| *** wanghao has joined #openstack-mogan | 00:16 | |
| zhenguo | morning mogan! | 00:54 |
|---|---|---|
| zhenguo | liusheng: cool | 00:56 |
| *** liujiong has joined #openstack-mogan | 00:56 | |
| liusheng | zhenguo: lol | 00:56 |
| zhenguo | liusheng: gate broken again? | 00:57 |
| liusheng | zhenguo: not sure | 00:58 |
| liusheng | zhenguo: seems devstack installed failed | 00:58 |
| zhenguo | liusheng: yes | 00:59 |
| liusheng | zhenguo: this situation occured sometimes | 00:59 |
| liusheng | zhenguo: but don't know shy | 00:59 |
| liusheng | zhenguo: why | 00:59 |
| zhenguo | liusheng: seems not related to us | 01:00 |
| *** openstackgerrit has joined #openstack-mogan | 01:02 | |
| openstackgerrit | liusheng proposed openstack/mogan master: Support running api under uWSGI https://review.openstack.org/482057 | 01:02 |
| wanghao | morning! | 01:17 |
| wanghao | hi guys, patch for managing existing bm is ready, welcome to review, part-one: https://review.openstack.org/#/c/479660/ | 01:20 |
| zhenguo | wanghao: awesome, thanks for the hard work! | 01:21 |
| wanghao | zhenguo: you're welcom | 01:23 |
| *** wanghao_ has joined #openstack-mogan | 01:25 | |
| *** wanghao has quit IRC | 01:25 | |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Fix circular reference error when attaching interface https://review.openstack.org/482757 | 01:35 |
| *** wanghao has joined #openstack-mogan | 01:56 | |
| Xinran | zhenguo, liusheng sorry for the delay. I was not in company yestoday. and thanks for you guy's effort | 01:56 |
| Xinran | zhenguo, liusheng very appreciate | 01:56 |
| liusheng | Xinran: you're welcome :) | 01:56 |
| *** wanghao_ has quit IRC | 01:58 | |
| zhenguo | liusheng: np | 02:03 |
| zhenguo | Xinran^^ | 02:03 |
| zhenguo | wrong person | 02:03 |
| zhenguo | liusheng, Xinran: recheck still got failed | 02:06 |
| liusheng | :( | 02:06 |
| zhenguo | liusheng: but another error | 02:06 |
| zhenguo | liusheng: let me recheck again | 02:06 |
| zhenguo | liusheng: how's the placement patch | 02:09 |
| liusheng | zhenguo: modifying | 02:10 |
| zhenguo | liusheng: ok, let's try to land it by today | 02:11 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the compute_node and compute_port objects and db interfaces https://review.openstack.org/478406 | 02:14 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of listing availability zone api and clean some legacy code https://review.openstack.org/478403 | 02:14 |
| openstackgerrit | liusheng proposed openstack/mogan master: Clean the methods about updating node resources to Mogan db https://review.openstack.org/478357 | 02:14 |
| openstackgerrit | liusheng proposed openstack/mogan master: Get rid of node listing api of Mogan https://review.openstack.org/478361 | 02:14 |
| openstackgerrit | liusheng proposed openstack/mogan master: Refactor the scheduler to use placement service https://review.openstack.org/477426 | 02:14 |
| zhenguo | luyao, Xinran: on my tests, there are still some problems with attaching interfaces, like when ironic attach vif failed, the vif and neutron port info will still associated the server, you can test with some corner cases. | 02:15 |
| zhenguo | luyao, Xinran: and the new devstack will provide two ports for each ironic node, so you can test attach without detaching the port. | 02:15 |
| Xinran | zhenguo, I will check it | 02:19 |
| zhenguo | Xinran: thanks | 02:19 |
| *** litao__ has joined #openstack-mogan | 02:22 | |
| openstackgerrit | liusheng proposed openstack/mogan master: Support running api under uWSGI https://review.openstack.org/482057 | 02:23 |
