng has quit IRC | 18:36 | |
| *** shaohe_feng has joined #openstack-mogan | 18:37 | |
| *** harlowja has joined #openstack-mogan | 18:54 | |
| *** harlowja has quit IRC | 20:19 | |
| *** harlowja has joined #openstack-mogan | 20:22 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!