openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/313743 | 00:03 |
---|---|---|
*** shaohe_feng has joined #openstack-nova | 00:04 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 00:05 |
*** rbridgeman_ has joined #openstack-nova | 00:07 | |
*** shaohe_feng has quit IRC | 00:08 | |
*** mriedem_afk has quit IRC | 00:08 | |
*** mriedem has joined #openstack-nova | 00:08 | |
*** armax has joined #openstack-nova | 00:08 | |
*** shaohe_feng has joined #openstack-nova | 00:08 | |
*** rbridgeman__ has joined #openstack-nova | 00:10 | |
*** Jeffrey4l_ has joined #openstack-nova | 00:10 | |
armax | mriedem: ping re https://review.openstack.org/#/c/299028/ | 00:11 |
armax | mriedem: you got a sec? | 00:11 |
mriedem | armax: sure | 00:12 |
armax | kevinbenton and I were going over the two backports and we were considering the possibility to bring these into Kilo in a way that’s possible to disable the filters if things go awry | 00:13 |
*** rbridgeman_ has quit IRC | 00:13 | |
* tbachman looks at his “to do” list, realizes that he had something on his calendar for nova :-o | 00:13 | |
kevinbenton | so it would be configurable with a cfg option essentially | 00:13 |
armax | because the fact that Kilo will be closed after this fix will prevent us from doing anything about it | 00:13 |
armax | the cherry picks are not clean anyway | 00:14 |
armax | and especially the ovs change looks totally different from the master fix | 00:14 |
mriedem | although the functional tests pass | 00:15 |
armax | so I see two approaches, one a) not allow the backport b) allow the backport with the ability to switch it off if things go bad | 00:15 |
*** rfolco has quit IRC | 00:15 | |
armax | mriedem: true, but the extent of the coverage is limited (single node not at scale, etc) | 00:15 |
mriedem | yeah | 00:15 |
mriedem | we've backported workaround bug fixes in nova before where they have a config option to enable/disable them | 00:15 |
armax | mriedem: I am a chicken, and backporting this as is makes me a bit unease | 00:16 |
armax | uneasy | 00:16 |
armax | mriedem: if there’s a precendent and since it’s a bw compact config change, perhaps that’s the way we allow it, and we can obviously have it on by default | 00:16 |
mriedem | if we're uneasy about it, to the point that we want to add a config option to disable it, i'm also totally fine with just saying we're not going to do them and do kilo eol | 00:17 |
mriedem | if people need them downstream, they could get them from what's proposed at their own risk | 00:17 |
kevinbenton | true, at least we know they will always patch cleanly since kilo is EOL | 00:17 |
armax | mriedem, kevinbenton: yes, the patch is there for consumption by anyone | 00:18 |
*** shaohe_feng has quit IRC | 00:18 | |
kevinbenton | do we have a pointer to it in the security advisory? | 00:18 |
*** gongysh has quit IRC | 00:19 | |
*** shaohe_feng has joined #openstack-nova | 00:19 | |
mriedem | there isn't a published ossa yet | 00:19 |
mriedem | there is a draft of the description in the bug report by tristanC | 00:20 |
mriedem | i think i've seen OSSAs that say it impacts a certain release but it's not fixed in that release because EOL | 00:20 |
armax | so are ruling out option b)? if kevinbenton put effort in respinning the patches with that approach, I am in | 00:21 |
mriedem | actually, e.g. https://bugs.launchpad.net/nova/+bug/1516765 we just marked juno as won't fix | 00:21 |
openstack | Launchpad bug 1516765 in OpenStack Compute (nova) liberty "[OSSA 2016-002] xenapi: volume_utils._parse_volume_info can leak connection password via StorageError (CVE-2015-8749)" [High,Fix committed] - Assigned to Matt Riedemann (mriedem) | 00:21 |
kevinbenton | it wouldn't take too long to add the config option, i could finish it up within the hour | 00:21 |
mriedem | and the ossa just didn't mention juno http://lists.openstack.org/pipermail/openstack-announce/2016-January/000916.html | 00:22 |
mriedem | up to you guys, i'm fine either way | 00:22 |
armax | mriedem: thanks, it’s good to know we’d be not in breach of some code of conduct :) | 00:22 |
kevinbenton | ok, armax, do you want me to respin with the config option? | 00:23 |
armax | kevinbenton: if there’s a prior example and you have the cycles to respin that now, we’d have time until the end of today to put them in the merge queue | 00:23 |
kevinbenton | armax: ack | 00:25 |
armax | kevinbenton: otherwise we’ll just live with the documented issue and a pointer to a fix | 00:25 |
*** diogogmt has joined #openstack-nova | 00:25 | |
*** shaohe_feng has quit IRC | 00:28 | |
*** shaohe_feng has joined #openstack-nova | 00:29 | |
*** Sukhdev has quit IRC | 00:30 | |
*** Sukhdev has joined #openstack-nova | 00:31 | |
*** abitha has quit IRC | 00:31 | |
alex_xu | alaski: yeah, those policy check for legacy v2 api will be removed, it's on my todo. | 00:33 |
*** ducttape_ has joined #openstack-nova | 00:35 | |
*** browne has joined #openstack-nova | 00:36 | |
*** Sukhdev has quit IRC | 00:37 | |
alex_xu | mriedem: just found we return 400 for marker not found https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/servers.py#L421 | 00:37 |
mriedem | alex_xu: :( | 00:38 |
mriedem | same for flavors | 00:38 |
*** shaohe_feng has quit IRC | 00:38 | |
alex_xu | mriedem: just found that also | 00:39 |
mriedem | i guess we need to update the servers inc then since i made sdague change that | 00:39 |
alex_xu | yea, nvm, i can cook a patch for the doc later | 00:39 |
*** shaohe_feng has joined #openstack-nova | 00:40 | |
*** salv-orlando has joined #openstack-nova | 00:42 | |
*** nic has quit IRC | 00:42 | |
*** ajo has joined #openstack-nova | 00:44 | |
openstackgerrit | Alex Xu proposed openstack/nova: Remove 404 for list and details action servers https://review.openstack.org/314355 | 00:44 |
*** markmc has joined #openstack-nova | 00:45 | |
alex_xu | mriedem: ^ done for this 1 mins, not later anymore... | 00:45 |
*** mriedem1 has joined #openstack-nova | 00:47 | |
*** ducttape_ has quit IRC | 00:47 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/313743 | 00:48 |
*** mriedem has quit IRC | 00:48 | |
*** salv-orlando has quit IRC | 00:48 | |
*** shaohe_feng has quit IRC | 00:49 | |
*** dmellado has joined #openstack-nova | 00:49 | |
*** shaohe_feng has joined #openstack-nova | 00:50 | |
*** Swami has quit IRC | 00:53 | |
*** cheneydc has joined #openstack-nova | 00:55 | |
*** unicell1 has quit IRC | 00:55 | |
*** unicell has joined #openstack-nova | 00:56 | |
*** mriedem1 is now known as mriedem | 00:56 | |
*** shaohe_feng has quit IRC | 00:59 | |
*** hoangcx has joined #openstack-nova | 00:59 | |
*** gongysh has joined #openstack-nova | 00:59 | |
*** shaohe_feng has joined #openstack-nova | 01:00 | |
*** iceyao has joined #openstack-nova | 01:01 | |
*** sdake has joined #openstack-nova | 01:02 | |
*** iceyao_ has joined #openstack-nova | 01:05 | |
*** fishcried has joined #openstack-nova | 01:05 | |
*** mriedem1 has joined #openstack-nova | 01:06 | |
*** fishcried has quit IRC | 01:07 | |
*** iceyao has quit IRC | 01:08 | |
*** mriedem has quit IRC | 01:08 | |
*** shaohe_feng has quit IRC | 01:09 | |
*** ducttape_ has joined #openstack-nova | 01:10 | |
*** shaohe_feng has joined #openstack-nova | 01:10 | |
*** ljxiash has joined #openstack-nova | 01:14 | |
*** baoli has quit IRC | 01:16 | |
*** browne has quit IRC | 01:16 | |
*** baoli has joined #openstack-nova | 01:16 | |
*** apoorvad has quit IRC | 01:16 | |
*** unicell1 has joined #openstack-nova | 01:17 | |
*** unicell has quit IRC | 01:17 | |
*** shaohe_feng has quit IRC | 01:19 | |
*** shaohe_feng has joined #openstack-nova | 01:20 | |
*** baoli has quit IRC | 01:22 | |
*** kengo_sakai has joined #openstack-nova | 01:23 | |
*** mriedem1 has quit IRC | 01:24 | |
*** ljxiash has quit IRC | 01:25 | |
*** ducttape_ has quit IRC | 01:25 | |
*** zenoway has joined #openstack-nova | 01:26 | |
*** jinxing has joined #openstack-nova | 01:28 | |
*** thorst_ has joined #openstack-nova | 01:28 | |
*** shaohe_feng has quit IRC | 01:30 | |
*** zenoway has quit IRC | 01:30 | |
*** EinstCrazy has joined #openstack-nova | 01:31 | |
*** shaohe_feng has joined #openstack-nova | 01:33 | |
*** baoli has joined #openstack-nova | 01:34 | |
*** yuywz__ is now known as yuywz | 01:36 | |
*** ctrath has joined #openstack-nova | 01:38 | |
*** sacharya has joined #openstack-nova | 01:39 | |
*** mdrabe has joined #openstack-nova | 01:40 | |
*** shaohe_feng has quit IRC | 01:40 | |
*** shaohe_feng has joined #openstack-nova | 01:41 | |
*** roeyc has joined #openstack-nova | 01:41 | |
*** sacharya has quit IRC | 01:44 | |
*** diogogmt has quit IRC | 01:44 | |
*** salv-orlando has joined #openstack-nova | 01:44 | |
*** baoli has quit IRC | 01:48 | |
*** baoli has joined #openstack-nova | 01:48 | |
*** ducttape_ has joined #openstack-nova | 01:49 | |
*** BjoernT has joined #openstack-nova | 01:49 | |
*** gongysh has quit IRC | 01:50 | |
*** pwnall has quit IRC | 01:50 | |
*** shaohe_feng has quit IRC | 01:50 | |
*** salv-orlando has quit IRC | 01:51 | |
*** shaohe_feng has joined #openstack-nova | 01:51 | |
*** Daisy has joined #openstack-nova | 01:53 | |
*** unicell1 has quit IRC | 01:55 | |
*** shaohe_feng has quit IRC | 02:01 | |
*** shaohe_feng has joined #openstack-nova | 02:01 | |
*** roeyc has quit IRC | 02:01 | |
*** sacharya has joined #openstack-nova | 02:02 | |
*** yamahata has joined #openstack-nova | 02:03 | |
*** baoli has quit IRC | 02:03 | |
*** EinstCrazy has quit IRC | 02:05 | |
*** EinstCrazy has joined #openstack-nova | 02:08 | |
*** fishcried has joined #openstack-nova | 02:09 | |
*** shaohe_feng has quit IRC | 02:11 | |
*** gjayavelu has quit IRC | 02:11 | |
*** shaohe_feng has joined #openstack-nova | 02:12 | |
*** ijw_ has quit IRC | 02:12 | |
*** zenoway has joined #openstack-nova | 02:13 | |
*** ayoung has quit IRC | 02:17 | |
*** zenoway has quit IRC | 02:17 | |
*** woodster_ has quit IRC | 02:18 | |
*** houming has joined #openstack-nova | 02:18 | |
*** ctrath has quit IRC | 02:19 | |
*** shaohe_feng has quit IRC | 02:21 | |
*** shaohe_feng has joined #openstack-nova | 02:22 | |
*** gongysh has joined #openstack-nova | 02:23 | |
*** amotoki has quit IRC | 02:25 | |
*** hieulq_ has joined #openstack-nova | 02:25 | |
*** vilobhmm11 has joined #openstack-nova | 02:27 | |
*** shaohe_feng has quit IRC | 02:31 | |
*** shaohe_feng has joined #openstack-nova | 02:32 | |
*** armax_ has joined #openstack-nova | 02:34 | |
*** ljxiash has joined #openstack-nova | 02:35 | |
*** armax has quit IRC | 02:37 | |
*** armax_ is now known as armax | 02:37 | |
openstackgerrit | Alex Xu proposed openstack/nova: Remove 404 for list and details actions of servers https://review.openstack.org/314355 | 02:38 |
*** shaohe_feng has quit IRC | 02:42 | |
*** ljxiash has quit IRC | 02:42 | |
*** hieulq_ has quit IRC | 02:42 | |
*** shaohe_feng has joined #openstack-nova | 02:42 | |
*** allen_gao has quit IRC | 02:44 | |
*** ducttape_ has quit IRC | 02:46 | |
*** allen_gao has joined #openstack-nova | 02:48 | |
*** zenoway has joined #openstack-nova | 02:49 | |
*** vtech has quit IRC | 02:51 | |
*** coreywright has quit IRC | 02:51 | |
*** shaohe_feng has quit IRC | 02:52 | |
*** amotoki has joined #openstack-nova | 02:52 | |
*** coreywright has joined #openstack-nova | 02:52 | |
*** shaohe_feng has joined #openstack-nova | 02:52 | |
*** salv-orlando has joined #openstack-nova | 02:53 | |
*** BjoernT has quit IRC | 02:53 | |
*** tjones has joined #openstack-nova | 02:53 | |
*** zenoway has quit IRC | 02:53 | |
*** salv-orlando has quit IRC | 02:54 | |
*** vtech has joined #openstack-nova | 02:56 | |
*** vladikr has quit IRC | 02:57 | |
*** allen_gao has quit IRC | 02:57 | |
*** amotoki has quit IRC | 03:00 | |
*** shaohe_feng has quit IRC | 03:02 | |
*** shaohe_feng has joined #openstack-nova | 03:03 | |
*** allen_gao has joined #openstack-nova | 03:04 | |
*** Sukhdev has joined #openstack-nova | 03:05 | |
*** rhagarty has quit IRC | 03:06 | |
*** sdake has quit IRC | 03:07 | |
*** amotoki has joined #openstack-nova | 03:10 | |
*** shaohe_feng has quit IRC | 03:12 | |
*** shaohe_feng has joined #openstack-nova | 03:13 | |
*** rfolco has joined #openstack-nova | 03:17 | |
*** akshai has joined #openstack-nova | 03:18 | |
*** Jeffrey4l_ has quit IRC | 03:18 | |
*** ljxiash has joined #openstack-nova | 03:18 | |
*** Jeffrey4l_ has joined #openstack-nova | 03:19 | |
*** gongysh has quit IRC | 03:20 | |
*** gjayavelu has joined #openstack-nova | 03:22 | |
*** gjayavelu has quit IRC | 03:22 | |
*** gjayavelu has joined #openstack-nova | 03:22 | |
*** shaohe_feng has quit IRC | 03:23 | |
*** shaohe_feng has joined #openstack-nova | 03:23 | |
*** zenoway has joined #openstack-nova | 03:24 | |
*** rbridgeman__ has quit IRC | 03:25 | |
*** moshele has joined #openstack-nova | 03:26 | |
*** EinstCrazy has quit IRC | 03:28 | |
*** EinstCrazy has joined #openstack-nova | 03:28 | |
*** links has joined #openstack-nova | 03:29 | |
*** zenoway has quit IRC | 03:29 | |
*** dkliban has left #openstack-nova | 03:30 | |
*** amotoki has quit IRC | 03:31 | |
*** shaohe_feng has quit IRC | 03:33 | |
*** thorst_ has quit IRC | 03:33 | |
*** jinxing has quit IRC | 03:33 | |
*** shaohe_feng has joined #openstack-nova | 03:33 | |
*** thorst_ has joined #openstack-nova | 03:34 | |
*** Vek has quit IRC | 03:37 | |
*** thorst_ has quit IRC | 03:38 | |
*** amotoki has joined #openstack-nova | 03:39 | |
*** dimtruck is now known as zz_dimtruck | 03:40 | |
*** zz_dimtruck is now known as dimtruck | 03:40 | |
*** ayoung has joined #openstack-nova | 03:42 | |
*** thorst_ has joined #openstack-nova | 03:42 | |
*** shaohe_feng has quit IRC | 03:43 | |
openstackgerrit | zhufl proposed openstack/nova: Correct some misspell words in nova https://review.openstack.org/314406 | 03:43 |
openstackgerrit | Takashi NATSUME proposed openstack/nova: WIP: Add versioned instance usage notifications https://review.openstack.org/279416 | 03:44 |
*** shaohe_feng has joined #openstack-nova | 03:46 | |
*** EinstCra_ has joined #openstack-nova | 03:46 | |
*** EinstCrazy has quit IRC | 03:48 | |
*** zenoway has joined #openstack-nova | 03:49 | |
*** shaohe_feng has quit IRC | 03:53 | |
*** zenoway has quit IRC | 03:54 | |
*** shaohe_feng has joined #openstack-nova | 03:54 | |
*** dimtruck is now known as zz_dimtruck | 03:55 | |
*** prateek has joined #openstack-nova | 03:56 | |
*** salv-orlando has joined #openstack-nova | 03:58 | |
*** salv-orlando has quit IRC | 04:02 | |
*** shaohe_feng has quit IRC | 04:04 | |
*** vilobhmm11 has quit IRC | 04:04 | |
*** Daisy has quit IRC | 04:05 | |
*** Daisy has joined #openstack-nova | 04:05 | |
*** shaohe_feng has joined #openstack-nova | 04:08 | |
*** oomichi_ has joined #openstack-nova | 04:08 | |
*** Daisy_ has joined #openstack-nova | 04:09 | |
*** xenogear has quit IRC | 04:10 | |
*** Daisy has quit IRC | 04:10 | |
*** vilobhmm11 has joined #openstack-nova | 04:10 | |
*** zz_dimtruck is now known as dimtruck | 04:10 | |
*** Daisy_ has quit IRC | 04:12 | |
*** Daisy has joined #openstack-nova | 04:12 | |
*** vilobhmm11 has quit IRC | 04:13 | |
*** sdake has joined #openstack-nova | 04:13 | |
*** shaohe_feng has quit IRC | 04:14 | |
*** shaohe_feng has joined #openstack-nova | 04:14 | |
*** Daisy has quit IRC | 04:17 | |
*** dimtruck is now known as zz_dimtruck | 04:20 | |
*** vilobhmm11 has joined #openstack-nova | 04:20 | |
*** shaohe_feng has quit IRC | 04:24 | |
*** shaohe_feng has joined #openstack-nova | 04:25 | |
*** vilobhmm11 has quit IRC | 04:25 | |
*** unicell has joined #openstack-nova | 04:26 | |
*** shaohe_feng has quit IRC | 04:34 | |
*** iceyao_ has quit IRC | 04:35 | |
*** shaohe_feng has joined #openstack-nova | 04:35 | |
*** iceyao has joined #openstack-nova | 04:35 | |
*** irenab has quit IRC | 04:36 | |
*** sheel has joined #openstack-nova | 04:44 | |
*** shaohe_feng has quit IRC | 04:45 | |
*** shaohe_feng has joined #openstack-nova | 04:45 | |
openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Change hashes for latest oslo.versionedobjects https://review.openstack.org/314413 | 04:52 |
*** mdrabe has quit IRC | 04:53 | |
*** rfolco has quit IRC | 04:53 | |
*** shaohe_feng has quit IRC | 04:55 | |
*** shaohe_feng has joined #openstack-nova | 04:56 | |
*** salv-orlando has joined #openstack-nova | 04:57 | |
*** jwcroppe has quit IRC | 04:57 | |
*** jwcroppe has joined #openstack-nova | 04:58 | |
*** fragatina has quit IRC | 04:59 | |
*** akshai has quit IRC | 05:00 | |
*** Sukhdev has quit IRC | 05:01 | |
*** jwcroppe has quit IRC | 05:02 | |
*** shaohe_feng has quit IRC | 05:05 | |
*** shaohe_feng has joined #openstack-nova | 05:06 | |
*** sdake has quit IRC | 05:06 | |
*** armax has quit IRC | 05:08 | |
*** ljxiash has quit IRC | 05:08 | |
oomichi_ | alex_xu: can you take a look at https://review.openstack.org/#/c/314413/ ? | 05:08 |
oomichi_ | alex_xu: the unit test is failing on the gate, and the patch fixes it. | 05:09 |
oomichi_ | as the passing the test | 05:09 |
*** ljxiash has joined #openstack-nova | 05:10 | |
*** prateek has quit IRC | 05:10 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/313743 | 05:12 |
*** ljxiash has quit IRC | 05:14 | |
openstackgerrit | Moshe Levi proposed openstack/nova: pci: Deprecate is_new from pci request https://review.openstack.org/310990 | 05:14 |