| luyao | zhenguo, ok, so just clean the info if attach vif failed? | 02:23 |
| zhenguo | luyao: not sure, maybe better to do some cleanup | 02:24 |
| zhenguo | luyao: you finished your thesis? | 02:24 |
| luyao | zhenguo, not yet, do some improvement now ：) | 02:25 |
| zhenguo | luyao: hah | 02:26 |
| luyao | zhenguo, I will fix this bug asap | 02:27 |
| zhenguo | luyao: thanks, and found this bug https://bugs.launchpad.net/mogan/+bug/1703592 | 02:28 |
| openstack | Launchpad bug 1703592 in Mogan "Circular reference error when attaching interface" [Undecided,In progress] - Assigned to Zhenguo Niu (niu-zglinux) | 02:28 |
| luyao | zhenguo, you have fixed this bug? | 02:32 |
| zhenguo | luyao: I think so | 02:32 |
| luyao | zhenguo, hah : ) | 02:32 |
| zhenguo | luyao: need you to confirm with it, hah | 02:33 |
| luyao | zhenguo, OK, : D | 02:33 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan-specs master: Node aggregates https://review.openstack.org/470927 | 02:37 |
| *** wanghao_ has joined #openstack-mogan | 02:43 | |
| *** wanghao has quit IRC | 02:43 | |
| *** wanghao has joined #openstack-mogan | 02:54 | |
| *** wanghao__ has joined #openstack-mogan | 02:55 | |
| *** wanghao has quit IRC | 02:55 | |
| *** wanghao_ has quit IRC | 02:58 | |
| zhenguo | liusheng: the detaching interface patch seems still got failed with test_attach_detach_interface | 03:08 |
| *** wanghao has joined #openstack-mogan | 03:10 | |
| zhenguo | liusheng, shaohe_feng: please help to +1 this https://review.openstack.org/#/c/482484/ | 03:12 |
| *** wanghao__ has quit IRC | 03:14 | |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Clean up unneeded server=True in model_query https://review.openstack.org/482781 | 03:23 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Clean up unneeded server=True in model_query https://review.openstack.org/482781 | 03:34 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Adds aggregates DB model and API https://review.openstack.org/482786 | 03:43 |
| shaohe_feng | zhenguo: OK | 03:47 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Trivial-Fix: Correct KeypairExists exception message https://review.openstack.org/482806 | 05:26 |
| openstackgerrit | Xinran WANG proposed openstack/mogan master: Add filters to server list API https://review.openstack.org/473323 | 05:55 |
| openstackgerrit | wanghao proposed openstack/mogan master: Refactor exception raise by using mogan exception https://review.openstack.org/475971 | 06:01 |
| *** liujiong has quit IRC | 06:19 | |
| *** liujiong has joined #openstack-mogan | 06:19 | |
| *** openstackgerrit has quit IRC | 06:48 | |
| *** liujiong_lj has joined #openstack-mogan | 06:50 | |
| -openstackstatus- NOTICE: FYI, ask.openstack.org is down, review.o.o is slow - please have patience until this is fixed | 06:52 | |
| *** liujiong has quit IRC | 06:54 | |
| *** openstackgerrit has joined #openstack-mogan | 07:11 | |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Remove log translations https://review.openstack.org/482825 | 07:11 |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Remove log translations https://review.openstack.org/482825 | 07:15 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Adds aggregates DB model and API https://review.openstack.org/482786 | 07:26 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan-specs master: Node aggregates https://review.openstack.org/470927 | 07:37 |
| -openstackstatus- NOTICE: Gerrit is going to be restarted, due to low performance | 07:49 | |
