| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-01-19 | |||
| 19:39:54 | sean-k-mooney | sdague: mind shareing? whats the limitation? | |
| 19:40:48 | sdague | sean-k-mooney: in order to match jgit's merge behavior, zuul is using a git merge strategy that falls over after 6 way merge | |
| 19:41:12 | openstackgerrit | Jay Pipes proposed openstack/os-vif: Correct py27 assumption that filter() returns list https://review.openstack.org/422838 | |
| 19:41:14 | jaypipes | dims: ^^ | |
| 19:41:26 | sdague | which means when you get a whole stream of unchained commits eventually they merge fail | |
| 19:41:46 | sdague | requirements upper-constraints auto generated items fall into that category | |
| 19:42:11 | sean-k-mooney | sdague: ah ok | |
| 19:42:41 | sean-k-mooney | and im guessing if we iterativly merged the patches instade of 6+ way merge it would break compaiblity with jgit | |
| 19:43:20 | sdague | sean-k-mooney: yeh, it would instead look like it was going to pass, get to top of queue, and gerrit would fail it | |
| 19:43:29 | sdague | which used to happen | |
| 19:44:01 | sdague | this fails things in a few more situations though, so it's kind of a toss up as to when you get kicked in the pants | |
| 19:44:46 | sean-k-mooney | sdague: actully looking at the zull gate queue specifcally for the patch mriedem was interested in https://review.openstack.org/#/c/422709 | |
| 19:45:14 | sean-k-mooney | there are 2 other requirements changes in front of it that will cause the rest of the queue to restart when they merge | |
| 19:45:22 | sdague | no | |
| 19:45:40 | sdague | the symptom was a merge fail when it entered the gate queue | |
| 19:45:47 | sdague | so it was a black dot | |
| 19:45:49 | sean-k-mooney | is there any reason we currently allow multiple changes to the same repo to be cogated | |
| 19:46:06 | sdague | sean-k-mooney: as opposed to? | |
| 19:46:41 | sean-k-mooney | not starting the subsequent job until the first one passed or filed. | |
| 19:47:07 | sean-k-mooney | currentlyu we waste vms runing jobs that will be restarted when the first patch merges | |
| 19:47:33 | openstackgerrit | Sergey Nikitin proposed openstack/nova: do not merge https://review.openstack.org/422841 | |
| 19:47:35 | sdague | sean-k-mooney: no, we don't | |
| 19:47:48 | sdague | the merge is done early | |
| 19:47:54 | sdague | if it fails, no vms are started | |
| 19:48:30 | sean-k-mooney | if you go to http://status.openstack.org/zuul/ and filter on 422709 when the current top patch 422502 merges the requirements job belowit will be restarted | |
| 19:49:03 | sdague | sean-k-mooney: why do you think that it will be? | |
| 19:49:21 | sean-k-mooney | i just watch the q restart all jobs | |
| 19:49:30 | sdague | yeh, that's because top of gate failed | |
| 19:49:44 | sean-k-mooney | oh ok i tagut it passed sorry | |
| 19:50:04 | sdague | I'm pretty sure it's because top of gate failed | |
| 19:51:04 | sdague | https://review.openstack.org/#/c/292570/ just got a -2 10 minutes ago | |
| 19:51:06 | sean-k-mooney | i wasnet paying enough attention to note down which patch it was so not sure. i noticed that other patches merged and the queue did not resate so your proably correct in assuming that it failed | |
| 19:55:38 | dims | jaypipes : does "if dup_filter:" on an iterator work properly? | |
| 20:07:15 | jaypipes | dims: not sure... | |
| 20:07:22 | jaypipes | (and I need to run out for a couple hours...) | |
| 20:14:19 | sean-k-mooney | dims: it looks like an iterator over an empty list or full list both evaluate as true | |
| 20:15:50 | sean-k-mooney | dims: im not sure what the input to the lamda fuction is but if filter retruns an interator it will always enter the if dup_filter block | |
| 20:16:20 | dims | sean-k-mooney : so when we do a filter, immediately turn it into a list would work | |
| 20:22:43 | openstackgerrit | Timofey Durakov proposed openstack/nova: Make all datastamps formats equal https://review.openstack.org/421981 | |
| 20:24:28 | tdurakov | hi folks, who could review small fix for notification timestam formats? | |
| 20:27:39 | tdurakov | gibi: ping | |
| 20:52:03 | mriedem | nova meeting in 8 minutes | |
| 20:53:45 | jroll | omg I'm so excited | |
| 20:54:36 | tdurakov | mriedem: if you have a minute, could you please a patch above? | |
| 20:58:13 | mriedem | tdurakov: does that not require a change to the samples? | |
| 21:00:24 | tdurakov | mriedem: according to the previous run(see ps1) I'd say it doesn't(surprise!!!) | |
| 21:00:30 | tdurakov | at least for me | |
| 21:09:55 | slaweq | mriedem: hello, can You have a look on https://review.openstack.org/#/c/420176/ ? | |
| 21:09:58 | slaweq | thx in advance | |
| 21:23:13 | tdurakov | mriedem: so, for the that notifications fix... I assume there should be no notification versions bump, as object versions are remain the same, right? | |
| 21:27:08 | mriedem | tdurakov: well, the format on the thing getting sent has changed | |
| 21:28:37 | tdurakov | mriedem: what is the proper way to bump versions then? | |