*** Daisy has joined #openstack-nova | 05:15 | |
*** shaohe_feng has quit IRC | 05:15 | |
*** shaohe_feng has joined #openstack-nova | 05:16 | |
*** fragatina has joined #openstack-nova | 05:17 | |
openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Change hashes for latest oslo.versionedobjects https://review.openstack.org/314413 | 05:18 |
oomichi_ | dansmith: johnthetubaguy: can you take a look at https://review.openstack.org/#/c/314413/ ? | 05:19 |
*** gcb has joined #openstack-nova | 05:19 | |
oomichi_ | dansmith: johnthetubaguy: I am not a specialist about the area, and happy if you see it | 05:19 |
*** lpetrut has joined #openstack-nova | 05:20 | |
*** fragatina has quit IRC | 05:22 | |
*** shaohe_feng has quit IRC | 05:26 | |
*** iceyao_ has joined #openstack-nova | 05:28 | |
*** sudipto has joined #openstack-nova | 05:28 | |
*** irenab has joined #openstack-nova | 05:28 | |
*** shaohe_feng has joined #openstack-nova | 05:29 | |
*** apoorv has joined #openstack-nova | 05:29 | |
*** iceyao has quit IRC | 05:31 | |
*** itamarl has joined #openstack-nova | 05:31 | |
*** fragatina has joined #openstack-nova | 05:33 | |
*** sudipto has quit IRC | 05:33 | |
*** fragatina has quit IRC | 05:33 | |
*** roeyc has joined #openstack-nova | 05:33 | |
*** fragatina has joined #openstack-nova | 05:34 | |
*** Vek has joined #openstack-nova | 05:34 | |
*** jichen has joined #openstack-nova | 05:34 | |
*** shaohe_feng has quit IRC | 05:36 | |
*** shaohe_feng has joined #openstack-nova | 05:37 | |
*** prateek has joined #openstack-nova | 05:37 | |
*** moshele has quit IRC | 05:39 | |
*** oomichi_ has quit IRC | 05:41 | |
*** roeyc has quit IRC | 05:42 | |
*** shaohe_feng has quit IRC | 05:46 | |
*** shaohe_feng has joined #openstack-nova | 05:47 | |
*** sudipto has joined #openstack-nova | 05:49 | |
*** roeyc has joined #openstack-nova | 05:50 | |
*** itamarl has quit IRC | 05:53 | |
*** shaohe_feng has quit IRC | 05:56 | |
*** shaohe_feng has joined #openstack-nova | 05:57 | |
*** sridharg has joined #openstack-nova | 05:59 | |
*** lpetrut has quit IRC | 05:59 | |
*** mjura has joined #openstack-nova | 05:59 | |
*** ljxiash has joined #openstack-nova | 06:02 | |
*** esracelik has joined #openstack-nova | 06:02 | |
*** sudipto has quit IRC | 06:03 | |
*** dedery has joined #openstack-nova | 06:04 | |
*** thorst_ has quit IRC | 06:06 | |
*** itamarl has joined #openstack-nova | 06:06 | |
openstackgerrit | Srushti Gadadare proposed openstack/nova: Fix invalid uuid warnings https://review.openstack.org/302658 | 06:06 |
*** thorst_ has joined #openstack-nova | 06:06 | |
*** ljxiash has quit IRC | 06:06 | |
*** shaohe_feng has quit IRC | 06:07 | |
*** ljxiash has joined #openstack-nova | 06:07 | |
*** shaohe_feng has joined #openstack-nova | 06:10 | |
*** amotoki has quit IRC | 06:10 | |
*** thorst_ has quit IRC | 06:11 | |
*** Oku_OS-away is now known as Oku_OS | 06:13 | |
*** paul-carlton1 has joined #openstack-nova | 06:15 | |
*** gongysh has joined #openstack-nova | 06:16 | |
*** shaohe_feng has quit IRC | 06:17 | |
*** shaohe_feng has joined #openstack-nova | 06:18 | |
*** sudipto has joined #openstack-nova | 06:21 | |
*** Daisy has quit IRC | 06:24 | |
*** Daisy has joined #openstack-nova | 06:24 | |
*** gcb has quit IRC | 06:25 | |
*** andreas_s has joined #openstack-nova | 06:26 | |
*** itamarl has quit IRC | 06:26 | |
*** rcernin has joined #openstack-nova | 06:26 | |
*** shaohe_feng has quit IRC | 06:27 | |
*** shaohe_feng has joined #openstack-nova | 06:28 | |
*** Daisy has quit IRC | 06:28 | |
*** gongysh has quit IRC | 06:31 | |
*** vtech has quit IRC | 06:34 | |
*** vilobhmm11 has joined #openstack-nova | 06:34 | |
*** shaohe_feng has quit IRC | 06:37 | |
*** roeyc has quit IRC | 06:38 | |
*** jdurgin1 has joined #openstack-nova | 06:38 | |
*** shaohe_feng has joined #openstack-nova | 06:40 | |
*** moshele has joined #openstack-nova | 06:41 | |
*** moshele has quit IRC | 06:42 | |
*** moshele has joined #openstack-nova | 06:42 | |
*** itamarl has joined #openstack-nova | 06:45 | |
*** gongysh has joined #openstack-nova | 06:45 | |
*** shaohe_feng has quit IRC | 06:48 | |
*** shaohe_feng has joined #openstack-nova | 06:48 | |
*** migi has joined #openstack-nova | 06:51 | |
*** migi has joined #openstack-nova | 06:51 | |
*** Daisy has joined #openstack-nova | 06:53 | |
*** sacharya has quit IRC | 06:53 | |
*** shaohe_feng has quit IRC | 06:58 | |
*** lpetrut has joined #openstack-nova | 06:59 | |
*** shaohe_feng has joined #openstack-nova | 06:59 | |
*** amotoki has joined #openstack-nova | 06:59 | |
openstackgerrit | zhufl proposed openstack/nova: Correct some misspell words in nova https://review.openstack.org/314406 | 07:00 |
*** jed56 has joined #openstack-nova | 07:00 | |
*** markus_z has joined #openstack-nova | 07:00 | |
*** mariusv has quit IRC | 07:00 | |
*** mariusv has joined #openstack-nova | 07:01 | |
*** sudipto has quit IRC | 07:03 | |
*** tesseract has joined #openstack-nova | 07:03 | |
*** mpavone has joined #openstack-nova | 07:03 | |
*** laszloh has joined #openstack-nova | 07:03 | |
*** salv-orlando has quit IRC | 07:05 | |
*** gjayavelu has quit IRC | 07:06 | |
*** iceyao_ has quit IRC | 07:07 | |
*** iceyao has joined #openstack-nova | 07:07 | |
*** jamielennox is now known as jamielennox|away | 07:08 | |
*** zenoway has joined #openstack-nova | 07:08 | |
*** shaohe_feng has quit IRC | 07:08 | |
*** fishcried has quit IRC | 07:09 | |
*** shaohe_feng has joined #openstack-nova | 07:09 | |
*** vtech has joined #openstack-nova | 07:10 | |
*** omnipresent has joined #openstack-nova | 07:15 | |
*** zenoway has quit IRC | 07:16 | |
*** nkrinner_afk is now known as nkrinner | 07:16 | |
*** shaohe_feng has quit IRC | 07:18 | |
*** jdurgin1 has quit IRC | 07:19 | |
*** shaohe_feng has joined #openstack-nova | 07:19 | |
*** zenoway has joined #openstack-nova | 07:23 | |
tdurakov | hi, folks, could we finally merge this patch? | 07:24 |
tdurakov | https://review.openstack.org/#/c/287997/ | 07:24 |
*** lennyb has quit IRC | 07:25 | |
*** omnipresent has quit IRC | 07:25 | |
*** lpetrut has quit IRC | 07:27 | |
*** shaohe_feng has quit IRC | 07:29 | |
*** shaohe_feng has joined #openstack-nova | 07:29 | |
*** josecastroleon has joined #openstack-nova | 07:34 | |
*** lennyb has joined #openstack-nova | 07:34 | |
*** hieulq has quit IRC | 07:37 | |
*** mnestratov has joined #openstack-nova | 07:37 | |
*** markus_z has quit IRC | 07:38 | |
*** shaohe_feng has quit IRC | 07:39 | |
*** jlanoux has joined #openstack-nova | 07:39 | |
bauzas | good morning Nova | 07:39 |
*** gongysh has quit IRC | 07:39 | |
*** shaohe_feng has joined #openstack-nova | 07:39 | |
*** matrohon has joined #openstack-nova | 07:45 | |
*** cheneydc1 has joined #openstack-nova | 07:49 | |
*** jamielennox|away is now known as jamielennox | 07:49 | |
*** shaohe_feng has quit IRC | 07:49 | |
*** vilobhmm11 has quit IRC | 07:49 | |
*** shaohe_feng has joined #openstack-nova | 07:50 | |
*** belmoreira has joined #openstack-nova | 07:51 | |
*** cheneydc has quit IRC | 07:51 | |
*** cheneydc1 is now known as cheneydc | 07:51 | |
*** liverpooler has joined #openstack-nova | 07:52 | |
*** hieulq has joined #openstack-nova | 07:52 | |
*** sacharya has joined #openstack-nova | 07:54 | |
*** PaulMurray has joined #openstack-nova | 07:55 | |
*** palexster has quit IRC | 07:57 | |
*** sacharya has quit IRC | 07:58 | |
*** shaohe_feng has quit IRC | 07:59 | |
*** zzzeek has quit IRC | 08:00 | |
*** zzzeek has joined #openstack-nova | 08:00 | |
*** shaohe_feng has joined #openstack-nova | 08:00 | |
*** jaosorior has joined #openstack-nova | 08:01 | |
*** dedery_ has joined #openstack-nova | 08:02 | |
*** dedery__ has joined #openstack-nova | 08:03 | |
*** haukebruno has joined #openstack-nova | 08:03 | |
*** palexster has joined #openstack-nova | 08:04 | |
*** voelzmo has joined #openstack-nova | 08:04 | |
*** dedery__ has quit IRC | 08:04 | |
*** dedery has quit IRC | 08:04 | |
*** dedery__ has joined #openstack-nova | 08:04 | |
*** dedery_ has quit IRC | 08:06 | |
*** danpb has joined #openstack-nova | 08:06 | |
mdbooth | Would the real paul-carlton paul-carlton1 paul-carlton2 please stand up? | 08:09 |
*** shaohe_feng has quit IRC | 08:10 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 08:10 |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 08:10 |
*** tpepper has quit IRC | 08:10 | |
andrearosa | mdbooth: he is not in the office yet, probably on a train | 08:11 |
*** pkoniszewski has quit IRC | 08:11 | |
openstackgerrit | Andrea Rosa proposed openstack/nova: Adding device_present field to bdm table https://review.openstack.org/268053 | 08:12 |
openstackgerrit | Weiwei Gu proposed openstack/nova: migration volume failed for invalid type https://review.openstack.org/314455 | 08:13 |
*** tpepper1 has joined #openstack-nova | 08:14 | |
*** shaohe_feng has joined #openstack-nova | 08:14 | |
*** pkoniszewski has joined #openstack-nova | 08:15 | |
*** mvk has quit IRC | 08:16 | |
*** yamahata has quit IRC | 08:20 | |
*** shaohe_feng has quit IRC | 08:20 | |
*** shaohe_feng has joined #openstack-nova | 08:21 | |
*** vipuls has quit IRC | 08:22 | |
*** lucas-dinner is now known as lucasagomes | 08:22 | |
*** tpepper1 has quit IRC | 08:22 | |
*** tpepper1 has joined #openstack-nova | 08:23 | |
*** iceyao has quit IRC | 08:23 | |
*** lpetrut has joined #openstack-nova | 08:24 | |
*** vipuls has joined #openstack-nova | 08:25 | |
*** paul-carlton1 has quit IRC | 08:25 | |
*** iceyao has joined #openstack-nova | 08:28 | |
*** lpetrut has quit IRC | 08:28 | |
*** lpetrut has joined #openstack-nova | 08:29 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove mox in test_block_device_mapping_v1.py https://review.openstack.org/308746 | 08:29 |
*** sfinucan has joined #openstack-nova | 08:29 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in test_pipelib.py https://review.openstack.org/308806 | 08:30 |
*** shaohe_feng has quit IRC | 08:30 | |
*** daemontool__ is now known as daemontool | 08:30 | |
*** shaohe_feng has joined #openstack-nova | 08:31 | |
johnthetubaguy | oomichi: just going to take a look at that patch now | 08:34 |
johnthetubaguy | oomichi: I guess the gate is totally broken right now then? | 08:34 |
*** jistr has joined #openstack-nova | 08:37 | |
johnthetubaguy | oomichi: oh, I see, it will break once the upper constraint is lifted | 08:39 |
*** jwcroppe has joined #openstack-nova | 08:39 | |
mnestratov | good morning | 08:39 |
*** shaohe_feng has quit IRC | 08:40 | |
*** yassine has joined #openstack-nova | 08:41 | |
mnestratov | johnthetubaguy: please remove your -2 on https://review.openstack.org/#/c/190843/ | 08:41 |
*** shaohe_feng has joined #openstack-nova | 08:41 | |
*** yassine is now known as Guest59809 | 08:41 | |
mnestratov | related bp has been reapproved https://blueprints.launchpad.net/nova/+spec/libvirt-vzstorage-volume-support | 08:42 |
johnthetubaguy | mnestratov: sure thing, the BP is approved | 08:42 |
*** amotoki has quit IRC | 08:42 | |
mnestratov | johnthetubaguy: thanks a lot | 08:42 |
*** amotoki has joined #openstack-nova | 08:43 | |
*** paul-carlton1 has joined #openstack-nova | 08:44 | |
*** jwcroppe has quit IRC | 08:44 | |
*** mvk has joined #openstack-nova | 08:45 | |
*** derekh has joined #openstack-nova | 08:47 | |
*** gcb has joined #openstack-nova | 08:50 | |
*** shaohe_feng has quit IRC | 08:51 | |
openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: use os-brick for volume related operations https://review.openstack.org/273504 | 08:52 |
*** vnogin has joined #openstack-nova | 08:55 | |
*** shaohe_feng has joined #openstack-nova | 08:55 | |
openstackgerrit | Alexis Lee proposed openstack/nova: Make 'debug' option reconfigurable in Nova https://review.openstack.org/280851 | 08:56 |
yuywz | hi folks, I have a question about mox and mock: how to replace a class with another like 'mox.stubs.Set' does using mock? | 08:58 |
yuywz | for example we can replace 'oslo_messaging.Notifier' with 'fake_notifer.FakeNotifier' with self.mox.stubs.Set(oslo_messaging, 'Notifier', fake_notifier.FakeNotifier) | 08:59 |
yuywz | but how to do this using mock? | 09:00 |
*** sacharya has joined #openstack-nova | 09:00 | |
*** sdake has joined #openstack-nova | 09:01 | |
*** shaohe_feng has quit IRC | 09:01 | |
*** shaohe_feng has joined #openstack-nova | 09:01 | |
*** ljxiash has quit IRC | 09:03 | |
*** ljxiash has joined #openstack-nova | 09:03 | |
johnthetubaguy | yuywz: take a look at: https://github.com/openstack/nova/blob/master/nova/test.py#L269 | 09:03 |
*** Guest59809 is now known as yassou | 09:04 | |
*** sacharya has quit IRC | 09:04 | |
openstackgerrit | Srushti Gadadare proposed openstack/nova: Fix invalid uuid warnings https://review.openstack.org/302658 | 09:04 |
*** roeyc has joined #openstack-nova | 09:06 | |
*** itamarl_ has joined #openstack-nova | 09:06 | |
yuywz | johnthetubaguy: thx, but I've tried that and seems stub_out only works with single method like self.stub_out('nova.compute.api.API.rebuild', fake_rebuild) | 09:07 |
*** ljxiash has quit IRC | 09:07 | |
*** itamarl has quit IRC | 09:08 | |
*** itamarl_ is now known as itamarl | 09:09 | |
*** mnestratov has quit IRC | 09:09 | |
*** roeyc has quit IRC | 09:10 | |
yuywz | johnthetubaguy: I tried this self.useFixture(fixtures.MonkeyPatch('oslo_messaging.Notifier', fake_notifier.FakeNotifier)), but it did not work as expected | 09:11 |
*** shaohe_feng has quit IRC | 09:11 | |
yuywz | and self.mox.stubs.Set(oslo_messaging, 'Notifier', fake_notifier.FakeNotifier) works | 09:11 |
*** tpepper1 has quit IRC | 09:13 | |
*** vipul- has joined #openstack-nova | 09:13 | |
*** vipuls has quit IRC | 09:13 | |
*** pkoniszewski has quit IRC | 09:14 | |
johnthetubaguy | yuywz: hmm, thats odd, I was just taking a glance at the code we use underneath: https://github.com/testing-cabal/fixtures/blob/master/fixtures/_fixtures/monkeypatch.py | 09:14 |
*** jwcroppe has joined #openstack-nova | 09:15 | |
*** tpepper has joined #openstack-nova | 09:17 | |
*** shaohe_feng has joined #openstack-nova | 09:17 | |
johnthetubaguy | yuywz: where are you using that code, depends what self is at that point, I am just looking at here: https://github.com/openstack/nova/blob/master/nova/tests/unit/fake_notifier.py#L78 | 09:18 |
*** pkoniszewski has joined #openstack-nova | 09:18 | |
*** jwcroppe has quit IRC | 09:20 | |
johnthetubaguy | if you have a patch in gerrit, its easier to see the context and help, most of the time I would mock out only the method I care about, I guess, not sure if that helps here | 09:20 |
*** shaohe_feng has quit IRC | 09:21 | |
yuywz | johnthetubaguy: thanks for your help, sorry I don't have a patch at hand, will upload one later | 09:22 |
johnthetubaguy | yuywz: no worries | 09:22 |
*** shaohe_feng has joined #openstack-nova | 09:22 | |
*** tpepper has quit IRC | 09:23 | |
*** tpepper has joined #openstack-nova | 09:24 | |
*** tbachman has quit IRC | 09:26 | |
*** salv-orlando has joined #openstack-nova | 09:27 | |
*** Daisy has quit IRC | 09:27 | |
*** Daisy has joined #openstack-nova | 09:28 | |
*** daemontool has quit IRC | 09:28 | |
*** Drankis has joined #openstack-nova | 09:29 | |
openstackgerrit | Laszlo Hegedus proposed openstack/nova: Add versioned exception notifications https://review.openstack.org/310700 | 09:30 |
*** tbachman has joined #openstack-nova | 09:30 | |
*** shaohe_feng has quit IRC | 09:32 | |
*** salv-orlando has quit IRC | 09:32 | |
*** jlanoux has quit IRC | 09:32 | |
*** palexster has quit IRC | 09:32 | |
*** shaohe_feng has joined #openstack-nova | 09:33 | |
*** Daisy_ has joined #openstack-nova | 09:33 | |
*** jlanoux has joined #openstack-nova | 09:36 | |
*** Daisy has quit IRC | 09:36 | |
*** salv-orlando has joined #openstack-nova | 09:38 | |
*** salv-orlando has quit IRC | 09:38 | |
*** salv-orlando has joined #openstack-nova | 09:38 | |
*** markus_z has joined #openstack-nova | 09:41 | |
*** shaohe_feng has quit IRC | 09:42 | |
*** shaohe_feng has joined #openstack-nova | 09:45 | |
markus_z | nova-bugs team meeting in 10 min in #openstack-meeting-4. New contributors are welcome as well. https://wiki.openstack.org/wiki/Meetings/Nova/BugsTeam | 09:49 |
*** Daisy_ has quit IRC | 09:49 | |
*** Daisy has joined #openstack-nova | 09:50 | |
*** iceyao_ has joined #openstack-nova | 09:51 | |
*** iceyao has quit IRC | 09:51 | |
*** shaohe_feng has quit IRC | 09:52 | |
*** shaohe_feng has joined #openstack-nova | 09:53 | |
*** tbachman has quit IRC | 09:55 | |
openstackgerrit | Laszlo Hegedus proposed openstack/nova: Add versioned exception notifications https://review.openstack.org/310700 | 09:59 |
*** cheneydc has quit IRC | 09:59 | |
*** roeyc has joined #openstack-nova | 10:01 | |
markus_z | nova-bugs meeting started in #openstack-meeting-4 | 10:01 |
*** shaohe_feng has quit IRC | 10:02 | |
*** mvk has quit IRC | 10:03 | |
*** shaohe_feng has joined #openstack-nova | 10:03 | |
*** mvk has joined #openstack-nova | 10:04 | |
*** houming_ has joined #openstack-nova | 10:05 | |
*** roeyc has quit IRC | 10:05 | |
*** houming has quit IRC | 10:06 | |
*** houming_ is now known as houming | 10:06 | |
*** gcb has quit IRC | 10:06 | |
*** daemontool has joined #openstack-nova | 10:08 | |
*** cdent has joined #openstack-nova | 10:09 | |
*** jaosorior has quit IRC | 10:09 | |
*** iceyao_ has quit IRC | 10:09 | |
*** jaosorior has joined #openstack-nova | 10:09 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 10:10 |
*** jwcroppe has joined #openstack-nova | 10:10 | |
*** sdague has joined #openstack-nova | 10:12 | |
*** shaohe_feng has quit IRC | 10:13 | |
openstackgerrit | Sean Dague proposed openstack/nova: Complete method verification of os-security-group-rules https://review.openstack.org/314264 | 10:13 |