| zhenguo | hi all, please help to review the placement patch https://review.openstack.org/#/c/477426/ | 07:53 |
| -openstackstatus- NOTICE: Gerrit has been successfully restarted | 07:56 | |
| zhenguo | liusheng: hi, do you have a plan to add socat console support? | 07:57 |
| liusheng | zhenguo: not yet, I am not very fimilar with it :( | 07:58 |
| zhenguo | liusheng: which will also make it possible for console logs support for us | 07:58 |
| zhenguo | liusheng: haha, just an ironic api to get the console url like shellinabox | 07:58 |
| zhenguo | liusheng: the logic is done by ironic | 07:59 |
| liusheng | zhenguo: will learn it | 07:59 |
| zhenguo | liusheng: need to use physical env for testing | 08:00 |
| liusheng | zhenguo: our physical env need to be restack | 08:00 |
| zhenguo | liusheng: yes | 08:00 |
| zhenguo | liusheng: no rush, just want to introduce it before Pike | 08:00 |
| zhenguo | liusheng: but I think it's easy | 08:01 |
| liusheng | zhenguo: hah, will try | 08:01 |
| zhenguo | liusheng: thanks | 08:01 |
| zhenguo | liusheng: will this break our gate https://review.openstack.org/#/c/482484/ ? | 08:01 |
| liusheng | zhenguo: you can help to take a look at the uwsgi support patch if you get time | 08:01 |
| zhenguo | liusheng: sure | 08:02 |
| zhenguo | liusheng: but after the placement one | 08:02 |
| zhenguo | liusheng: i'm testing it | 08:02 |
| liusheng | zhenguo: actually, I am not very sure, just believe you, hah | 08:02 |
| *** liujiong_lj is now known as liujiong | 08:02 | |
| zhenguo | liusheng: I think our devstack plugin will override the resource_class | 08:03 |
| zhenguo | liusheng: which flavor you are using for tempest test | 08:03 |
| liusheng | zhenguo: ok, thanks, but I assume it need a matched client side patch of the uwsgi support patch | 08:03 |
| liusheng | zhenguo: the "small" | 08:03 |
| zhenguo | liusheng: I will get rid of the small flavor | 08:03 |
| liusheng | zhenguo: yes, I have a TODO in the tempest, that we'd better to get rid of the "small" flavor in tempest | 08:04 |
| zhenguo | liusheng: with this https://review.openstack.org/#/c/482430/ | 08:04 |
| zhenguo | liusheng: we will got a flavor like baremtal_16cpu_1024ram_1024disk | 08:05 |
| zhenguo | liusheng: as ironic will set the resource class itsself like this | 08:05 |
| liusheng | zhenguo: but the flavor name can be simple, right ? | 08:05 |
| liusheng | zhenguo: we just need to match the resource class | 08:05 |
| zhenguo | liusheng: yes, but 'small' seems not that specific | 08:06 |
| zhenguo | liusheng: I like the pattern what ironic used, hah | 08:06 |
| liusheng | zhenguo: yes, I don't mean "small" haha | 08:06 |
| zhenguo | liusheng: so, you will get the 'small' on tempest? | 08:06 |
| zhenguo | liusheng: or just using what devstack created | 08:07 |
| liusheng | zhenguo: may you can try to clean something | 08:08 |
| liusheng | zhenguo: sorry, ignore above words | 08:08 |
| zhenguo | liusheng: will try to test the patch with new project-config | 08:08 |
| zhenguo | liusheng: btw, the detaching patch seems still got failed | 08:08 |
| liusheng | zhenguo: yes, it is easy to create a specific flavor for tempest only | 08:08 |
| liusheng | zhenguo: yes, I have no idea about our unstable gate | 08:09 |
| liusheng | zhenguo: :( | 08:09 |
| zhenguo | liusheng: lol | 08:09 |
| zhenguo | liusheng: seems match a recheck bug, let me recheck it | 08:09 |
| liusheng | zhenguo: need to dig and improve it | 08:09 |
| zhenguo | liusheng: seems most are not related ot us, right? | 08:10 |