| 21:31:31 | mriedem | tdurakov: i'm not sure in this case, and that's why i'm surprised the existing samples tests don't fail | |
| 21:31:40 | mriedem | btw there is a pep8 failure on that change | |
| 21:31:46 | tdurakov | already fixed | |
| 21:31:55 | tdurakov | will resubmit in a minute | |
| 21:32:02 | tdurakov | checking py3 too | |
| 21:33:23 | tdurakov | mriedem: tbh, I'm surprised that nova explicitly handle datetime formatting for notifications, I'd expect it to be managed by serializer | |
| 21:38:59 | tdurakov | mriedem: https://github.com/openstack/nova/blob/master/nova/objects/instance.py#L103 - is this todo valid? | |
| 21:51:18 | mriedem | tdurakov: yes, but will require a ton of work to get there | |
| 21:51:49 | tdurakov | should I rollback that to instance.get(...) ? | |
| 21:52:04 | mriedem | tdurakov: if that's what the getattr was for then it's fine to leave getattr | |
| 21:52:33 | tdurakov | mriedem: yes, that was for that todo | |
| 21:53:54 | openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix path parameters in os-hypervisors.inc https://review.openstack.org/416911 | |
| 21:54:11 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 7 test cases https://review.openstack.org/395436 | |
| 21:54:26 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 3 test cases https://review.openstack.org/394738 | |
| 21:54:36 | openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix parameters in os-hosts.inc https://review.openstack.org/416444 | |
| 21:54:47 | openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix deprecated proxy API parameters https://review.openstack.org/416882 | |
| 21:55:00 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Avoid forcing translation on logging calls https://review.openstack.org/413876 | |
| 21:55:15 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Enable cold migration with target host(1/2) https://review.openstack.org/408955 | |
| 21:55:42 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Enable cold migration with target host(2/2) https://review.openstack.org/408964 | |
| 21:56:33 | openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix keypair API parameters https://review.openstack.org/420626 | |
| 21:56:48 | openstackgerrit | Takashi NATSUME proposed openstack/nova: Add functional tests for cold migration to same host https://review.openstack.org/414926 | |
| 21:56:59 | openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Add parameters in cold migrate action https://review.openstack.org/410042 | |
| 21:57:11 | openstackgerrit | Takashi NATSUME proposed openstack/python-novaclient: Microversion 2.42 - Enable cold migration with target host https://review.openstack.org/406707 | |
| 22:03:21 | openstackgerrit | Timofey Durakov proposed openstack/nova: Make all datastamps formats equal https://review.openstack.org/421981 | |
| 22:03:55 | tdurakov | mriedem: fix for the pep8 is done, I wonder do we have any issues with py35? | |
| 22:09:03 | mriedem | tdurakov: yes, py35 job is broken until https://review.openstack.org/#/c/422709/ is merged | |
| 22:09:47 | tdurakov | mriedem: ok, anything else on the change? | |
| 22:11:52 | mriedem | tdurakov: i'd like gibi to take a look at it, i'm not ready to vote either way right now | |
| 22:11:58 | mriedem | kind of distracted with other things atm | |
| 22:14:50 | cfriesen | does anyone know of a bug open for the fact that running "nova show <instance>" as the admin user gives useless output if the instance was booted with a large user-data file? | |
| 22:16:04 | cfriesen | (the encoded user-data file contents are included in the output, which messes up the formatting) | |
| 22:18:52 | cfriesen | I guess the solution is to explicitly specify the "--wrap" option? | |
| 22:51:50 | openstackgerrit | Steven Webster proposed openstack/nova: Release PCI devices on drop_move_claim() https://review.openstack.org/370374 | |
| 22:54:49 | mriedem | mox3 fix is #2 in the gate, | |
| 22:54:53 | mriedem | so we should be unblocked real soon now | |
| 22:58:36 | mriedem | and it's merged | |
| 23:10:26 | mriedem | jaypipes: easy peasy for you https://review.openstack.org/#/c/410737/ | |
| 23:11:12 | jaypipes | mwhahaha: sorry, had to run out. but bottom line I think we're closer to each other's viewpoints than either of us realize. | |
| 23:11:28 | mwhahaha | jaypipes: it's all good | |
| 23:12:03 | jaypipes | mwhahaha: and I really do appreciate yours and others feedback. It is extremely valuable. We just need to do more (and earlier) of these types of conversations. | |
| 23:13:23 | jaypipes | mriedem: awesomesauce on the mox3 gate.. | |
| 23:16:12 | jaypipes | mriedem: done. | |
| 23:26:36 | bauzas | jaypipes: back there | |
| 23:26:46 | bauzas | jaypipes: I just saw the meeting logs | |
| 23:27:02 | bauzas | jaypipes: are you planning to present something at the Summit for the placement ? | |
| 23:27:16 | bauzas | jaypipes: if so, I'd love to help you if you accept it | |
| 23:27:51 | jaypipes | bauzas: yes, I already submitted two talks. | |
| 23:28:25 | bauzas | jaypipes: meh, okay | |