*** shaohe_feng has joined #openstack-nova | 10:13 | |
*** jwcroppe has quit IRC | 10:15 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: objects: Removes base code that already exists in o.vo https://review.openstack.org/307041 | 10:17 |
*** eglynn has joined #openstack-nova | 10:17 | |
*** tbachman has joined #openstack-nova | 10:17 | |
johnthetubaguy | sdague: I was thinking about that change to remove the tenant_id from the URL, so we just talk about things in addition to the service catalog | 10:18 |
johnthetubaguy | sdague: I am assuming thats waiting till after this sprint? | 10:18 |
*** kolesovdv has joined #openstack-nova | 10:18 | |
sdague | johnthetubaguy: yeh | 10:18 |
sdague | otherwise it's just a giant merge conflict for everyone | 10:18 |
johnthetubaguy | yeah, ack | 10:19 |
johnthetubaguy | sdague: this patch is blocking the merge of an approved docs patch: https://review.openstack.org/#/c/313959/ | 10:19 |
*** gcb has joined #openstack-nova | 10:19 | |
johnthetubaguy | sdague: the expected 404 on lists where the marker is currently outputting 400 instead, what do you think we should do in the docs for the moment? I think we have some folks adding 404, and some folks excluding it | 10:22 |
*** shaohe_feng has quit IRC | 10:23 | |
*** shaohe_feng has joined #openstack-nova | 10:24 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: Compute: Adds driver disk_gb instance overhead estimation https://review.openstack.org/234283 | 10:25 |
openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Fixes disk overhead claim issue https://review.openstack.org/246307 | 10:25 |
*** claudiub has joined #openstack-nova | 10:25 | |
*** smatzek has joined #openstack-nova | 10:25 | |
*** liverpooler has quit IRC | 10:26 | |
*** liverpooler has joined #openstack-nova | 10:26 | |
*** jlanoux_ has joined #openstack-nova | 10:29 | |
*** jlanoux has quit IRC | 10:30 | |
*** EinstCra_ has quit IRC | 10:30 | |
*** smatzek has quit IRC | 10:30 | |
*** tbachman_ has joined #openstack-nova | 10:30 | |
sdague | johnthetubaguy: yeh, it's a good question. I hadn't really been thinking too hard on those error codes when we got started. I kind of felt like we need a separate error codes section to talk them through in detail. | 10:31 |
*** EinstCrazy has joined #openstack-nova | 10:31 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: Fixes python 3 unit tests https://review.openstack.org/278810 | 10:31 |
openstackgerrit | Claudiu Belu proposed openstack/nova: Fixes python 3 urllib quote / unquote usage https://review.openstack.org/278809 | 10:31 |
johnthetubaguy | sdague: I noticed your nice comments on the how errors happen in one section, I could see that being a thing | 10:31 |
openstackgerrit | Claudiu Belu proposed openstack/nova: compute: fixes python 3 related unit tests https://review.openstack.org/300262 | 10:32 |
sdague | yeh, servers was complicated enough I didn't want to loose what I'd discovered | 10:32 |
johnthetubaguy | +1 | 10:32 |
*** smatzek has joined #openstack-nova | 10:33 | |
*** shaohe_feng has quit IRC | 10:33 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: libvirt: fixes python 3 related unit tests https://review.openstack.org/278834 | 10:33 |
openstackgerrit | Claudiu Belu proposed openstack/nova: libvirt: fixes python 3 related unit tests https://review.openstack.org/278834 | 10:33 |
openstackgerrit | Claudiu Belu proposed openstack/nova: numa: Fixes NUMA topology related unit tests https://review.openstack.org/278835 | 10:34 |
*** shaohe_feng has joined #openstack-nova | 10:34 | |
*** tbachman has quit IRC | 10:34 | |
*** tbachman_ is now known as tbachman | 10:34 | |
*** EinstCrazy has quit IRC | 10:35 | |
*** hoangcx has quit IRC | 10:37 | |
*** sdake has quit IRC | 10:42 | |
*** Daisy_ has joined #openstack-nova | 10:42 | |
*** shaohe_feng has quit IRC | 10:43 | |
johnthetubaguy | sdague: I wasn't sure about this one: https://review.openstack.org/#/c/314287/1/api-ref/source/os-fping.inc@31 am I miss-reading something? | 10:44 |
*** shaohe_feng has joined #openstack-nova | 10:44 | |
sdague | ah, yeh, 404 probably shouldn't be in that list. | 10:44 |
*** Daisy has quit IRC | 10:46 | |
*** rodrigods has quit IRC | 10:49 | |
*** rodrigods has joined #openstack-nova | 10:49 | |
openstackgerrit | John Garbutt proposed openstack/nova: Complete method verification of os-fping https://review.openstack.org/314287 | 10:49 |
sdague | https://review.openstack.org/#/c/313532 would be really good to get in, mriedem thought he found a thing, but the code is just confusing | 10:50 |
johnthetubaguy | sdague: talking a look | 10:50 |
*** Daisy_ has quit IRC | 10:50 | |
*** Daisy has joined #openstack-nova | 10:51 | |
*** abalutoiu has joined #openstack-nova | 10:51 | |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 10:52 |
*** shaohe_feng has quit IRC | 10:54 | |
*** shaohe_feng has joined #openstack-nova | 10:54 | |
*** Daisy has quit IRC | 10:56 | |
*** laszloh has quit IRC | 10:58 | |
*** gcb has quit IRC | 11:00 | |
*** sacharya has joined #openstack-nova | 11:01 | |
*** shaohe_feng has quit IRC | 11:04 | |
*** shaohe_feng has joined #openstack-nova | 11:05 | |
*** jwcroppe has joined #openstack-nova | 11:05 | |
*** sacharya has quit IRC | 11:05 | |
*** sdake has joined #openstack-nova | 11:07 | |
openstackgerrit | John Garbutt proposed openstack/nova: api-ref: complete verification for extensions resource https://review.openstack.org/314101 | 11:09 |
*** mnestratov has joined #openstack-nova | 11:10 | |
*** jwcroppe has quit IRC | 11:11 | |
*** gcb has joined #openstack-nova | 11:12 | |
*** shaohe_feng has quit IRC | 11:14 | |
*** shaohe_feng has joined #openstack-nova | 11:15 | |
openstackgerrit | Sylvain Bauza proposed openstack/nova: Verify requested_destination in the scheduler https://review.openstack.org/314509 | 11:17 |
*** dims has quit IRC | 11:20 | |
*** jaosorior has quit IRC | 11:20 | |
*** jaosorior has joined #openstack-nova | 11:20 | |
*** swamireddy has quit IRC | 11:23 | |
*** palexster has joined #openstack-nova | 11:24 | |
*** shaohe_feng has quit IRC | 11:24 | |
*** dims has joined #openstack-nova | 11:25 | |
*** kengo_sakai has quit IRC | 11:25 | |
*** shaohe_feng has joined #openstack-nova | 11:25 | |
*** rtheis has joined #openstack-nova | 11:29 | |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 11:30 |
*** gcb has quit IRC | 11:33 | |
*** shaohe_feng has quit IRC | 11:35 | |
*** shaohe_feng has joined #openstack-nova | 11:35 | |
*** apoorv has quit IRC | 11:36 | |
*** jaosorior has quit IRC | 11:42 | |
*** jaypipes has quit IRC | 11:43 | |
*** jaosorior has joined #openstack-nova | 11:43 | |
*** allen_gao has quit IRC | 11:44 | |
*** shaohe_feng has quit IRC | 11:45 | |
*** jaypipes has joined #openstack-nova | 11:45 | |
*** shaohe_feng has joined #openstack-nova | 11:45 | |
sdague | johnthetubaguy: https://review.openstack.org/#/c/314257/ is an easy one | 11:45 |
*** gcb has joined #openstack-nova | 11:47 | |
*** baoli has joined #openstack-nova | 11:50 | |
*** lucasagomes is now known as lucas-hungry | 11:52 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Pass BDM in swap_volume method of LibvirtDriver https://review.openstack.org/257135 | 11:52 |
*** baoli_ has joined #openstack-nova | 11:53 | |
openstackgerrit | Sean Dague proposed openstack/nova: Denormalize extensions for clarity https://review.openstack.org/314521 | 11:53 |
*** shaohe_feng has quit IRC | 11:55 | |
*** baoli has quit IRC | 11:56 | |
*** shaohe_feng has joined #openstack-nova | 11:56 | |
*** sdake has quit IRC | 11:56 | |
*** gcb has quit IRC | 11:58 | |
sdague | johnthetubaguy: https://review.openstack.org/314521 hopefully would have made the confusion mriedem ran into less confusing | 11:59 |
*** roeyc has joined #openstack-nova | 12:02 | |
*** sdake has joined #openstack-nova | 12:02 | |
*** shaohe_feng has quit IRC | 12:05 | |
*** shaohe_feng has joined #openstack-nova | 12:06 | |
*** houming has quit IRC | 12:06 | |
*** roeyc has quit IRC | 12:07 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Set 'serial' to new volume ID in swap volumes https://review.openstack.org/257135 | 12:07 |
*** PaulMurray has quit IRC | 12:08 | |
*** ozialien10 has quit IRC | 12:08 | |
*** sdake_ has joined #openstack-nova | 12:08 | |
*** ozialien10 has joined #openstack-nova | 12:09 | |
*** sdake has quit IRC | 12:11 | |
*** gcb has joined #openstack-nova | 12:11 | |
*** haukebruno has quit IRC | 12:12 | |
tdurakov | where this folder is used in nova: https://github.com/openstack/nova/blob/master/api-ref/ ? | 12:12 |
johnthetubaguy | tdurakov: in the doc build system, see tox.ini | 12:13 |
tdurakov | johnthetubaguy: ok, and where is it available then? | 12:14 |
tdurakov | here http://developer.openstack.org/api-ref-compute-v2.1.html ? | 12:14 |
johnthetubaguy | tdurakov: was just getting that url, its here: http://developer.openstack.org/api-guide/compute/ | 12:14 |
johnthetubaguy | tdurakov: the one you linked is in the api-site, be we are currently moving that | 12:14 |
johnthetubaguy | tdurakov: oh, wait, did I get mixed up there | 12:15 |
johnthetubaguy | yeah, I did | 12:15 |
johnthetubaguy | tdurakov: it goes here: http://developer.openstack.org/api-ref/compute/ | 12:15 |
*** rfolco has joined #openstack-nova | 12:15 | |
*** shaohe_feng has quit IRC | 12:16 | |
johnthetubaguy | tdurakov: its soon replacing the api-site stuff that is here: http://developer.openstack.org/api-ref-compute-v2.1.html | 12:16 |
*** pradk has joined #openstack-nova | 12:16 | |
*** shaohe_feng has joined #openstack-nova | 12:16 | |
sdague | tdurakov: we're doing a sprint this week to try to get through verifying the content before replacing the links | 12:17 |
tdurakov | johnthetubaguy: that makes sense, thank you for details, the original problem: couldn't find this line here: http://developer.openstack.org/api-ref-compute-v2.1.html | 12:17 |
tdurakov | sdague: could be interesting^ | 12:17 |
*** sree has joined #openstack-nova | 12:17 | |
tdurakov | this line: https://github.com/openstack/nova/blob/master/api-ref/source/servers-admin-action.inc#L130 | 12:17 |
sdague | tdurakov: right, that content has been split as of 3 weeks ago | 12:18 |
johnthetubaguy | tdurakov: for completeness, this is the old stuff we are soon to delete: https://github.com/openstack/api-site/tree/master/api-ref/src/wadls/compute-api/src/v2.1/wadl | 12:20 |
tdurakov | sdague: acked, one more question, is it about latest api version only? | 12:20 |
sdague | tdurakov: no, there is some support for microversions, and more will be added | 12:21 |
tdurakov | sdague: how is it possible to check api for prev microversion for example? | 12:22 |
*** svenkat has joined #openstack-nova | 12:22 | |
tdurakov | johnthetubaguy: i believe you remember context of this change: https://review.openstack.org/#/c/287997/ could you please review it again? | 12:23 |
sdague | tdurakov: http://docs-draft.openstack.org/32/313532/9/check/gate-nova-api-ref/5ed64d6//api-ref/build/html/#create-server | 12:24 |
*** aysyd has joined #openstack-nova | 12:24 | |
johnthetubaguy | tdurakov: deep in other topics right now, but will try get to that | 12:25 |
tdurakov | johnthetubaguy: that would be great, thanks in advance! | 12:26 |
*** shaohe_feng has quit IRC | 12:26 | |
*** shaohe_feng has joined #openstack-nova | 12:26 | |
tdurakov | sdague: acked | 12:26 |
*** sree has quit IRC | 12:28 | |
*** ivase has joined #openstack-nova | 12:33 | |
ildikov | johnthetubaguy: hi | 12:33 |
johnthetubaguy | ildikov: hi | 12:33 |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 12:35 |
ildikov | johnthetubaguy: I just wanted to check whether the Thursday 1700UTC slot work for you for the CInder-Nova follow up meetings | 12:35 |
ildikov | johnthetubaguy: that is planned to be IRC meeting only | 12:35 |
*** shaohe_feng has quit IRC | 12:36 | |
ildikov | johnthetubaguy: I uploaded an updated version for the multiattach spec, if theres any issue with it we can discuss as well beyond the follow ups on each item | 12:36 |
johnthetubaguy | ildikov: its tricky for me on the thursday, I wasn't really sure I was needed to keep things moving along though? | 12:37 |
*** shaohe_feng has joined #openstack-nova | 12:37 | |
*** liverpooler has quit IRC | 12:38 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: libvirt: fixes python 3 related unit tests https://review.openstack.org/278834 | 12:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: numa: Fixes NUMA topology related unit tests https://review.openstack.org/278835 | 12:39 |
ildikov | johnthetubaguy: I will not force you :) | 12:40 |
*** burgerk has joined #openstack-nova | 12:40 | |
ildikov | johnthetubaguy: but if you would like to then I can try to find another slot | 12:40 |
ildikov | johnthetubaguy: also if you have concerns regarding the spec and cannot attend the meeting, can you add comments on the review please? | 12:41 |
openstackgerrit | Moshe Levi proposed openstack/nova: pci: Move PCI devices and PCI requests into migration context https://review.openstack.org/307124 | 12:42 |
johnthetubaguy | ildikov: will do, do you have the spec link handy? | 12:43 |
ildikov | johnthetubaguy: https://review.openstack.org/#/c/304681/ | 12:44 |
ildikov | johnthetubaguy: thanks | 12:44 |
johnthetubaguy | ildikov: ah, good good, I will try get to that one today | 12:44 |
ildikov | johnthetubaguy: tnx much | 12:44 |
*** thorst_ has joined #openstack-nova | 12:45 | |
*** dedery__ has quit IRC | 12:45 | |
*** shaohe_feng has quit IRC | 12:46 | |
*** shaohe_feng has joined #openstack-nova | 12:47 | |
openstackgerrit | John Garbutt proposed openstack/nova: Get support matrix notes from the ini file https://review.openstack.org/264763 | 12:48 |
*** thorst_ has quit IRC | 12:49 | |
*** salv-orlando has quit IRC | 12:49 | |
*** palexster has quit IRC | 12:49 | |
*** thorst_ has joined #openstack-nova | 12:49 | |
*** palexster has joined #openstack-nova | 12:50 | |
*** lucas-hungry is now known as lucasagomes | 12:51 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Allow user to define the shutdown method through API when stop instances https://review.openstack.org/293790 | 12:51 |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 12:51 |
*** Daisy has joined #openstack-nova | 12:51 | |
*** sree has joined #openstack-nova | 12:53 | |
johnthetubaguy | markus_z: I am wondering if we should really have an official deprecation cycle on this one, we should be OK to drop that image service code though: https://review.openstack.org/#/c/314146 | 12:53 |
*** jwcroppe has joined #openstack-nova | 12:53 | |
markus_z | johnthetubaguy: Hm? I'm already removing them completely without deprecation. | 12:54 |
*** edmondsw has joined #openstack-nova | 12:55 | |
markus_z | Or are you concerned about the release note? | 12:56 |
*** shaohe_feng has quit IRC | 12:57 | |
johnthetubaguy | markus_z: I was worried about the removing them, because we were talking about making the service fail to start if you have an "invalid" config | 12:57 |
*** shaohe_feng has joined #openstack-nova | 12:57 | |
*** EinstCrazy has joined #openstack-nova | 12:57 | |
*** Daisy has quit IRC | 12:57 | |
*** jlanoux has joined #openstack-nova | 12:58 | |
*** jwcroppe has quit IRC | 12:58 | |
markus_z | johnthetubaguy: Ah, ok, got it. I've read your first sentence wrongly. | 12:59 |
*** psargent has joined #openstack-nova | 12:59 | |
markus_z | johnthetubaguy: We don't lose anything if we chose to deprecate it first, I'm fine with that. Can you leave me a -1 with a short note on that? | 13:00 |
*** gjayavelu has joined #openstack-nova | 13:00 | |
*** jlanoux_ has quit IRC | 13:00 | |
*** amotoki has quit IRC | 13:01 | |
*** sacharya has joined #openstack-nova | 13:02 | |
johnthetubaguy | markus_z: totally, let a note | 13:02 |
johnthetubaguy | left^ | 13:02 |
markus_z | thanks, I'm going to update this today | 13:02 |
markus_z | jaypipes: Do you have a minute? | 13:03 |
*** dave-mccowan has joined #openstack-nova | 13:03 | |
jaypipes | markus_z: yup! | 13:03 |
*** salv-orlando has joined #openstack-nova | 13:03 | |
markus_z | jaypipes: I'm looking at https://github.com/openstack/nova/blob/master/nova/image/download/file.py to move the options and it looks a little strange to me. | 13:04 |
markus_z | jaypipes: IIUC one can introduce new nova.conf sections by another config option CONF.image_file_url.filesystems | 13:05 |
*** rlrossit has joined #openstack-nova | 13:05 | |
markus_z | jaypipes: And I'm wondering if this is really something which gets used in production | 13:05 |
*** andymaier_ has joined #openstack-nova | 13:05 | |
markus_z | jaypipes: Feels like "bring your own interface" | 13:06 |
markus_z | and makes deprecation harder | 13:06 |
*** nkrinner is now known as nkrinner_afk | 13:06 | |
markus_z | long story short, is this file download from glance really a thing? | 13:06 |