| liusheng | zhenguo: yes, only one situation that the console tempest fail sometimes | 08:11 |
| zhenguo | liusheng: ironic is now removing the ssh driver | 08:12 |
| liusheng | zhenguo: how it effect us | 08:12 |
| zhenguo | liusheng: how to we test console | 08:13 |
| zhenguo | liusheng: seems fake ipmidriver can't support console now | 08:13 |
| liusheng | zhenguo: I remember it can only work in physical environment, right ? | 08:14 |
| zhenguo | liusheng: you may need to update the doc about how to test console with ssh driver enabled | 08:14 |
| zhenguo | liusheng: yes | 08:14 |
| zhenguo | liusheng: but ssh driver can provide shellinabox console | 08:14 |
| zhenguo | liusheng: if ssh driver got removed, I think we should make socat as default console, wdyt? and we need to add console log support | 08:15 |
| liusheng | zhenguo: not sure how Nova test it, if there any console tempest based on Nova+Ironic ? | 08:15 |
| zhenguo | liusheng: seems not, I think the tempest is with very low coverage | 08:15 |
| liusheng | zhenguo: yes | 08:15 |
| zhenguo | liusheng: socat console need websocket proxy, we can just copy from nova | 08:16 |
| zhenguo | lol | 08:16 |
| liusheng | zhenguo: so the socat console can avoid these problems ? | 08:16 |
| liusheng | hah | 08:16 |
| zhenguo | liusheng: no | 08:16 |
| zhenguo | liusheng: we can't test console without physical env... | 08:17 |
| zhenguo | liusheng: both for shellinabox and socat | 08:17 |
| liusheng | zhenguo: hmm... | 08:17 |
| zhenguo | liusheng: as we only left a virtualbmc for vm which can't provide console | 08:17 |
| * liusheng seems Mogan develop is expensive.. | 08:18 | |
| zhenguo | liusheng: hah | 08:19 |
| zhenguo | liusheng: maybe virtualbmc can add that support in future | 08:19 |
| liusheng | zhenguo: it has been implemented in Ironic ? | 08:20 |
| zhenguo | liusheng: a sub project of ironic | 08:20 |
| liusheng | zhenguo: oh, need to learn more | 08:20 |
| zhenguo | liusheng: no need to learn, just do it, hah | 08:20 |
| liusheng | hah | 08:21 |
| zhenguo | liusheng, shaohe_feng: I added some trivalfix patches, please help to review, which are simple | 08:28 |
| zhenguo | liusheng: for the keypair patch, I want' to correct the wrong 'KeyPaire' work in error message | 08:37 |
| liusheng | zhenguo: yes, just replied you | 08:37 |
| zhenguo | lusheng: and seems nova itself it's not consistent for the exception name | 08:38 |
| zhenguo | liusheng: I find all exceptions for keypair in nova use Keypair other than that one | 08:38 |
| liusheng | zhenguo: hah | 08:38 |
| liusheng | zhenguo: so maybe you can just fix typo or change all them to consistent | 08:39 |
| zhenguo | liusheng: on exception side, with the patch, it got consistency, lol | 08:40 |
| liusheng | zhenguo: ... alright | 08:41 |
| zhenguo | liusheng: hah | 08:41 |
| liusheng | zhenguo: we may need to confirm if gate got broken after project-config changes | 08:41 |
| zhenguo | liusheng: yes, I just recheked the patch | 08:42 |
| zhenguo | liusheng: but seems need to wait for some time to make project-config take effect | 08:42 |
| liusheng | zhenguo: doesn't it effect immediately ? | 08:43 |
| zhenguo | liusheng: no | 08:43 |
| zhenguo | liusheng: maybe half hour? | 08:43 |
| liusheng | zhenguo: oh | 08:43 |
| *** liujiong has quit IRC | 08:47 | |
| *** liujiong has joined #openstack-mogan | 08:47 | |
| openstackgerrit | liusheng proposed openstack/python-moganclient master: Fix the gate broken by osc-lib updates https://review.openstack.org/482857 | 08:53 |
| zhenguo | liusheng: see my comments for the placement patch, | 09:03 |