*** sacharya has quit IRC | 13:06 | |
jaypipes | markus_z: yes, it's awful. | 13:06 |
markus_z | The author of that is not active anymore, otherwise I would ping him | 13:07 |
jaypipes | markus_z: RAX is the only one that was using it, AFAIK. perhaps johnthetubaguy could comment on that? | 13:07 |
*** shaohe_feng has quit IRC | 13:07 | |
*** cdent has quit IRC | 13:07 | |
*** shaohe_feng has joined #openstack-nova | 13:07 | |
jaypipes | markus_z: my preference would be for that code to DIAF, frankly. | 13:08 |
markus_z | jaypipes: yeah, my preference too :) | 13:08 |
*** akshai has joined #openstack-nova | 13:08 | |
johnthetubaguy | jaypipes: I don't think we have ever used that, we use glance | 13:09 |
markus_z | jaypipes: Funnily, those options never got mentioned in the nova.conf.sample nor in the config reference because of the obscure way the are offered. | 13:09 |
jaypipes | johnthetubaguy: that *is* Glance :) | 13:10 |
*** mdrabe has joined #openstack-nova | 13:10 | |
jaypipes | johnthetubaguy: it's just an abomination of code... | 13:10 |
markus_z | It's glance through the backdoor IIUC | 13:10 |
jaypipes | markus_z: correct. it relies on a Glance v2 API for multiple image location properties and a hard-coded string prefix in the URIs for those image locations. | 13:11 |
johnthetubaguy | jaypipes: hmm, OK, so I should look closer, I thought that was the file backend stuff, we use the swift backend | 13:11 |
*** edmondsw has quit IRC | 13:11 | |
voelzmo | Hey @sdague, any advice on how we could proceed with https://bugs.launchpad.net/nova/+bug/1483132 ? | 13:11 |
openstack | Launchpad bug 1483132 in OpenStack Compute (nova) "ssh-keygen-to-Paramiko change breaks third-party tools" [Undecided,Won't fix] | 13:11 |
*** ctrath has joined #openstack-nova | 13:11 | |
johnthetubaguy | jaypipes: is this the thing ceph subclasses? | 13:11 |
jaypipes | markus_z: all of that functionality should be in glance_store now, but we (Nova) have never integrated with glance_store appropriately. | 13:11 |
jaypipes | johnthetubaguy: no. it's the "image download plugins" that John Bresnahan and Eddie Sheffield wrote. | 13:12 |
voelzmo | paramiko has released 2.x with a fix moving back to DER encoding of keys, I guess we could update paramiko in nova to 2.x and make golang-based tooling happy again | 13:12 |
*** swamireddy has joined #openstack-nova | 13:12 | |
*** swamireddy has quit IRC | 13:13 | |
*** sree has quit IRC | 13:13 | |
sdague | voelzmo: yeh, I think we should bump to paramiko > 2 | 13:14 |
sdague | does that pass for projects? | 13:14 |
sdague | voelzmo: I reopened the bug, thanks for getting it sorted upstream | 13:14 |
johnthetubaguy | jaypipes: OK, looks funky, anyways, we generally use the XenServer plugins to do all the downloads and uploads to glance | 13:14 |
jaypipes | markus_z: in any case, we can't get rid of it without a deprecation cycle anyway. in the meantime, some code can be pushed that pulls those conf options out of that file and into the common nova/conf/ directory. | 13:14 |
voelzmo | @sdague: I've no idea on how to do this or how to test things, I'm just one of those guys suffering from that it is broken. Happy to submit a PR if I knew where to do that, though | 13:15 |
jaypipes | johnthetubaguy: yeah, that's why I always questioned whether that code was ever run anywhere.. | 13:15 |
johnthetubaguy | jaypipes: +1 on never run by us, AFAIK | 13:15 |
markus_z | jaypipes: yep, was thinking the same, wasn't entirely sure how much effort to spent | 13:15 |
*** eharney has joined #openstack-nova | 13:15 | |
markus_z | jaypipes: Thanks! I'll make a patch for a move and another for the deprecation | 13:16 |
johnthetubaguy | markus_z: that sounds good, deprecate it | 13:16 |
jaypipes | coolio. | 13:16 |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 13:16 |
*** shaohe_feng has quit IRC | 13:17 | |
*** kengo_sakai has joined #openstack-nova | 13:17 | |
sdague | dims: you about? | 13:17 |
*** jamesden_ has joined #openstack-nova | 13:17 | |
sdague | dims: why did you take paramiko out of https://review.openstack.org/#/c/314423/ ? | 13:17 |
*** kengo_sakai has quit IRC | 13:17 | |
*** yamahata has joined #openstack-nova | 13:18 | |
*** shaohe_feng has joined #openstack-nova | 13:18 | |
*** kengo_sakai has joined #openstack-nova | 13:18 | |
sdague | voelzmo: ok, so in looking at the test results of https://review.openstack.org/#/c/314423/ | 13:19 |
openstackgerrit | Paul Carlton proposed openstack/nova-specs: Use libvirt storage pools https://review.openstack.org/310505 | 13:19 |
sdague | there will need to be a patch in nova that supports either 1.16 or 2.0 so that nova can roll forward gracefully here | 13:20 |
sdague | then we can bring up the paramiko requirements | 13:20 |
*** thorst_ has quit IRC | 13:20 | |
*** daemontool_ has joined #openstack-nova | 13:21 | |
*** jichen has quit IRC | 13:21 | |
voelzmo | @sdague: I see 'paramiko' mentioned in the commit msg of the above change, but no paramiko related change seems to be done in there? What am I missing? | 13:22 |
voelzmo | Ah, first patchset tried to update paramiko to 2.0 | 13:22 |
sdague | right, and in the n-api logs you can see it failes | 13:22 |
*** vtech has quit IRC | 13:23 | |
sdague | http://logs.openstack.org/23/314423/1/check/gate-tempest-dsvm-full/5b356ed/logs/screen-n-api.txt.gz?level=ERROR#_2016-05-10_06_49_00_548 | 13:23 |
*** mriedem has joined #openstack-nova | 13:23 | |
*** links has quit IRC | 13:23 | |
*** daemontool has quit IRC | 13:24 | |
voelzmo | something like @coreywright added to the original bug for nova? https://bugs.launchpad.net/nova/+bug/1483132/comments/10 | 13:25 |
openstack | Launchpad bug 1483132 in OpenStack Compute (nova) "ssh-keygen-to-Paramiko change breaks third-party tools" [Low,Confirmed] | 13:25 |
openstackgerrit | Laszlo Hegedus proposed openstack/nova: Replace root volume during rebuild https://review.openstack.org/305079 | 13:25 |
sdague | voelzmo: probably, though I'm not sure why it wasn't sumbitted via gerrit | 13:27 |
*** shaohe_feng has quit IRC | 13:27 | |
*** gcb has quit IRC | 13:28 | |
mriedem | sdague: on https://review.openstack.org/#/c/314521/1/nova/api/openstack/compute/disk_config.py@148 shouldn't body_deprecated default to None? | 13:28 |
mriedem | like in access_ips and the way it was before the refactor | 13:28 |
sdague | mriedem: honestly, it's thrown away | 13:28 |
sdague | it was defaulting to None mostly so that different methods that don't have it could use it | 13:28 |
sdague | if we weren't going to be deleting extensions, I'd go and fix the unit tests to pass the 3rd parameter | 13:29 |
sdague | because, in real activity, that parameter isn't optional | 13:29 |
sdague | but because it was made optional in common code, unit tests started expecting it | 13:29 |
*** shaohe_feng has joined #openstack-nova | 13:30 | |
*** tongli has joined #openstack-nova | 13:30 | |
openstackgerrit | Sean Dague proposed openstack/nova: Denormalize extensions for clarity https://review.openstack.org/314521 | 13:31 |
*** hshiina has joined #openstack-nova | 13:31 | |
sdague | mriedem: oh, I just fixed the typo comment | 13:31 |
sdague | if you want to reapprove | 13:31 |
mriedem | done | 13:32 |
mriedem | now i just have to come to terms with the bottom patch | 13:32 |
sdague | mriedem: yep | 13:32 |
*** thorst_ has joined #openstack-nova | 13:32 | |
sdague | it's added stuff in parameters.yaml that other people want to use, so getting it in moves the ball forward. | 13:32 |
sdague | and we can continue to polish things in the future | 13:33 |
dansmith | johnthetubaguy: fwiw, we consider anything that changes the hashes unexpectedly in o.vo to be a breakage, unless we can't help it | 13:34 |
openstackgerrit | Ludovic Beliveau proposed openstack/nova: Remove DictCompat from InstancePCIRequests object https://review.openstack.org/311870 | 13:34 |
dansmith | johnthetubaguy: just re: your comment about it not happening again.. I think this is the first time, and it was unintentional, a fix just hadn't merged | 13:34 |
*** takedakn has joined #openstack-nova | 13:34 | |
openstackgerrit | Laszlo Hegedus proposed openstack/nova: Replace root volume during rebuild https://review.openstack.org/305079 | 13:34 |
*** gcb has joined #openstack-nova | 13:34 | |
Kevin_Zheng | mriedem: Hi, could you please review this one again? https://review.openstack.org/#/c/295595/ | 13:35 |
mriedem | Kevin_Zheng: yeah | 13:35 |
johnthetubaguy | dansmith: no worries, I was curious, I certainly didn't remember that before | 13:35 |
openstackgerrit | Sean Dague proposed openstack/nova: api-ref: complete verification of servers-action-crash-dump.inc https://review.openstack.org/314566 | 13:36 |
mriedem | sdague: you could do the same denormalize in the personality extension | 13:36 |
*** efried has joined #openstack-nova | 13:36 | |
sdague | mriedem: yeh, I can stack up that patch if you want | 13:36 |
claudiub | mriedem: hello. since this is a feature parity feature, does it necessarely require a spec? https://blueprints.launchpad.net/nova/+spec/hyperv-vm-diagnostics | 13:36 |
mriedem | claudiub: i don't think so, however, | 13:37 |
mriedem | claudiub: we were just talked yesterday that we need a microversion in the diagnostics API to use the newer version of it which has the standardized values | 13:37 |
mriedem | s/talked/talking/ | 13:37 |
mriedem | and that would need a spec | 13:37 |
*** shaohe_feng has quit IRC | 13:38 | |
mriedem | claudiub: so adding the support to the hyper-v driver should only be with the newer version (get_instance_diagnostics) | 13:38 |
mriedem | but the rest api doesn't call that today, we need a microversion to add that | 13:38 |
*** shaohe_feng has joined #openstack-nova | 13:38 | |
claudiub | mriedem: i see. is there any work being done on the microversion? | 13:39 |
sdague | claudiub: no, honestly, the gap was only really discovered in doing the api-ref sprint | 13:40 |
*** thorst__ has joined #openstack-nova | 13:40 | |
*** burt has joined #openstack-nova | 13:40 | |
sdague | the infrastructure was all built for nova api v3, but rolled back when we went down the 2.1 path, and never got reproposed | 13:40 |
*** thorst__ has quit IRC | 13:40 | |
*** edmondsw has joined #openstack-nova | 13:41 | |
mriedem | sdague: so i'm trying to sort out how these server_update extensions work, and it looks like in the servers api it loads the extensions from: | 13:41 |
mriedem | EXTENSION_UPDATE_NAMESPACE = 'nova.api.v21.extensions.server.update' | 13:41 |
kashyap | Cells job seems to be failing, is it just me? | 13:41 |
mriedem | kashyap: it's keystone | 13:41 |
* kashyap reads quickly greps the scrollback for any discussion on it | 13:41 | |
sdague | mriedem: one sec | 13:41 |
claudiub | jaypipes: hello. got a minute? | 13:41 |
mriedem | sdague: nvm i see | 13:41 |
mriedem | nova.api.v21.extensions.server.update = | 13:41 |
mriedem | access_ips = nova.api.openstack.compute.access_ips:AccessIPs | 13:41 |
mriedem | disk_config = nova.api.openstack.compute.disk_config:DiskConfig | 13:41 |
mriedem | i was expecting server_update, not server.update | 13:41 |
*** thorst_ has quit IRC | 13:41 | |
mriedem | so those extensions register, servers.py loads them on init, and calls them when server_update is called | 13:42 |
kashyap | mriedem: Oh, so it's a known issue? | 13:42 |
mriedem | passing in the request | 13:42 |
sdague | mriedem: https://github.com/openstack/nova/blob/20e1d6c46446dba0e63b820afaa0660a24427afe/nova/api/openstack/compute/servers.py#L69 | 13:42 |
sdague | mriedem: yep | 13:43 |
claudiub | mriedem: also, should I add the "Previously-approved: Mitaka" line in the commit message? In PS2, you said that I should remove it. https://review.openstack.org/#/c/286520/3 | 13:43 |
openstackgerrit | Paul Carlton proposed openstack/nova-specs: Use libvirt storage pools https://review.openstack.org/310505 | 13:43 |
mriedem | passes the request and mutable dict that we set values in which goes to instance.update | 13:43 |
mriedem | ok | 13:43 |
sdague | yeh | 13:43 |
*** PaulMurray has joined #openstack-nova | 13:43 | |
*** zengine has joined #openstack-nova | 13:43 | |
*** swamireddy has joined #openstack-nova | 13:44 | |
*** cdelatte has joined #openstack-nova | 13:44 | |
mriedem | claudiub: for host capabilities, the diff was not trivial, so i don't think previously-approved really applies here | 13:44 |
mriedem | claudiub: especially given all of the other specs around standardizing and exposing capabilities over the rest api | 13:45 |
claudiub | mriedem: sure, but you said in PS3 that I should add it, which is why I'm asking this now. :) | 13:45 |
mriedem | claudiub: i'd like to see the scheduler subteam sort that out. i think edleafe said he's working on sorting out the various capabilities specs yesterday. | 13:45 |
mriedem | claudiub: ah, sorry for the confusion | 13:46 |
PaulMurray | tdurakov, are you around ? | 13:46 |
tdurakov | PaulMurray: hey | 13:46 |
*** thorst_ has joined #openstack-nova | 13:46 | |
*** mdrabe_ has joined #openstack-nova | 13:46 | |
kashyap | This is the test that seems to be failing consistently for me in the Cells job: tempest.api.compute.admin.test_migrations.MigrationsAdminTest.test_resize_server_revert_deleted_flavor | 13:47 |
kashyap | "Server 35054938-f044-4230-b499-2043cd2a3e20 failed to reach VERIFY_RESIZE status and task state "None" within the required time (196 s)" | 13:47 |
*** gszasz has joined #openstack-nova | 13:47 | |
mriedem | kashyap: new test in tempest/ | 13:47 |
mriedem | ? | 13:47 |
PaulMurray | tdurakov, hi - I wanted to go over CI in LM meeting today - will you be able to do that ? | 13:47 |
tdurakov | PaulMurray: there were spring holidays in Russia | 13:47 |
mriedem | kashyap: actually i added that test | 13:47 |
kashyap | mriedem: Fetching the Tempest Git locally to see | 13:47 |
*** mdrabe has quit IRC | 13:47 | |
kashyap | Ah | 13:47 |
tdurakov | PaulMurray: mostly working on the spec now | 13:48 |
*** shaohe_feng has quit IRC | 13:48 | |
kashyap | mriedem: It's for this change (you're familiar with): https://review.openstack.org/#/c/314012/ | 13:48 |
mriedem | kashyap: tempest.api.compute.admin.test_migrations.MigrationsAdminTest.test_resize_server_revert_deleted_flavor ... SKIPPED: Resize not available. | 13:48 |
tdurakov | so, there will be little update | 13:48 |
tdurakov | for ci | 13:48 |
*** jwcroppe has joined #openstack-nova | 13:48 | |
mriedem | hmm, maybe the devstack change to skip those tests for cells isn't in stable/liberty | 13:48 |
*** kengo_sakai has quit IRC | 13:48 | |
*** shaohe_feng has joined #openstack-nova | 13:48 | |
PaulMurray | ok - but worth catching up - we may need to do something related to storage pools | 13:49 |
kashyap | mriedem: Ah, there's a check in DevStack to skip it | 13:49 |
kashyap | Let me go look | 13:49 |
*** swamireddy has quit IRC | 13:49 | |
PaulMurray | continue in the meeting in a while | 13:49 |
mriedem | kashyap: https://github.com/openstack-dev/devstack/blob/stable/mitaka/lib/tempest#L390 | 13:49 |
* kashyap clicks | 13:50 | |
*** thorst_ has quit IRC | 13:50 | |
mriedem | we need this on stable/liberty https://review.openstack.org/#/c/242592/ | 13:50 |
*** thorst_ has joined #openstack-nova | 13:50 | |
mriedem | there could be a few of them that need backporting | 13:50 |
kashyap | Ah-ha | 13:51 |
*** thorst_ has quit IRC | 13:51 | |
tdurakov | PaulMurray: I'll be on weekly for live-migration, happy to help with storage pools | 13:51 |
mriedem | kashyap: it's https://review.openstack.org/#/c/314576/ and https://review.openstack.org/#/c/242592/ | 13:51 |
mriedem | i'll stack those up | 13:52 |
*** esracelik has quit IRC | 13:52 | |
kashyap | mriedem: Ah, you already created the backports | 13:52 |
*** thorst_ has joined #openstack-nova | 13:52 | |
mriedem | kashyap: yeah, i broke it so i can fix it | 13:52 |
coreywright | sdague (voelzmo): i didn't submit that support-both-paramiko-1.x-and-2.x patch because i figured nova would just create a paramiko 2.x-only fix when the global constraints was upgraded rather than the ugly paramiko version testing and import within a function that my patch contains | 13:52 |
coreywright | but i can submit it | 13:52 |
coreywright | by way of gerrit | 13:52 |
*** jwcroppe has quit IRC | 13:53 | |
kashyap | mriedem: Excellent. It can get a straight-forward approval | 13:53 |
sdague | coreywright: yeh, we need to because nova's tests are blocking the paramiko version bump | 13:53 |
coreywright | it's also ugly in that it will have a code path that's untested (as i doubt we are going to test in the gates with both paramiko 1.x and 2.x) | 13:53 |
*** sbelous_ has quit IRC | 13:53 | |
sdague | so it's needed to roll past the issue | 13:53 |
coreywright | sdague: gotcha | 13:53 |
*** sbelous has joined #openstack-nova | 13:54 | |
sdague | coreywright: probably put in a comment that it's temporary until paramiko > 2 is required, then we'll make that change on global-requirements after | 13:54 |
sdague | once landed, we can remove it | 13:54 |
coreywright | yes, i have comments in the commit message and the patch itself about what to do after the upgrade to paramiko 2.x | 13:55 |
mriedem | kashyap: here is the backport https://review.openstack.org/#/c/314581/ | 13:55 |
*** vladikr has joined #openstack-nova | 13:55 | |
* kashyap clicks | 13:56 | |
*** thorst_ has quit IRC | 13:56 | |
*** takedakn has quit IRC | 13:56 | |
PaulMurray | Live Migration meeting on #openstack-meeting-3 in a couple of minutes PaulMurray paul-carlton2 eliqiao shaohe_feng pkoniszewski jlanoux yuntongjin mdbooth kashyap kmartin claudiub tobasco tdurakov andreas_s abhishekk diana_clarke | 13:56 |
*** yamahata has quit IRC | 13:57 | |
kashyap | sdague: Might want to fast-approve this DevStack change from mriedem? Clean backport, fixes a real issue: https://review.openstack.org/#/c/314581/1 | 13:57 |
*** yamahata has joined #openstack-nova | 13:57 | |
sdague | kashyap: yeh, I just did | 13:57 |
kashyap | sdague: Damn, you already did it | 13:57 |
kashyap | :-) | 13:57 |
kashyap | Thanks! | 13:57 |
*** prateek has quit IRC | 13:58 | |
*** thorst_ has joined #openstack-nova | 13:58 | |
*** shaohe_feng has quit IRC | 13:58 | |
*** ametts has joined #openstack-nova | 13:58 | |
*** shaohe_feng has joined #openstack-nova | 13:59 | |
*** tonytan4ever has joined #openstack-nova | 13:59 | |
*** hoangcx_ has joined #openstack-nova | 13:59 | |
*** tbachman has quit IRC | 13:59 | |
*** jwcroppe has joined #openstack-nova | 14:00 | |
*** amotoki has joined #openstack-nova | 14:01 | |
*** afazekas has quit IRC | 14:01 | |
*** afazekas has joined #openstack-nova | 14:01 | |
openstackgerrit | Brian Elliott proposed openstack/nova: Fix service version to update the DB https://review.openstack.org/314230 | 14:01 |
belliott | dansmith: ^ test added for your viewing pleasure | 14:02 |
*** hoangcx_ is now known as hoangcx | 14:02 | |
dansmith | belliott: cool, thanks | 14:03 |
*** roeyc has joined #openstack-nova | 14:03 | |
dansmith | mriedem: we need to backport that to mitaka ^ | 14:03 |
*** tonytan4ever has quit IRC | 14:03 | |
belliott | lots of mox tests in there | 14:04 |
*** obondarev has quit IRC | 14:04 | |
*** mtanino has joined #openstack-nova | 14:04 | |
openstackgerrit | Ludovic Beliveau proposed openstack/nova: Enable python34 tests for nova.tests.unit.pci.test_manager and test_stats https://review.openstack.org/310976 | 14:05 |
*** obondarev_ has joined #openstack-nova | 14:05 | |
*** irenab has quit IRC | 14:05 | |
jaypipes | claudiub: I do now :) | 14:06 |
*** ducttape_ has joined #openstack-nova | 14:06 | |
mriedem | dansmith: ok. was going to review https://review.openstack.org/#/c/302117/ after the live migration meeting and then i'll get to that. | 14:06 |
*** iceyao has joined #openstack-nova | 14:06 | |
dansmith | mriedem: oh sweet, I was going to annoy you about that too | 14:07 |
*** ducttape_ has quit IRC | 14:07 | |
mriedem | is there anything else? | 14:07 |
mriedem | :P | 14:07 |
*** ducttape_ has joined #openstack-nova | 14:07 | |
*** roeyc has quit IRC | 14:07 | |
mriedem | dansmith: maybe you want to review this in the meantime https://review.openstack.org/#/c/295595/ | 14:07 |
claudiub | jaypipes: cool. :) You said during the summit that you want a new table specifically for host_capabilities | 14:08 |
*** zz_dimtruck is now known as dimtruck | 14:08 | |
dansmith | mriedem: ugh | 14:08 |
*** hoangcx has quit IRC | 14:08 | |
*** shaohe_feng has quit IRC | 14:08 | |
*** ccarmack has joined #openstack-nova | 14:08 | |
*** whenry has joined #openstack-nova | 14:08 | |
*** shaohe_feng has joined #openstack-nova | 14:09 | |
claudiub | jaypipes: i have a question on how it will look. Since you probably want it to be fast, will it contain all the capabilities in a single table? Or have a table with id, capability_name and then a cross-table with compute_node_id, capability_id? | 14:09 |
jaypipes | claudiub: the latter. cross table with resource_provider_id, capability_id | 14:09 |
claudiub | jaypipes: ok, cool. good to know. | 14:11 |
openstackgerrit | Corey Wright proposed openstack/nova: crypto: Add support for Paramiko 2.x https://review.openstack.org/314592 | 14:11 |
coreywright | sdague, voelzmo: paramiko 2.x support: https://review.openstack.org/314592 | 14:12 |
jaypipes | claudiub: edleafe is working on a spec that cleans up the representation of much of that stuff. | 14:13 |
jaypipes | claudiub: in the DB/model layer | 14:13 |
jaypipes | edleafe: feel free to link to spec :) | 14:14 |
mriedem | claudiub: i updated the whiteboard for https://blueprints.launchpad.net/nova/+spec/hyperv-vm-diagnostics on the direction | 14:14 |
mriedem | claudiub: a microversoin for the diagnostics api should be an easy add | 14:14 |
* edleafe is in never-ending meetings today | 14:14 | |
markus_z | more meeting for all the people! | 14:16 |
markus_z | *meetings | 14:16 |
markus_z | ;) | 14:16 |
claudiub | jaypipes: edleafe: yeah, that could be useful, actually. | 14:17 |
kashyap | markus_z: The other day Matt Booth pointed me to this excellent "Stop the bleeding of $$ from meetings" ticker https://tobytripp.github.io/meeting-ticker/ | 14:18 |
markus_z | kashyap: muahaha :D | 14:18 |
*** shaohe_feng has quit IRC | 14:19 | |
*** lpetrut has quit IRC | 14:19 | |
*** shaohe_feng has joined #openstack-nova | 14:19 | |
*** lpetrut has joined #openstack-nova | 14:19 | |
claudiub | mriedem: cool, will let abalutoiu know. I don't think there's a blueprint for that yet, right? | 14:20 |
*** gjayavelu has quit IRC | 14:20 | |
mriedem | claudiub: nope | 14:20 |
mriedem | just came up yesterday | 14:20 |
mriedem | the microversion will require a spec | 14:20 |
mriedem | but it's pretty straight-forward | 14:20 |
claudiub | indeed. | 14:20 |
openstackgerrit | Sean Dague proposed openstack/nova: Denormalize personality extension https://review.openstack.org/314600 | 14:22 |
*** swamireddy has joined #openstack-nova | 14:23 | |
sdague | coreywright: https://review.openstack.org/#/c/314595/ should test your patch, and let us move forward | 14:23 |
markus_z | hm, life's ironic sometimes... have to go to a meeting, back in ~ 1h | 14:23 |
*** markus_z has quit IRC | 14:23 | |
openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: virtuozzo instance rescue mode support https://review.openstack.org/217679 | 14:24 |
*** jaosorior has quit IRC | 14:24 | |
openstackgerrit | Claudiu Belu proposed openstack/nova: Compute: Adds driver disk_gb instance overhead estimation https://review.openstack.org/234283 | 14:27 |
edleafe | claudiub: Here is the spec for passing qualitative requirements. https://review.openstack.org/#/c/313784/ | 14:28 |
edleafe | claudiub: I don't have anything yet for changing how they are stored. | 14:28 |
*** shaohe_feng has quit IRC | 14:29 | |
*** shaohe_feng has joined #openstack-nova | 14:29 | |
dansmith | alaski: good catch on that last patch, but unfortunately, the mistake was in the create-instance patch.. I never intended for it to be NULL if we create with no keypair | 14:29 |
mriedem | mikal: you might want to review this https://review.openstack.org/#/c/313696/ | 14:29 |
openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: fix disk size calculation for VZ container instances https://review.openstack.org/222567 | 14:30 |
dansmith | mriedem: I just tossed a supportive -1 on there | 14:33 |
mriedem | <3 | 14:33 |
openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: virtuozzo instance resize support https://review.openstack.org/182257 | 14:34 |
claudiub | edleafe: thanks, will take a look. | 14:36 |
*** pushkaru has joined #openstack-nova | 14:37 | |
alaski | dansmith: a new patch shoved in before the migration would be fine, just as long as create starts populating before the migration | 14:38 |
doffm | edleafe: About yout qualative requirements spec. Have you posted anything to the operators list to get feedback yet? | 14:38 |
dansmith | alaski: I'm fixing it in place already | 14:38 |
*** burgerk has quit IRC | 14:38 | |
dansmith | alaski: still waiting on doffm to update the spec anyway | 14:38 |
doffm | dansmith: Meeting soon. Will do so before lunch. | 14:39 |
dansmith | doffm: cool | 14:39 |
*** thorst_ has quit IRC | 14:39 | |
edleafe | doffm: no, but that's a good idea. Did anything jump out at you from that spec? | 14:39 |
*** shaohe_feng has quit IRC | 14:39 | |
*** thorst_ has joined #openstack-nova | 14:39 | |
*** shaohe_feng has joined #openstack-nova | 14:40 | |
*** belmoreira has quit IRC | 14:40 | |
doffm | edleafe: I think users will like it, but i'm not sure about operators, as billing gets more complicated, and they will have more combinations of instances floating around. | 14:42 |
*** jamie_h has joined #openstack-nova | 14:42 | |
doffm | Would be good to see some operator feedback. | 14:42 |
*** thorst_ has quit IRC | 14:44 | |
openstackgerrit | Dan Smith proposed openstack/nova: Create Instances with keypairs https://review.openstack.org/312751 | 14:45 |
openstackgerrit | Dan Smith proposed openstack/nova: Add Keypairs to the API database https://review.openstack.org/312755 | 14:45 |
openstackgerrit | Dan Smith proposed openstack/nova: Make Keypair object favor the API database https://review.openstack.org/312756 | 14:45 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to move keypairs from main to API database https://review.openstack.org/313664 | 14:45 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to store keypairs with instances https://review.openstack.org/313060 | 14:45 |
*** pwnall has joined #openstack-nova | 14:46 | |
edleafe | doffm: yeah, that's along the lines of what I expected. Do you see any other way to get away from the kind of flavor explosion that adding qualitative requirements will result in? | 14:47 |
*** shaohe_feng has quit IRC | 14:49 | |
*** shaohe_feng has joined #openstack-nova | 14:50 | |
*** cdent has joined #openstack-nova | 14:52 | |
doffm | edleafe: I guess provide the tools to operators to state what options are available on what flavors. | 14:52 |
*** pcaruana has joined #openstack-nova | 14:53 | |
*** mariojv has left #openstack-nova | 14:53 | |
*** thorst_ has joined #openstack-nova | 14:54 | |
*** mriedem is now known as mriedem_meeting | 14:55 | |
*** ctrath has quit IRC | 14:56 | |
*** voelzmo has quit IRC | 14:57 | |
*** abalutoiu has quit IRC | 14:58 | |
*** pwnall has quit IRC | 14:58 | |
*** thorst_ has quit IRC | 14:58 | |
*** pwnall has joined #openstack-nova | 14:58 | |
*** shaohe_feng has quit IRC | 15:00 | |
*** davidgiluk has joined #openstack-nova | 15:00 | |
*** ctrath has joined #openstack-nova | 15:00 | |
*** shaohe_feng has joined #openstack-nova | 15:00 | |
PaulMurray | mdbooth, paul-carlton pkoniszewski I think that discussion is a long and tangled one | 15:00 |
mdbooth | So, I think these are all the same spec: https://review.openstack.org/#/c/248358/ https://review.openstack.org/#/c/306561/ https://review.openstack.org/#/c/301509/ | 15:00 |
pkoniszewski | PaulMurray: yeah, it is | 15:00 |
*** zengine has quit IRC | 15:01 | |
mdbooth | They're all about causing a migration to complete. They should be designed together. | 15:01 |
mdbooth | They all require each other as context. | 15:01 |
tdurakov | tonyb: hi, are you around? | 15:01 |
davidgiluk | mdbooth: Yeh there's some subtelty - postcopy can force completion, auto-converge might not | 15:01 |
paul-carlton | maybe, happy to fold the other two into my spec if you like but I am concerned that will lead to nothing getting approved | 15:02 |
pkoniszewski | well, post copy and auto converge specs are about adding a flag for admin to force these two things | 15:02 |
mdbooth | davidgiluk: But it's just that: subtlety | 15:02 |
mdbooth | pkoniszewski: Right, and having the design of those 2 flags in 2 different specs is bad | 15:02 |
PaulMurray | mdbooth, paul-carlton pkoniszewski I want to be in this discussion but I have to go get my daughter - go on without me, but I will catch up later | 15:02 |
davidgiluk | mdbooth: Yeh I agree they're related | 15:02 |
pkoniszewski | paul-carlton specs is about implementing some kind of an automated mechanism in nova to take care of migrations | 15:02 |
mdbooth | Because they're both highly related | 15:02 |
paul-carlton | better to approve post-copy, it is simple to do | 15:02 |
danpb | davidgiluk: well auto-converge will converge - as it eventually throttles to 99% | 15:02 |
pkoniszewski | whats more is that auto converge might not force to complete | 15:02 |
danpb | unless you have awful network bandwidth | 15:03 |
pkoniszewski | even if you throttle down by 99% it might still not end | 15:03 |
pkoniszewski | and 100% is pause | 15:03 |
davidgiluk | pkoniszewski: Yeh I'm OK with them being separate specs, as long as they turn into something that cooperates between them | 15:03 |
danpb | i agree with mdbooth though, we should really just have one spec IMHO | 15:03 |
*** thorst_ has joined #openstack-nova | 15:03 | |
danpb | it doesn;t make sense to try to design use of each feature in isolation | 15:03 |
PaulMurray | danpb, mdbooth I agree - the question was about how long it will take | 15:04 |
PaulMurray | I can't see that being agreed in next couple of weeks | 15:04 |
pkoniszewski | danpb: do we want to put flags and all automated stuff in one spec? + all things requires for post-copt support in nova? | 15:04 |
danpb | PaulMurray: imho it'll take longer to get approved with 3 specs than with 1 | 15:04 |
pkoniszewski | danpb: instead of doing it in small, and honestly doable, small pieces? | 15:04 |
PaulMurray | if it has to be agreed first then so be it, but just thinking practically | 15:04 |
paul-carlton | danpb ok, happy to fold the auto-converge and post-copy into my spec if you would prefer | 15:04 |
danpb | not least because it makes it harder for the reviewer to see wtf the big picture is when you have to switch back & forth between reading 3 specs | 15:04 |
PaulMurray | we still need to work out how to expose | 15:04 |
*** Jeffrey4l_ has quit IRC | 15:05 | |
danpb | PaulMurray: what do you mean by that ? | 15:05 |
PaulMurray | danpb, which bit ? | 15:05 |
PaulMurray | how to expose ? | 15:05 |
danpb | PaulMurray: "we still need to work out how to expose" | 15:05 |
paul-carlton | Following, danpb's comments on my spec I think we have a reasonable solution which could be approved and implemented | 15:06 |
*** mdrabe_ has quit IRC | 15:06 | |
PaulMurray | I meant the discussion about how the options are presented through the API | 15:06 |
danpb | there's nothing to be exposed in the API at all | 15:06 |
mdbooth | danpb: That's really my issue. I have difficulty getting the context, and I've been actively trying to follow it. | 15:06 |
*** xenogear has joined #openstack-nova | 15:06 | |
danpb | this is exclusively something to be done inside the nova libvirt driver | 15:06 |
paul-carlton | it pretty much requires the post-copy feature to be effective, auto-converge is not particularly useful from what I'm hearing | 15:07 |
danpb | yes & no - auto-converge is good at convergance at the cost of high guest impact | 15:07 |
*** gagehugo has joined #openstack-nova | 15:07 | |
danpb | so if you have a choice between post-copy + auto-converge then pick post-copy | 15:07 |
mdbooth | danpb: I think you showed that post-copy has lower impact than auto-converge, right? | 15:08 |
* PaulMurray I really do have to go | 15:08 | |
danpb | nb that post-copy requires a newer qemu version than auto-converge | 15:08 |
* PaulMurray will be back soon | 15:08 | |
danpb | so for some people auto-converge will be the only available option | 15:08 |
mdbooth | The other thing is, I believe my understanding of the risk in post-copy was wrong. | 15:08 |
paul-carlton | Yes we already have live-migration-force-complete, we'd just change the implementation when post copy is available to do that instead of pause | 15:09 |
mdbooth | I'd been hearing talk of migrations 'failing' 90% of the time, which was used in the context of an argument against post-copy | 15:09 |
danpb | mdbooth: there is a risk in post-copy, but I believe the severity of the risk is rather over-blown | 15:09 |
mdbooth | But the 'failure' here is actually failure to converge | 15:09 |
luis5tb | but, besides forcing completion, why post-copy cannot be the default option (if the admin decides so) | 15:09 |
mdbooth | Which wouldn't happen with post-copy | 15:09 |
*** sacharya has joined #openstack-nova | 15:09 | |
*** shaohe_feng has quit IRC | 15:10 | |
mdbooth | So yes, to me the additional failure mode of post-copy actually seems pretty much irrelevant | 15:10 |
davidgiluk | mdbooth: Right, with postcopy you get no convergence failures, but if you lose network during the postcopy phase you're stuck | 15:10 |
mdbooth | davidgiluk: Right. But if you lose network without postcopy you're also stuck. | 15:10 |
mdbooth | So it's a failure case either way. | 15:10 |
danpb | that risk of network failure is pretty tiny if you have redundancy in your network too | 15:10 |
davidgiluk | mdbooth: No | 15:10 |
paul-carlton | post-copy exposes the instance to the risk of failure but a reboot will recover it | 15:10 |
davidgiluk | mdbooth: If you lose network on the migration network then the source can carry on running | 15:11 |
*** shaohe_feng has joined #openstack-nova | 15:11 | |
*** rmart04 has joined #openstack-nova | 15:11 | |
davidgiluk | mdbooth: That's no longer the case with postcopy (We do have a GSoC student looking at recovery from net failure) | 15:11 |
mdbooth | davidgiluk: What are the chances of losing the migration network, but not all other networking, though? | 15:11 |