| zhenguo | liusheng: if we just delete allocations, the rp can be scheduled again, but the ironic node is in deleting progress, which will cause error. | 09:04 |
| liusheng | zhenguo: I have replied you in previous patch | 09:04 |
| liusheng | zhenguo: I agree we'd better to improve that, but I think it should be done in another patch to add the waiting mechanism in server deletion | 09:04 |
| liusheng | zhenguo: wdyt ? | 09:05 |
| zhenguo | liusheng: how to wait? | 09:05 |
| liusheng | zhenguo: add a time to periodically query the node state | 09:05 |
| zhenguo | liusheng: as I understand the simple way is to get rid of the rp, and wait it come back when the node is available again | 09:05 |
| zhenguo | liusheng: why we need that | 09:05 |
| liusheng | zhenguo: why need to get rid of rp ? I think we should release the rp after deleting | 09:06 |
| zhenguo | liusheng: but only if the ironic node got available the rp should be release | 09:07 |
| liusheng | zhenguo: we need to add a timer to wait the state of node to become "available" and then delete the allocation to release the rp | 09:08 |
| zhenguo | liusheng: a periodic task wait for that costs too much, | 09:08 |
| liusheng | zhenguo: no | 09:08 |
| liusheng | zhenguo: cost what ? | 09:08 |
| zhenguo | liusheng: it's a loop | 09:08 |
| zhenguo | liusheng: and for physical env it may take many hours to be done as the disk erase | 09:09 |
| liusheng | zhenguo: even though you also need to delete the allocation | 09:10 |
| zhenguo | liusheng: yes, I propose to also delete the rp | 09:10 |
| liusheng | zhenguo: the resources updating cannot delete the allocation | 09:10 |
| liusheng | zhenguo: ... | 09:10 |
| zhenguo | liusheng: yes, delete rp and allocations | 09:11 |
| liusheng | zhenguo: that looks weird.. | 09:12 |
| zhenguo | liusheng: why | 09:12 |
| zhenguo | liusheng: but I can't find a way to handle it gracefully | 09:12 |
| liusheng | zhenguo: the allocation in placement was degined to handle the resource consumation | 09:12 |
| zhenguo | liusheng: I really take many hours to delete a node in our productions | 09:12 |
| zhenguo | liusheng: yes its is | 09:13 |
| liusheng | zhenguo: similar situation, the deployment may also take many hours, a waiting task I don't think cost much resourcs | 09:14 |
| zhenguo | liusheng: why delpyment will take that long | 09:14 |
| shaohe_feng | zhenguo: OK | 09:14 |
| zhenguo | liusheng: it will be less than 20 mins | 09:14 |
| *** liujiong has quit IRC | 09:15 | |
| *** liujiong has joined #openstack-mogan | 09:15 | |
| zhenguo | liusheng: deleting node will erase all disks, which may take a day or long to finish | 09:15 |
| zhenguo | liusheng: even if you don't delete the rp when deleting server ,it will get deleted when update resources, right? | 09:17 |
| liusheng | zhenguo: hmm.. another side, the resource updating won't only update the nodes in "available" state | 09:17 |
| zhenguo | liusheng: sure, you will delete rp if it's not 'available' | 09:18 |
| zhenguo | liusheng: but no handle allocations there | 09:18 |
| liusheng | zhenguo: no, the current implemenation won't delete the not available nodes | 09:19 |
| liusheng | zhenguo: it only delete the abnormal state nodes | 09:20 |
| zhenguo | liusheng: which is abnormal state | 09:20 |
| liusheng | zhenguo: that I think is the designe intention of placement | 09:21 |
| liusheng | zhenguo: e.g. state of a node is available, but instance_uuid of the node is not None | 09:21 |
| liusheng | zhenguo: or nodes in error state | 09:22 |
| zhenguo | liusheng: seems yes | 09:22 |