danpb | mdbooth: its non-zero, but IMHO small enough that most people won't care | 15:12 |
*** sridharg has quit IRC | 15:12 | |
danpb | particularly if they have network bonding failover | 15:12 |
*** sudipto has joined #openstack-nova | 15:12 | |
mdbooth | Right. It's pretty much theoretical afaict. | 15:12 |
*** irenab has joined #openstack-nova | 15:12 | |
davidgiluk | mdbooth: Yeh I mean it's just a risk, it's there but how paranoid depends on your workload/personality/etc | 15:12 |
mdbooth | If you lose network, it's likely because the cleaner unplugged it (metaphorically) | 15:12 |
danpb | and in cloud we tell people their VMs can die unrecoverably at any time regardless | 15:12 |
mdbooth | In which case you lose the lot | 15:12 |
paul-carlton | luis5tb, I think the default should be the status quo, i.e. no post-copy, if the cloud operator wants to turn it on then they should do so | 15:12 |
mdbooth | And in that case your instance is useless anyway | 15:12 |
danpb | people can call them "pets" but they have to accept there's a non-zero chance of their pet going away | 15:12 |
danpb | paul-carlton: i rather disagree - the default should be post-copy | 15:13 |
luis5tb | paul-carlton: I agree, but if the operator want to change it, we should provide a way (at nova.conf) to do it, right? | 15:13 |
danpb | paul-carlton we should not default to a config that is guaranteed to never complete for many workloads | 15:13 |
*** mageshgv has joined #openstack-nova | 15:13 | |
danpb | we should default to post-copy because it "just works" and let people turn it off if they absolutely insist they don't want it | 15:13 |
*** jdurgin1 has joined #openstack-nova | 15:14 | |
*** burgerk has joined #openstack-nova | 15:14 | |
mdbooth | Also, post-copy only needs to transfer the working set to be fast again | 15:14 |
paul-carlton | danpb ok by me, release note can cover risks | 15:14 |
*** abalutoiu has joined #openstack-nova | 15:14 | |
* davidgiluk would default to disabling it for now and then if people find they use it a lot and are happy then we can always flip to enable by default later? | 15:14 | |
danpb | we generally try to avoid having config options that amount to telling the user to set "make_it_work=yes" | 15:15 |
mdbooth | You might have a workaround to turn it off | 15:15 |
luis5tb | and then, if post-copy is enabled by default, it should change migration to postcopy mode as soon as possible? | 15:16 |
danpb | luis5tb: no, there should be active monitoring by nova to decide when its appropriate | 15:16 |
danpb | luis5tb: if you switch to post-copy too soon you uneccessarily penalize performance | 15:16 |
dansmith | IMHO, changing the default right now is probably not the best plan | 15:16 |
dansmith | based on the feedback in the room when we discussed it | 15:16 |
danpb | you want to run in pre-copy mode for some period of time and monitor progress - if it stops making progress then you should switch to post-copy | 15:17 |
mdbooth | dansmith: This was my point earlier, though. I think the feedback in the room was based on misinformation. | 15:17 |
dansmith | mdbooth: it doesn't really matter I don't think | 15:17 |
danpb | dansmith: i wasn't there to hear what was said, but i don't see a compelling reason not to enable it by default honestly | 15:17 |
dansmith | mdbooth: changing defaults always has some impact and aspect of surprise | 15:17 |
mdbooth | dansmith: I think it's grown unnecessarily into a bogeyman, which very few people will ever care about in practise. | 15:18 |
dansmith | danpb: it immediately makes live migrations more risky | 15:18 |
danpb | it is clearly better than the current situation in that it actually works reliably | 15:18 |
dansmith | I hear you | 15:18 |
danpb | dansmith: the risk is overblown really | 15:18 |
*** mdrabe has joined #openstack-nova | 15:19 | |
paul-carlton | per https://review.openstack.org/#/c/306561 the proposal is that unless the operator indicates that the migration should be completed at all costs post-copy will only be used if the live-migration-force-complete operation is used | 15:19 |
danpb | yes, there's a theoretical risk, but in practice if you're trying to evacuate a machine which is dieing | 15:19 |
dansmith | danpb: one of you said default it to post-copy and then later allow going back to the original model, or a workaround, or whatever, and I don't think that's a reasonable change to make | 15:19 |
danpb | you have the actual real risk of loosing all vms because you can't get them off the dieing machine in time | 15:19 |
*** irenab has quit IRC | 15:19 | |
dansmith | danpb: there are plenty of reasons to live migrate other than imminent death of the host | 15:19 |
openstackgerrit | Karen Bradshaw proposed openstack/nova: api-ref: server shelve action https://review.openstack.org/314629 | 15:20 |
*** shaohe_feng has quit IRC | 15:20 | |
danpb | dansmith: and there are plenty of other failures that can hit VMs at any time in nova when no migration is taking place | 15:20 |
*** tbachman has joined #openstack-nova | 15:20 | |
dansmith | danpb: I know | 15:20 |
*** shaohe_feng has joined #openstack-nova | 15:20 | |
paul-carlton | which is why I'm proposing the live-migration 'importance' flag to allow the operator to say this is a vital migration rather than a best effort one | 15:20 |
danpb | that are more likely than the new failure mode of post-copy | 15:20 |
mdbooth | dansmith: Either way, I think all of these things can be discussed in no more than 1 spec :) | 15:21 |
mdbooth | There are currently 3 | 15:21 |
*** whenry has quit IRC | 15:21 | |
dansmith | mdbooth: surely :) | 15:21 |
mdbooth | paul-carlton: I personally favour moving the other specs into your spec | 15:22 |
*** kbyrne has quit IRC | 15:22 | |
paul-carlton | danpb, I think you suggested a one spec approach too? | 15:23 |
*** chuck__ is now known as zul | 15:24 | |
paul-carlton | if so and luis5tb agrees I'll update my spec to incorporate the other two? | 15:24 |
danpb | i've just -2d the two supplementary specs | 15:25 |
paul-carlton | ok, I'll update mine | 15:25 |
danpb | so we have paul-carlton's as the main point | 15:25 |
*** jlanoux has quit IRC | 15:25 | |
*** hemnafk is now known as hemna | 15:27 | |
*** armax has joined #openstack-nova | 15:27 | |
luis5tb | ok | 15:28 |
*** iceyao has quit IRC | 15:28 | |
*** sacharya_ has joined #openstack-nova | 15:29 | |
*** shaohe_feng has quit IRC | 15:30 | |
*** sacharya has quit IRC | 15:31 | |
*** abalutoiu has quit IRC | 15:31 | |
*** shaohe_feng has joined #openstack-nova | 15:31 | |
tdurakov | mriedem_meeting: with the help of #openstack-infra folks we could you multinode with xenial, so I'll change image for live-migration job and we could compare it's stability | 15:32 |
*** zenoway has quit IRC | 15:32 | |
*** zenoway has joined #openstack-nova | 15:33 | |
*** kbyrne has joined #openstack-nova | 15:36 | |
*** irenab has joined #openstack-nova | 15:36 | |
*** mjura has quit IRC | 15:37 | |
*** andreas_s has quit IRC | 15:37 | |
*** zenoway has quit IRC | 15:38 | |
*** rena9067 has joined #openstack-nova | 15:38 | |
*** andymaier_ has quit IRC | 15:39 | |
*** aysyd has quit IRC | 15:39 | |
*** flwang has quit IRC | 15:39 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Work with latest paramiko 2.0.0 https://review.openstack.org/314637 | 15:40 |
*** kbyrne has quit IRC | 15:40 | |
*** itamarl has quit IRC | 15:40 | |
*** shaohe_feng has quit IRC | 15:41 | |
*** amotoki has quit IRC | 15:41 | |
openstackgerrit | Sean Dague proposed openstack/nova: Drop paramiko < 2 compat code https://review.openstack.org/314639 | 15:41 |
*** shaohe_feng has joined #openstack-nova | 15:41 | |
sdague | dims: that patch is already up tehre | 15:41 |
*** mpavone has quit IRC | 15:41 | |
dims | sdague : ah did not notice. give me a sec | 15:42 |
PaulMurray | danpb, so looks like I missed the whole discussion, but just to note, I think of config options and API the same way | 15:42 |
PaulMurray | danpb, looks like you propose we just start using post-copy | 15:43 |
sdague | dims: sure, I linked to it and -2ed yours. If you find issues, let's work on coreywright's patch | 15:43 |
sdague | I've got a global requirements change stacked in there as well so we can see it working, and then delete the compat code after upgrade | 15:43 |
*** moshele has quit IRC | 15:44 | |
*** aysyd has joined #openstack-nova | 15:44 | |
*** kolesovdv has quit IRC | 15:45 | |
jaypipes | sdague: hey, in Austin, you showed a link to some nice API devref documents that we are working on... do you have that link handy? | 15:45 |
*** markus_z has joined #openstack-nova | 15:45 | |
*** paul-carlton1 has quit IRC | 15:46 | |
openstackgerrit | Sarafraj Singh proposed openstack/nova: [WIP] api-ref: image.inc - Update method validation https://review.openstack.org/314268 | 15:46 |
*** woodster_ has joined #openstack-nova | 15:46 | |
*** yamahata has quit IRC | 15:47 | |
sdague | jaypipes: in tree it's nova/api-ref | 15:47 |
openstackgerrit | Anusha Unnam proposed openstack/nova: api-ref: os-volumes.inc method verification https://review.openstack.org/314325 | 15:47 |
sdague | it publishes to http://developer.openstack.org/api-ref/compute/ | 15:47 |
*** sdake_ has quit IRC | 15:47 | |
jaypipes | sdague: thanks Sean, appreciated. | 15:48 |
sdague | also, see all the api-ref emails, as we're sprinting this week to try to get the content up to snuff | 15:48 |
*** Oku_OS is now known as Oku_OS-away | 15:49 | |
*** lpetrut has quit IRC | 15:49 | |
*** gjayavelu has joined #openstack-nova | 15:50 | |
*** sdake has joined #openstack-nova | 15:50 | |
*** zenoway has joined #openstack-nova | 15:51 | |
*** shaohe_feng has quit IRC | 15:51 | |
markus_z | claudiub: Wanna have that in stable/mitaka? https://git.openstack.org/cgit/openstack/nova/commit/?id=94b40cf43ca3ab7d85fd15c8fda9baf8ed07f8a1 | 15:52 |
*** shaohe_feng has joined #openstack-nova | 15:52 | |
*** tesseract has quit IRC | 15:52 | |
*** sdake_ has joined #openstack-nova | 15:54 | |
*** Sukhdev has joined #openstack-nova | 15:55 | |
*** zenoway has quit IRC | 15:55 | |
claudiub | markus_z: hey, sure. it will reduce the amount of random py27 gate failures and it's an extremely simple commit. i think it's worth it. | 15:56 |
openstackgerrit | John Garbutt proposed openstack/nova: Config options: centralize scality libvirt options (15) https://review.openstack.org/309357 | 15:56 |
*** abalutoiu has joined #openstack-nova | 15:56 | |
markus_z | claudiub: okidoki, I finish another thing, then do a backport | 15:56 |
*** ccarmack has quit IRC | 15:56 | |
openstackgerrit | John Garbutt proposed openstack/nova: Config options: centralize smbfs libvirt options (16) https://review.openstack.org/309360 | 15:57 |
claudiub | markus_z: cool, let me know when you're done. :) | 15:57 |
openstackgerrit | John Garbutt proposed openstack/nova: Config options: centralize remotefs libvirt options (end) https://review.openstack.org/309363 | 15:57 |
*** sdake has quit IRC | 15:57 | |
*** PaulMurray has quit IRC | 15:57 | |
*** flwang has joined #openstack-nova | 15:58 | |
*** erhudy has joined #openstack-nova | 15:58 | |
tdurakov | markus_z: hi | 15:59 |
*** thorst_ has quit IRC | 15:59 | |
markus_z | tdurakov: hey ho | 15:59 |
tdurakov | markus_z: any updates on latest libvirt/qemu devstack plugin? | 15:59 |
*** davidgiluk has left #openstack-nova | 15:59 | |
erhudy | could i get somebody to consider the spec at https://review.openstack.org/#/c/311828/ and provide critical feedback | 16:00 |
markus_z | tdurakov: unfortunately not, sorry :( | 16:00 |
tdurakov | markus_z: np | 16:01 |
*** apoorvad has joined #openstack-nova | 16:01 | |
*** shaohe_feng has quit IRC | 16:01 | |
*** thorst_ has joined #openstack-nova | 16:01 | |
*** shaohe_feng has joined #openstack-nova | 16:02 | |
*** matrohon has quit IRC | 16:02 | |
markus_z | tdurakov: For the rest of the week I will only work a few hours per day, not sure if I can push it this week | 16:02 |
*** roeyc has joined #openstack-nova | 16:04 | |
*** penick has joined #openstack-nova | 16:04 | |
tdurakov | markus_z: trying to switch to xenial instead. | 16:05 |
markus_z | tdurakov: Should mitigate the issue for a while I guess, yeah | 16:06 |
*** ericksonsantos has joined #openstack-nova | 16:06 | |
*** roeyc has quit IRC | 16:08 | |
*** ctrath has quit IRC | 16:09 | |
openstackgerrit | Anusha Unnam proposed openstack/nova: api-ref: os-volumes.inc method verification https://review.openstack.org/314325 | 16:10 |
*** thorst_ has quit IRC | 16:10 | |
*** ctrath has joined #openstack-nova | 16:11 | |
*** thorst_ has joined #openstack-nova | 16:11 | |
*** thorst_ has quit IRC | 16:11 | |
*** shaohe_feng has quit IRC | 16:11 | |
*** ctrath has quit IRC | 16:12 | |
*** sacharya_ has quit IRC | 16:12 | |
*** jlanoux has joined #openstack-nova | 16:12 | |
*** ccarmack has joined #openstack-nova | 16:13 | |
*** daemontool_ has quit IRC | 16:13 | |
*** shaohe_feng has joined #openstack-nova | 16:14 | |
*** mriedem_meeting is now known as mriedem | 16:14 | |
mriedem | tdurakov: sounds good, thanks | 16:15 |
tdurakov | mriedem: both patches already got 2 +2 and W | 16:15 |
tdurakov | mriedem: hope to test them in a few hours | 16:15 |
*** ctrath has joined #openstack-nova | 16:16 | |
*** thorst_ has joined #openstack-nova | 16:17 | |
*** paul-carlton1 has joined #openstack-nova | 16:19 | |
openstackgerrit | Mark Doffman proposed openstack/nova-specs: Migrate keypair table to the API DB https://review.openstack.org/312619 | 16:20 |
dansmith | johnthetubaguy: I have literally never been able to auto-complete your username on gerrit, so can you see my question on this? https://review.openstack.org/#/c/313072/1 | 16:21 |
johnthetubaguy | dansmith: yeah, I think my move from citrix did something bad to my metadata | 16:21 |
jroll | dansmith: oh god | 16:21 |
johnthetubaguy | anyways, looking at your question | 16:21 |
doffm | alaski: I left the keypair 'instance-extra' work in the keypair migration spec. I hope thats OK. Did you want a different name for the specifications because of that? | 16:21 |
*** winston-d_ has joined #openstack-nova | 16:21 | |
dansmith | jroll: "oh god" as in "oh god, tell me about it!" ? | 16:21 |
*** shaohe_feng has quit IRC | 16:22 | |
johnthetubaguy | dansmith: I think "you just broke my billing" is another phrase, although I wish I knew for sure | 16:22 |
*** thorst_ has quit IRC | 16:22 | |
*** shaohe_feng has joined #openstack-nova | 16:22 | |
*** Drankis has quit IRC | 16:22 | |
dansmith | johnthetubaguy: yeah, I don't know how people do that, so .. | 16:23 |
jroll | dansmith: I just panic on our virt team's behalf when I see BandwidthUsage queries | 16:23 |
jroll | johnthetubaguy: off the top of my head, we emit a notification with the sum of those every 24h, so I agree it looks like a bit of lost revenue | 16:23 |
dansmith | jroll: ah | 16:24 |
*** Raildo_ has joined #openstack-nova | 16:24 | |
dansmith | johnthetubaguy: well, I'll let you coordinate with mriedem and lxsli on the way to fix the problem without losing money :) | 16:24 |
dansmith | jroll: I thought you were sympathizing with my hatred of trying to find johnthetubaguy on gerrit | 16:25 |
mriedem | dansmith: easy, start with john@ | 16:25 |
mriedem | and it fills in the rest | 16:25 |
mriedem | finding cburgess on gerrit, on the other hand, is nearly impossible | 16:25 |
openstackgerrit | Chris Dent proposed openstack/nova-specs: resource-providers: generic resource pools https://review.openstack.org/300176 | 16:26 |
cburgess | mriedem Do what now? | 16:26 |
dansmith | mriedem: okay, well, I usually look for johng<...> which fails, I guess I need to remember john@.. | 16:26 |
cdent | jaypipes, bauzas, mriedem ^^ adjusted the api spec to reflect my novella | 16:26 |
dansmith | just "john" also returns nothing useful of course | 16:26 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: move image_file_url download options https://review.openstack.org/314668 | 16:26 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: deprecate "file transfer" feature for Glance images https://review.openstack.org/314669 | 16:26 |
jroll | dansmith: no, though that used to be the case for me | 16:27 |
jroll | johnthetubaguy: infra can usually help fix the gerrit thing | 16:27 |
*** rena9067 has quit IRC | 16:28 | |
dansmith | well, john@ seems to work fine | 16:28 |
jroll | oh never mind, yours isn't broken | 16:28 |
lxsli | dansmith: it's case-sensitive | 16:28 |
jroll | mine would just error | 16:28 |
dansmith | I just wish gerrit search actually ... searched instead of just completed | 16:28 |
jroll | right :P | 16:28 |
mriedem | i find jroll via jim. | 16:28 |
mriedem | @ and . are key | 16:28 |
*** zengine has joined #openstack-nova | 16:28 | |
mriedem | james von rollenhagen esq IV | 16:28 |
mriedem | you'll never find him drinking with his pinky down | 16:29 |
*** Sukhdev has quit IRC | 16:29 | |
dansmith | jim. doesn't do it for me | 16:29 |
lxsli | johnthetubaguy: I have to go, leave a comment and I'll look at it tomorrow | 16:29 |
johnthetubaguy | lxsli: ack | 16:29 |
jroll | yeah, should be jim@ :P | 16:29 |
dansmith | yeah | 16:29 |
*** Sukhdev has joined #openstack-nova | 16:30 | |
mriedem | oh i was using the rax email then | 16:30 |