| liusheng | zhenguo: the rps or inventories in placement don't represent only the resources can be consumed | 09:23 |
| zhenguo | liusheng: what others it for? | 09:23 |
| liusheng | zhenguo: the also includes the resources have been consumed | 09:24 |
| zhenguo | liusheng: yes, I know, | 09:24 |
| zhenguo | liusheng: so we can provide node list for aggregates with that | 09:24 |
| zhenguo | liusheng: but wait for status is too long, that's why nova don't wait that | 09:26 |
| liusheng | zhenguo: hmm, if so how aobut improve the resource updating to also update allocations | 09:28 |
| zhenguo | liusheng: you mean check the node status? | 09:28 |
| liusheng | zhenguo: yes, and also update the allocations in placement | 09:29 |
| zhenguo | liusheng: if the node is avaialble, then delete the allocations | 09:29 |
| liusheng | zhenguo: if state of node is available, need to check if there is allocation items assinged to this node | 09:29 |
| zhenguo | liusheng: seems a good choice | 09:29 |
| liusheng | zhenguo: yes, | 09:29 |
| zhenguo | liusheng: please go ahead, hah | 09:30 |
| liusheng | zhenguo: ok, will try | 09:30 |
| zhenguo | liusheng: thanks | 09:30 |
| * zhenguo brb | 09:31 | |
| *** wanghao has quit IRC | 09:32 | |
| liusheng | zhenguo: please take a look at https://review.openstack.org/#/c/482857/ if you get time | 09:33 |
| zhenguo | liusheng: ok | 09:42 |
| zhenguo | liusheng: when deleting server, should we delete the allocations if the rp is gone? | 09:42 |
| zhenguo | liusheng: seems the project-config patch really break the gate, hah | 09:44 |
| liusheng | zhenguo: :( | 09:44 |
| liusheng | zhenguo: placement didn't provide an api to delete the allocations of a provider :( | 09:45 |
| liusheng | zhenguo: only provide list the allocations of a provider | 09:45 |
| zhenguo | liusheng: hah | 09:45 |
| liusheng | zhenguo: I guess it will provide that api | 09:45 |
| zhenguo | liusheng: if so, let's wait for that happen | 09:45 |
| openstackgerrit | Merged openstack/python-moganclient master: Fix the gate broken by osc-lib updates https://review.openstack.org/482857 | 09:50 |
| *** liujiong has quit IRC | 10:23 | |
| openstackgerrit | Merged openstack/mogan master: Trivial-Fix: Correct KeypairExists exception message https://review.openstack.org/482806 | 10:23 |
| openstackgerrit | Merged openstack/mogan master: Clean up unneeded server=True in model_query https://review.openstack.org/482781 | 10:24 |
| openstackgerrit | Merged openstack/mogan master: Fix circular reference error when attaching interface https://review.openstack.org/482757 | 10:24 |
| zhenguo | liusheng: seems the gate isn't broken | 11:21 |
| zhenguo | liusheng: need your +A for this https://review.openstack.org/#/c/482526/ | 11:34 |
| openstackgerrit | Zhenguo Niu proposed openstack/mogan master: Use IRONIC_USE_RESOURCE_CLASSES in devstack plugin https://review.openstack.org/482430 | 11:54 |
| * zhenguo away | 11:54 | |
| *** litao__ has quit IRC | 11:56 | |
| *** zhenguo_ has joined #openstack-mogan | 12:40 | |
| *** zhenguo has quit IRC | 12:41 | |
| *** zhenguo_ is now known as zhenguo | 12:41 | |
| *** liusheng has quit IRC | 15:42 | |
| *** liusheng has joined #openstack-mogan | 15:43 | |
| *** zhuli_ has joined #openstack-mogan | 16:00 | |
| *** zhuli_ has quit IRC | 18:40 | |
| *** harlowja has quit IRC | 19:17 | |
| *** harlowja has joined #openstack-mogan | 20:02 | |
| *** wanghao has joined #openstack-mogan | 23:54 | |
| *** zhuli_ has joined #openstack-mogan | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!