johnthetubaguy | jroll: yeah, I should ping them | 16:30 |
jroll | oh wow | 16:30 |
jroll | johnthetubaguy: nah, my mistake, yours isn't broken to the point of infra intervention | 16:30 |
*** chuck__ has joined #openstack-nova | 16:30 | |
*** abalutoiu has quit IRC | 16:31 | |
*** zul has quit IRC | 16:31 | |
*** chuck__ is now known as zul | 16:31 | |
*** ijw has joined #openstack-nova | 16:32 | |
*** shaohe_feng has quit IRC | 16:32 | |
*** shaohe_feng has joined #openstack-nova | 16:32 | |
markus_z | claudiub: https://review.openstack.org/#/c/314671/ | 16:32 |
*** jlanoux has quit IRC | 16:34 | |
jaypipes | cdent: thx | 16:35 |
cdent | jaypipes: I _think_ I got everything but it's hard to be 100% sure | 16:35 |
cdent | gah typo | 16:36 |
*** abitha has joined #openstack-nova | 16:36 | |
*** yamahata has joined #openstack-nova | 16:36 | |
openstackgerrit | Chris Dent proposed openstack/nova-specs: resource-providers: generic resource pools https://review.openstack.org/300176 | 16:37 |
*** jdillaman has quit IRC | 16:37 | |
*** sdake_ is now known as sdake | 16:38 | |
*** thorst_ has joined #openstack-nova | 16:38 | |
*** Swami has joined #openstack-nova | 16:38 | |
*** hieulq_ has joined #openstack-nova | 16:39 | |
*** sfinucan has quit IRC | 16:42 | |
*** shaohe_feng has quit IRC | 16:42 | |
*** thorst_ has quit IRC | 16:42 | |
*** EinstCrazy has quit IRC | 16:42 | |
*** shaohe_feng has joined #openstack-nova | 16:43 | |
*** diogogmt has joined #openstack-nova | 16:43 | |
*** rbridgeman has joined #openstack-nova | 16:43 | |
*** thorst_ has joined #openstack-nova | 16:44 | |
*** kengo_sakai has joined #openstack-nova | 16:45 | |
*** jaypipes has quit IRC | 16:45 | |
*** flaper87 has quit IRC | 16:46 | |
*** flaper87 has joined #openstack-nova | 16:46 | |
*** cdelatte has quit IRC | 16:46 | |
*** cdelatte has joined #openstack-nova | 16:47 | |
*** jaypipes has joined #openstack-nova | 16:47 | |
*** sacharya has joined #openstack-nova | 16:47 | |
*** jaypipes has quit IRC | 16:47 | |
*** thorst_ has quit IRC | 16:49 | |
*** kengo_sakai has quit IRC | 16:49 | |
alaski | doffm: that's fine. my comment on the name was that you mispelled keypairs | 16:49 |
*** jaypipes has joined #openstack-nova | 16:50 | |
mriedem | danpb: can you take a quick pass through mdbooth's libvirt image backend refactor spec before we approve? https://review.openstack.org/#/c/302117/ | 16:50 |
*** danpb has quit IRC | 16:50 | |
*** sacharya has quit IRC | 16:50 | |
*** derekh has quit IRC | 16:50 | |
*** sacharya has joined #openstack-nova | 16:50 | |
*** shaohe_feng has quit IRC | 16:52 | |
*** shaohe_feng has joined #openstack-nova | 16:53 | |
sdague | anyone want to take a quick look at this - https://review.openstack.org/#/c/314592/ so we can roll forward to paramiko 2? | 16:53 |
*** abalutoiu has joined #openstack-nova | 16:54 | |
*** rama_y has joined #openstack-nova | 16:54 | |
*** rbridgeman has quit IRC | 16:54 | |
*** unicell has quit IRC | 16:55 | |
*** delattec has joined #openstack-nova | 16:56 | |
*** mvk has quit IRC | 16:57 | |
mriedem | sdague: isn't there a cleaner way to get the version using pkg_resources? | 16:57 |
sdague | mriedem: possibly, but this code should only exist for about 4 days | 16:57 |
*** andymaier_ has joined #openstack-nova | 16:57 | |
dansmith | cdent: jaypipes: is the code up to migrate the existing inventory stuff to the api db? | 16:58 |
sdague | there is already the g-r change + the full removal of the v1 code stacked up | 16:58 |
cdent | dansmith: not that I'm aware of | 16:58 |
dansmith | cdent: who is working on it? | 16:59 |
cdent | dansmith: I don't know | 16:59 |
mriedem | sdague: ok, +W | 16:59 |
*** penick has quit IRC | 16:59 | |
*** cdelatte has quit IRC | 16:59 | |
cdent | it's only been discussd, as far as I know noone has "taken that action" | 16:59 |
dansmith | cdent: okay, it's blocking everything else, so ... seems like someone should be doing it | 17:00 |
* cdent dies in a fire of business speak | 17:00 | |
cdent | you want to do it dansmith ? | 17:00 |
dansmith | no, I was just hoping it was up for review | 17:00 |
*** vilobhmm11 has joined #openstack-nova | 17:00 | |
*** daemontool_ has joined #openstack-nova | 17:00 | |
cdent | I agree it needs to happen, but I'm not sure who has the cycles, _right now_ | 17:00 |
*** roeyc has joined #openstack-nova | 17:01 | |
dansmith | since it's blocking everything, I would think the set of people would be "anyone working on anything else resource-pools-related" | 17:01 |
mriedem | dansmith: on https://review.openstack.org/#/c/295595/ i think i'll throw that in the nova meeting for thursday | 17:01 |
dansmith | mriedem: okay | 17:02 |
cdent | that's great dansmith but I know at least I am carrying a huge amount of state that I can't set down this minute | 17:02 |
cdent | so it's not going to be me | 17:02 |
cdent | at least not in the next few days | 17:02 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: improve help text of database (related) options (1/3) https://review.openstack.org/301238 | 17:02 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: centralize section "database" + "api_database" https://review.openstack.org/263804 | 17:02 |
jaypipes | dansmith: I will do it. | 17:02 |
*** shaohe_feng has quit IRC | 17:03 | |
*** jaypipes has quit IRC | 17:03 | |
*** shaohe_feng has joined #openstack-nova | 17:03 | |
mriedem | rlrossit might like to get in on some online data migration action | 17:03 |
mriedem | but i think he's busy with internal things | 17:04 |
cdent | reality is a such a pita | 17:04 |
rlrossit | mriedem: oh you, voluntelling me for more things | 17:04 |
*** jaypipes has joined #openstack-nova | 17:04 | |
*** lpetrut has joined #openstack-nova | 17:06 | |
rlrossit | mriedem: is this scheduler migrating, or cells migrating? | 17:06 |
*** roeyc has quit IRC | 17:06 | |
openstackgerrit | Corey Wright proposed openstack/nova: Drop paramiko < 2 compat code https://review.openstack.org/314639 | 17:06 |
*** sacharya has quit IRC | 17:08 | |
*** sacharya has joined #openstack-nova | 17:08 | |
*** yassou has quit IRC | 17:10 | |
*** thorst_ has joined #openstack-nova | 17:11 | |
*** kfarr has joined #openstack-nova | 17:12 | |
*** shaohe_feng has quit IRC | 17:13 | |
doffm | alaski: Ohh. Ok I guess I'll fix that now. | 17:13 |
*** shaohe_feng has joined #openstack-nova | 17:13 | |
openstackgerrit | Mark Doffman proposed openstack/nova-specs: Migrate keypair table to the API DB https://review.openstack.org/312619 | 17:14 |
cdent | rlrossit: the things that was being discussed was migrating the inventory and allocations tables to api db | 17:14 |
dansmith | doffm: alaski: oops, I didn't notice that it wasn't fixed | 17:15 |
*** thorst_ has quit IRC | 17:15 | |
*** rmart04 has quit IRC | 17:15 | |
doffm | dansmith: alaski: Sorry. | 17:16 |
*** thorst_ has joined #openstack-nova | 17:16 | |
alaski | no worries | 17:16 |
*** gagehugo has quit IRC | 17:18 | |
*** nic has joined #openstack-nova | 17:18 | |
*** gagehugo has joined #openstack-nova | 17:19 | |
*** ccarmack has quit IRC | 17:20 | |
*** jistr has quit IRC | 17:21 | |
*** whenry has joined #openstack-nova | 17:21 | |
*** shaohe_feng has quit IRC | 17:23 | |
*** unicell has joined #openstack-nova | 17:23 | |
*** thorst_ has quit IRC | 17:23 | |
*** shaohe_feng has joined #openstack-nova | 17:24 | |
cdent | dansmith, jaypipes: is there an etherpad reference somewhere to what parts of the "inventories stuff" will need to be migrated? What's the status of the reosurce provider and resource provider aggregate tables? They are remaining in the cell db? | 17:24 |
*** daemontool_ has quit IRC | 17:25 | |
*** ccarmack has joined #openstack-nova | 17:25 | |
*** ccarmack has joined #openstack-nova | 17:26 | |
mriedem | resource provider aggregates moves to the api db because that's where aggregates are moving | 17:26 |
jaypipes | cdent: currently brainstorming this with dansmith in a /query window.. | 17:26 |
jaypipes | cdent: gimme a few :) | 17:26 |
cdent | it seems like quite a bit will need to move because we want the fk relationship across that resource_providers_aggregates table with both aggregates and resource_providers table | 17:28 |
cdent | jaypipes: bon chance | 17:28 |
*** gjayavelu has quit IRC | 17:29 | |
smcginnis | Just looking at this: https://github.com/openstack/governance/blob/master/reference/projects.yaml#L2516 | 17:29 |
smcginnis | Are stable releases done for python-novaclient? | 17:30 |
*** thorst_ has joined #openstack-nova | 17:30 | |
smcginnis | mriedem: ^^ | 17:30 |
*** penick has joined #openstack-nova | 17:30 | |
mriedem | smcginnis: yeah, as needed | 17:30 |
mriedem | but not often | 17:30 |
smcginnis | mriedem: So are newer release of the client not compatible with older ones? | 17:31 |
mriedem | smcginnis: we use semver for the client | 17:31 |
mriedem | so if we make a backward incompatible change, it's a major version bump | 17:31 |
openstackgerrit | Paul Carlton proposed openstack/nova-specs: Automatic Live Migration Completion https://review.openstack.org/306561 | 17:32 |
mriedem | the stable branch is really just for bug fixes on n-1 | 17:32 |
smcginnis | mriedem: OK, so you've had to do that for past releases, therefore need stable releases of the client for older ones? | 17:32 |
mriedem | yeah | 17:32 |
mriedem | smcginnis: e.g. https://github.com/openstack/python-novaclient/commit/bc213fd8347f20df3534f93e992e9325adaa2a63 | 17:32 |
*** ctrath has quit IRC | 17:32 | |
smcginnis | mriedem: OK, makes sense. We just always move forward with python-cinderclient, so trying to understand why it was different. | 17:32 |
*** paul-carlton1 has quit IRC | 17:33 | |
*** sambetts is now known as sambetts|afk | 17:33 | |
*** shaohe_feng has quit IRC | 17:33 | |
*** rooooook has quit IRC | 17:34 | |
*** shaohe_feng has joined #openstack-nova | 17:34 | |
*** rooooook has joined #openstack-nova | 17:34 | |
*** thorst_ has quit IRC | 17:34 | |
*** sudipto has quit IRC | 17:35 | |
*** ctrath has joined #openstack-nova | 17:35 | |
*** hieulq_ has quit IRC | 17:36 | |
*** hshiina has quit IRC | 17:36 | |
*** sudipto has joined #openstack-nova | 17:41 | |
*** thorst_ has joined #openstack-nova | 17:43 | |
*** mriedem1 has joined #openstack-nova | 17:43 | |
*** mriedem has quit IRC | 17:43 | |
*** shaohe_feng has quit IRC | 17:44 | |
*** shaohe_feng has joined #openstack-nova | 17:44 | |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: move s3 related options https://review.openstack.org/314146 | 17:45 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: deprecate s3 image service config options https://review.openstack.org/314697 | 17:45 |
*** neophy has joined #openstack-nova | 17:46 | |
*** whenry has quit IRC | 17:46 | |
*** mpavone has joined #openstack-nova | 17:46 | |
*** mpavone has quit IRC | 17:47 | |
*** thorst_ has quit IRC | 17:47 | |
*** jichen has joined #openstack-nova | 17:48 | |
*** ccarmack has quit IRC | 17:49 | |
*** lucasagomes is now known as lucas-afk | 17:49 | |
*** sudipto has quit IRC | 17:49 | |
*** gjayavelu has joined #openstack-nova | 17:50 | |
*** rlrossit has quit IRC | 17:50 | |
*** gjayavelu has quit IRC | 17:51 | |
*** markus_z has quit IRC | 17:52 | |
*** shaohe_feng has quit IRC | 17:54 | |
*** shaohe_feng has joined #openstack-nova | 17:54 | |
*** andymaier_ has quit IRC | 17:55 | |
*** lucas-afk is now known as lucasagomes | 17:55 | |
*** rlrossit has joined #openstack-nova | 17:55 | |
openstackgerrit | Robert Ellis proposed openstack/nova: Changed 202 for lock/unlock return code to 200. This addresses bug 1449811. 409 return code was not addressed as the compute node is already in that state so no corrective action is required. 404 is already covered by existing code. https://review.openstack.org/314267 | 17:56 |
openstack | bug 1449811 in OpenStack Compute (nova) "nova lock/unlock api return code is not accurate" [Wishlist,Confirmed] https://launchpad.net/bugs/1449811 - Assigned to Robert Ellis (bob-ellis42) | 17:56 |
*** lucasagomes is now known as lucas-dinner | 17:57 | |
*** Sukhdev_ has joined #openstack-nova | 17:57 | |
*** vilobhmm11 has quit IRC | 17:59 | |
*** rmart04 has joined #openstack-nova | 17:59 | |
*** Sukhdev has quit IRC | 18:00 | |
*** whenry has joined #openstack-nova | 18:01 | |
*** Sukhdev_ has quit IRC | 18:02 | |
*** thorst_ has joined #openstack-nova | 18:02 | |
*** shaohe_feng has quit IRC | 18:04 | |
*** shaohe_feng has joined #openstack-nova | 18:04 | |
*** zengine has quit IRC | 18:06 | |
*** chuck__ has joined #openstack-nova | 18:06 | |
*** zul has quit IRC | 18:07 | |
*** tonytan4ever has joined #openstack-nova | 18:07 | |
*** chuck__ is now known as zul | 18:07 | |
dansmith | mriedem1: you can drop your -2 on this now: https://review.openstack.org/#/c/312750/ | 18:08 |
dansmith | I don't want to start on any more migrations until that merges and gets the migration number reserved | 18:08 |
*** mriedem1 is now known as mriedem | 18:09 | |
mriedem | dansmith: is the spec approved now? | 18:09 |
mriedem | yes it is https://review.openstack.org/#/c/312619/ | 18:09 |
mriedem | well hells bells | 18:10 |
mriedem | bp is approved, -2 removed | 18:11 |
mriedem | peace be with you | 18:11 |
mriedem | dansmith: what migration number? | 18:12 |
dansmith | mriedem: the db migration number | 18:13 |
mriedem | that is either later in the series or doesn't exist yet | 18:13 |
dansmith | it's in the series | 18:13 |
*** rcernin has quit IRC | 18:13 | |
dansmith | https://review.openstack.org/#/c/312755/8 | 18:14 |
*** shaohe_feng has quit IRC | 18:14 | |
*** shaohe_feng has joined #openstack-nova | 18:15 | |
*** whenry has quit IRC | 18:17 | |
*** mvk has joined #openstack-nova | 18:17 | |
*** pushkaru has quit IRC | 18:18 | |
mriedem | the fernet test races must be extra bad in the cells v1 job | 18:21 |
* tbachman tries to parse that sentence | 18:22 | |
tbachman | mriedem: btw, I haven’t forgotten that I have a tempest test obligation | 18:23 |
tbachman | haven’t gotten to it yet, but hopefully mid-week | 18:23 |
mriedem | for shame | 18:23 |
*** shaohe_feng has quit IRC | 18:25 | |
*** shaohe_feng has joined #openstack-nova | 18:25 | |
*** thumpba has joined #openstack-nova | 18:25 | |
*** haplo37 has joined #openstack-nova | 18:26 | |
*** cdent has quit IRC | 18:28 | |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 18:30 |
auggy | sdague: jichen: I have a question about https://review.openstack.org/#/c/314320/2/api-ref/source/ips.inc | 18:30 |
*** lpetrut has quit IRC | 18:31 | |
*** sacharya has quit IRC | 18:31 | |
jichen | auggy: yes | 18:31 |
auggy | sdague: jichen: i am not sure how to represent that "addr" field | 18:31 |
auggy | should I add a new parameter? | 18:31 |
auggy | i realized the definition for "address" is "floating ip address" | 18:31 |
*** baoli_ has quit IRC | 18:32 | |
*** Daisy has joined #openstack-nova | 18:32 | |
auggy | is that what we want that definition to be, should we always assume anything that uses "address" is always going to use a floating ip? | 18:32 |
jichen | yes, I think so, at least ,the floating ip looks me is incorrect, my guess is to add a param in parameters.yaml | 18:32 |
jichen | auggy: my understanding is we need to distinguish them ,for example . you have a lot of 'ids' but server_id , user_id is different, we need separated them | 18:33 |
auggy | jichen: sdague: should we consider updating "address" somehow to distinguish between generic, static, and floating? | 18:33 |
auggy | it just seems potentially ambiguous right now | 18:34 |
*** gyee has joined #openstack-nova | 18:34 | |
auggy | jichen: sdague: in the meantime, should I just add "ip_address" as a new parameter and use that instead? | 18:34 |
*** lpetrut has joined #openstack-nova | 18:35 | |
*** shaohe_feng has quit IRC | 18:35 | |
jichen | auggy: um... I am not very sure, maybe sdague has more input, my guess is w probably need distinguish them , and I think ip_address works for me | 18:35 |
*** shaohe_feng has joined #openstack-nova | 18:36 | |
*** voelzmo has joined #openstack-nova | 18:36 | |
auggy | jichen: ok cool, i'll just update the patch with that change and a comment | 18:36 |
voelzmo | awesome, thanks @coreywright | 18:36 |
auggy | thanks ;) | 18:36 |
*** Daisy has quit IRC | 18:36 | |
jichen | auggy: thanks :) | 18:37 |
*** cdent has joined #openstack-nova | 18:38 | |
auggy | jichen: looks like there's already a param called "ip"! | 18:38 |
jichen | auggy: then maybe you can reuse it , if the meaning is same , I think we reuse them sometime | 18:39 |
auggy | jichen: do you know if you can have a parameter named the same for both body and path? | 18:39 |
*** haplo37 has quit IRC | 18:40 | |
auggy | the ip param i found is "in: path", so can i create a second one with "in: body"? | 18:40 |
jichen | auggy: I don't know ,I think we need use in:path and in: body both as sdague pointed out in the same patch | 18:41 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/313743 | 18:41 |
*** kfarr has quit IRC | 18:41 | |
*** kfarr has joined #openstack-nova | 18:41 | |
*** gagehugo has quit IRC | 18:41 | |
*** gagehugo has joined #openstack-nova | 18:41 | |
*** penick has quit IRC | 18:41 | |
*** penick has joined #openstack-nova | 18:41 | |
*** ctrath has quit IRC | 18:41 | |
*** ctrath has joined #openstack-nova | 18:41 | |
*** mriedem has quit IRC | 18:42 | |
*** mriedem has joined #openstack-nova | 18:42 | |
*** rlrossit has quit IRC | 18:42 | |
*** rlrossit has joined #openstack-nova | 18:42 | |
*** shaohe_feng has quit IRC | 18:42 | |
*** shaohe_feng has joined #openstack-nova | 18:42 | |
auggy | jichen: yeah, i am looking at parameters.yaml to see if there are any with the same name that have both in: body and in: path | 18:42 |
auggy | but so far everything has a unique name | 18:42 |
*** rcernin has joined #openstack-nova | 18:43 | |
jichen | auggy: ok, If I need a name, I typical created it :) | 18:44 |
*** thumpba has quit IRC | 18:44 | |
*** shaohe_feng has quit IRC | 18:45 | |
auggy | jichen: ah i see, i guess i would rather see if one exists i can use already :) | 18:45 |
*** thumpba has joined #openstack-nova | 18:46 | |
*** shaohe_feng has joined #openstack-nova | 18:46 | |
openstackgerrit | Sujitha proposed openstack/nova: Added new exception to handle CinderClientException https://review.openstack.org/290156 | 18:46 |
*** thorst_ has quit IRC | 18:48 | |
belliott | dansmith: so some tests failed, i assumed this is what your keystone recheck was supposed to fix? -- https://review.openstack.org/#/c/314230/ | 18:49 |
*** thorst_ has joined #openstack-nova | 18:49 | |
openstackgerrit | sean mooney proposed openstack/os-vif: Add support for vhost-user https://review.openstack.org/285613 | 18:49 |
mriedem | belliott: recheck again | 18:49 |
dansmith | belliott: aye | 18:49 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/313743 | 18:50 |
belliott | mriedem: wasn't sure what all the recheck commands are anymore | 18:51 |
mriedem | belliott: just recheck | 18:51 |
belliott | mriedem: :) | 18:51 |
*** zenoway has joined #openstack-nova | 18:51 | |
openstackgerrit | Augustina Ragwitz proposed openstack/nova: api-ref: ips.inc parameter verification https://review.openstack.org/314320 | 18:51 |
*** zenoway has quit IRC | 18:52 | |
*** zenoway has joined #openstack-nova | 18:52 | |
*** sdake has quit IRC | 18:52 | |
*** pushkaru has joined #openstack-nova | 18:53 | |
*** shaohe_feng has quit IRC | 18:55 | |
*** shaohe_feng has joined #openstack-nova | 18:56 | |
openstackgerrit | Dan Smith proposed openstack/nova: Create Instances with keypairs https://review.openstack.org/312751 | 18:57 |
openstackgerrit | Dan Smith proposed openstack/nova: Add keypairs to Instance object https://review.openstack.org/312750 | 18:57 |
openstackgerrit | Dan Smith proposed openstack/nova: Add Keypairs to the API database https://review.openstack.org/312755 | 18:57 |
openstackgerrit | Dan Smith proposed openstack/nova: Make Keypair object favor the API database https://review.openstack.org/312756 | 18:57 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to move keypairs from main to API database https://review.openstack.org/313664 | 18:57 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to store keypairs with instances https://review.openstack.org/313060 | 18:57 |
*** thorst_ has quit IRC | 18:57 | |
sdague | auggy: they all have to have unique names | 19:00 |
sdague | typically I've been appending _path or _query where appropriate | 19:00 |
*** xenogear has quit IRC | 19:00 | |
*** salv-orl_ has joined #openstack-nova | 19:02 | |
*** xyang1 has joined #openstack-nova | 19:02 | |
*** fragatina has quit IRC | 19:02 | |
*** thorst_ has joined #openstack-nova | 19:03 | |
*** eglynn has quit IRC | 19:03 | |
melwitt | mriedem: do you know if elastic-recheck is still working? I feel like I haven't seen it comment on reviews in a long time | 19:04 |
*** rmart04 has quit IRC | 19:04 | |
mriedem | melwitt: i haven't seen comments either | 19:04 |
*** salv-orlando has quit IRC | 19:05 | |
sdague | mriedem: failopotamus patch merged | 19:05 |
mriedem | sdague: yes! | 19:05 |
mriedem | doffm: ^ is what i was giggling about | 19:05 |
*** shaohe_feng has quit IRC | 19:06 | |
*** shaohe_feng has joined #openstack-nova | 19:07 | |
mriedem | dansmith: i'm blowing you away in commits here https://github.com/sdague/failopotamus/graphs/contributors | 19:07 |
mriedem | :) | 19:07 |
dansmith | mriedem: um, congrats? | 19:07 |
*** tbachman has quit IRC | 19:08 | |
*** thorst_ has quit IRC | 19:08 | |
*** dedery has joined #openstack-nova | 19:08 | |
openstackgerrit | Vilobh Meshram proposed openstack/nova-specs: Flavor Classes https://review.openstack.org/314731 | 19:09 |
*** thorst_ has joined #openstack-nova | 19:09 | |
*** fragatina has joined #openstack-nova | 19:09 | |
melwitt | so that's how you make the cool failure graph | 19:11 |
*** pcaruana has quit IRC | 19:12 | |
*** thorst_ has quit IRC | 19:14 | |
*** thorst_ has joined #openstack-nova | 19:15 | |
*** shaohe_feng has quit IRC | 19:16 | |
*** shaohe_feng has joined #openstack-nova | 19:19 | |
*** thorst_ has quit IRC | 19:20 | |
openstackgerrit | Robert Ellis proposed openstack/nova: Changed 202 for lock/unlock return code to 200. This addresses bug 1449811. 409 return code was not addressed as the compute node is already in that state so no corrective action is required. 404 is already covered by existing code. https://review.openstack.org/314267 | 19:21 |
openstack | bug 1449811 in OpenStack Compute (nova) "nova lock/unlock api return code is not accurate" [Wishlist,Confirmed] https://launchpad.net/bugs/1449811 - Assigned to Robert Ellis (bob-ellis42) | 19:21 |
*** thorst_ has joined #openstack-nova | 19:22 | |
*** sacharya has joined #openstack-nova | 19:22 | |
*** thorst_ has quit IRC | 19:23 | |
*** thorst_ has joined #openstack-nova | 19:23 | |
mriedem | dansmith: on https://review.openstack.org/#/c/314230/ so why weren't we doing this before? just didn't think of it? | 19:24 |
*** rmart04 has joined #openstack-nova | 19:24 | |
*** neophy has quit IRC | 19:24 | |
*** shaohe_feng has quit IRC | 19:26 | |
*** shaohe_feng has joined #openstack-nova | 19:27 | |
*** thorst_ has quit IRC | 19:30 | |
*** neophy has joined #openstack-nova | 19:30 | |
*** thorst_ has joined #openstack-nova | 19:30 | |
*** rmart04 has quit IRC | 19:32 | |
*** rmart04_ has joined #openstack-nova | 19:32 | |
*** tbachman has joined #openstack-nova | 19:32 | |
*** gszasz has quit IRC | 19:34 | |
*** thorst_ has quit IRC | 19:36 | |
*** voelzmo has quit IRC | 19:36 | |
*** shaohe_feng has quit IRC | 19:36 | |
*** dedery_ has joined #openstack-nova | 19:37 | |
*** shaohe_feng has joined #openstack-nova | 19:37 | |
*** thorst_ has joined #openstack-nova | 19:37 | |
*** dedery__ has joined #openstack-nova | 19:38 | |
cdent | mriedem: you ate an 80 year old man's spirit? what kind of crazy witch dr are you? | 19:38 |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 19:39 |
mriedem | cdent: i wanted his life force, or what was left of it | 19:39 |
*** dedery has quit IRC | 19:39 | |
openstackgerrit | Sean Dague proposed openstack/nova: Denormalize personality extension https://review.openstack.org/314600 | 19:40 |
cdent | I question your choices, sir. | 19:40 |
cdent | "Denormalize personality extension" <- best commit summary ever | 19:40 |
*** rmart04 has joined #openstack-nova | 19:40 | |
*** rmart04_ has quit IRC | 19:41 | |
*** dedery_ has quit IRC | 19:42 | |
*** thorst_ has quit IRC | 19:42 | |
*** shaohe_feng has quit IRC | 19:47 | |
sdague | :) | 19:47 |
*** shaohe_feng has joined #openstack-nova | 19:47 | |
*** rmart04_ has joined #openstack-nova | 19:48 | |
*** rmart04 has quit IRC | 19:49 | |
*** rmart04_ is now known as rmart04 | 19:49 | |
*** rcernin has quit IRC | 19:51 | |
*** daemontool_ has joined #openstack-nova | 19:51 | |
*** tongli has quit IRC | 19:53 | |
*** rmart04 has quit IRC | 19:55 | |
mriedem | neutron + libvirt + lvm https://review.openstack.org/#/c/314744/ | 19:56 |
mriedem | what could go wrong?! | 19:56 |
*** shaohe_feng has quit IRC | 19:57 | |
*** lpetrut has quit IRC | 19:57 | |
*** shaohe_feng has joined #openstack-nova | 19:57 | |
*** rcernin has joined #openstack-nova | 19:59 | |
*** rmart04 has joined #openstack-nova | 19:59 | |
kashyap | mriedem: Looks fairly straight-forward, no? :-) | 20:00 |
kashyap | The earth continues to spin | 20:01 |
openstackgerrit | Dan Smith proposed openstack/nova: Create Instances with keypairs https://review.openstack.org/312751 | 20:02 |
openstackgerrit | Dan Smith proposed openstack/nova: Add Keypairs to the API database https://review.openstack.org/312755 | 20:02 |
openstackgerrit | Dan Smith proposed openstack/nova: Make Keypair object favor the API database https://review.openstack.org/312756 | 20:02 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to move keypairs from main to API database https://review.openstack.org/313664 | 20:02 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to store keypairs with instances https://review.openstack.org/313060 | 20:02 |
dansmith | mriedem: jaypipes: don't let those highly stubbed tests in test_compute_api fool you.. they're not testing any of this.. there's lots of coverage of this path, especially in functional (and in tempest), as evidenced by the test failures that just pointed out I made a mind-o in my tweak for alaski ^^ :) | 20:02 |
dansmith | mriedem: jaypipes: i.e.: http://logs.openstack.org/51/312751/7/check/gate-nova-tox-db-functional/3f5230d/testr_results.html.gz | 20:03 |
scottda | Does anyone know why adding a new keyword arg to cinderclient Client init would cause nova be so unhappy? https://bugs.launchpad.net/nova/+bug/1580319 | 20:03 |
openstack | Launchpad bug 1580319 in OpenStack Compute (nova) "unreleased cinderclient fails volume-attach due to new Client parameter" [Undecided,New] | 20:03 |
*** rmart04 has quit IRC | 20:04 | |
auggy | sdague: cool thanks | 20:05 |
*** rmart04 has joined #openstack-nova | 20:05 | |
mriedem | scottda: well you have the stacktrace there | 20:05 |
auggy | sdague: cdent if we denormalize our personality extensions, does that mean we all become cybermen? | 20:05 |
*** daemontool_ has quit IRC | 20:06 | |
scottda | mriedem: Yes, I saw that. But I'm not sure where that decorator in /usr/local/lib/python2.7/dist-packages/positional/ is even coming from. Just wondering if anyone had seen this kind of thing before... | 20:06 |
cdent | WE ARE DEVO | 20:06 |
*** rmart04_ has joined #openstack-nova | 20:07 | |
jaypipes | dansmith: k | 20:07 |
openstackgerrit | Pushkar Umaranikar proposed openstack/nova: method verification for servers-action-fixed-ip https://review.openstack.org/314748 | 20:07 |
*** shaohe_feng has quit IRC | 20:07 | |
mriedem | scottda: sudo cp -rs /opt/stack/python-cinderclient/cinderclient /urr/local/lib/python2.7/dist-packages/ | 20:07 |
mriedem | you have a typo in there | 20:07 |
mriedem | /urr/local | 20:07 |
*** shaohe_feng has joined #openstack-nova | 20:08 | |
*** akshai has quit IRC | 20:08 | |
mriedem | also, i'd just pip install the cloned python-cinderclient | 20:08 |
mriedem | rather than copy things around | 20:08 |
kashyap | mriedem: dansmith: When you get a moment, mind taking a look at this (you both have more than good context) liberty backport: https://review.openstack.org/#/c/314012/4 | 20:09 |
scottda | mriedem: Thanks, I'll fix that in the bug report. But I just typed that in. The bug still exists . | 20:09 |
*** rmart04 has quit IRC | 20:09 | |
*** rmart04_ is now known as rmart04 | 20:09 | |
mriedem | scottda: riddle me this, does python-cinderclient go through functional dsvm testing? | 20:09 |
*** dharinic has joined #openstack-nova | 20:10 | |
mriedem | scottda: http://logs.openstack.org/41/301941/15/check/gate-cinderclient-dsvm-functional/4b56a7f/logs/testr_results.html | 20:10 |
mriedem | it just installs, it doesn't run tests | 20:10 |
mriedem | looks like http://logs.openstack.org/41/301941/15/gate/gate-tempest-dsvm-neutron-src-python-cinderclient/5e5e21a/logs/testr_results.html.gz tests some tihngs | 20:10 |
scottda | hmm | 20:11 |
scottda | mriedem: This looks like functional is running: http://logs.openstack.org/27/303627/9/check/gate-cinderclient-dsvm-functional/8c6916c/console.html.gz | 20:11 |
mriedem | scottda: is this bug report against a nova change you're working on to use api_version with cinderclient? | 20:12 |
*** haplo37 has joined #openstack-nova | 20:12 | |
scottda | mriedem: No, this is cinderclient changes to make the cinderclient aware of api_versions, and they are merged into master | 20:12 |
mriedem | ah yes http://logs.openstack.org/27/303627/9/check/gate-cinderclient-dsvm-functional/8c6916c/logs/testr_results.html.gz | 20:12 |
*** sacharya_ has joined #openstack-nova | 20:13 | |
*** rmart04_ has joined #openstack-nova | 20:14 | |
*** rmart04 has quit IRC | 20:14 | |
*** rmart04_ is now known as rmart04 | 20:14 | |
*** winston-d_ has quit IRC | 20:15 | |
*** sacharya has quit IRC | 20:16 | |
scottda | mriedem: So, there's obviously a testing gap. I don't think there's anything that tests Nova with latest cinderclient on master, just the latest released version. | 20:16 |
*** rama_y has quit IRC | 20:17 | |
*** shaohe_feng has quit IRC | 20:17 | |
mriedem | scottda: well, http://logs.openstack.org/41/301941/15/check/gate-tempest-dsvm-neutron-src-python-cinderclient/b6f11fc/ would be running tempest against latest nova + this commit in cinderclient | 20:17 |
mriedem | i see the commit is in pip freeze here http://logs.openstack.org/41/301941/15/check/gate-tempest-dsvm-neutron-src-python-cinderclient/b6f11fc/logs/pip-freeze.txt.gz | 20:18 |
mriedem | scottda: stale pyc's in your local devstack version? | 20:18 |
*** shaohe_feng has joined #openstack-nova | 20:18 | |
scottda | mriedem: Maybe | 20:19 |
mriedem | anyway, i wouldn't mess with copying files around, i'd pip install that bad mama jama locally | 20:19 |
scottda | mriedem: I'll try that. In the meantime, I'm just a bit paranoid about this one. | 20:19 |
dansmith | mriedem: um, isn't it "mamba jamba" ? | 20:19 |
mriedem | mamba jamba? | 20:19 |
mriedem | no | 20:19 |
mriedem | https://www.youtube.com/watch?v=qop26t6j2BE | 20:19 |
dansmith | http://www.urbandictionary.com/define.php?term=Mamba%20Jamba | 20:20 |
*** ccarmack has joined #openstack-nova | 20:20 | |
mriedem | i can't tell if this guy is serious or not | 20:20 |
*** rama_y has joined #openstack-nova | 20:21 | |
mriedem | here we go https://www.youtube.com/watch?v=6QIw1BQIvT4 | 20:21 |
mriedem | lbragstad: ^ i think you have a long lost brother | 20:21 |
openstackgerrit | jichenjc proposed openstack/nova: WIP: api-ref keypair update https://review.openstack.org/314502 | 20:21 |
raj_singh | sdague: I am working on image.inc file and saw some discrepancies in "normal response code" for get methods mentioned in file and one in api doc. Api doc says 200,203 while inc file has just 200. | 20:22 |
raj_singh | http://developer.openstack.org/api-ref-compute-v2.1.html | 20:22 |
sdague | raj_singh: 203 is never something we return | 20:22 |
sdague | I think it's something in the rax proxy layer | 20:22 |
raj_singh | sdague: so need to fix this in docs? | 20:23 |
sdague | yes | 20:23 |
*** rmart04 has quit IRC | 20:23 | |
raj_singh | thx | 20:23 |
*** abitha has quit IRC | 20:26 | |
openstackgerrit | Dan Smith proposed openstack/nova: Add Keypairs to the API database https://review.openstack.org/312755 | 20:26 |
openstackgerrit | Dan Smith proposed openstack/nova: Make Keypair object favor the API database https://review.openstack.org/312756 | 20:26 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to move keypairs from main to API database https://review.openstack.org/313664 | 20:26 |
openstackgerrit | Dan Smith proposed openstack/nova: Add online migration to store keypairs with instances https://review.openstack.org/313060 | 20:26 |
*** fragatina has quit IRC | 20:27 | |
*** shaohe_feng has quit IRC | 20:28 | |
*** ijw_ has joined #openstack-nova | 20:28 | |
*** ijw has quit IRC | 20:28 | |
*** shaohe_feng has joined #openstack-nova | 20:29 | |
*** thorst_ has joined #openstack-nova | 20:32 | |
*** baoli has joined #openstack-nova | 20:36 | |
*** ccarmack has quit IRC | 20:38 | |
*** shaohe_feng has quit IRC | 20:38 | |
*** shaohe_feng has joined #openstack-nova | 20:39 | |
lbragstad | mriedem that link makes my ears burn | 20:40 |
mriedem | lbragstad: you have an uncanny resemblance | 20:41 |
mriedem | to carl carlton | 20:41 |
*** baoli has quit IRC | 20:41 | |
*** baoli has joined #openstack-nova | 20:42 | |
openstackgerrit | jichenjc proposed openstack/nova: api-ref keypair update (os-keypairs.inc) https://review.openstack.org/314502 | 20:43 |
openstackgerrit | Brian Elliott proposed openstack/nova: Fix service version to update the DB https://review.openstack.org/314230 | 20:43 |
*** ccarmack has joined #openstack-nova | 20:45 | |
mikal | mriedem: I'll take a look at that spec post coffee | 20:45 |
mikal | Also, good morning | 20:46 |
*** jichen has quit IRC | 20:47 | |
mriedem | mikal: hi | 20:47 |
*** eharney has quit IRC | 20:47 | |
*** markstur has joined #openstack-nova | 20:48 | |
mriedem | mikal: i thought of you because of the vendor data thing | 20:48 |
*** shaohe_feng has quit IRC | 20:48 | |
*** baoli has quit IRC | 20:48 | |
mikal | Yep, makes sense to me | 20:49 |
*** baoli has joined #openstack-nova | 20:49 | |
*** baoli has quit IRC | 20:50 | |
markstur | mriedem, Suggestions for " | 20:50 |
markstur | Add instance host to evacuate terminate connector" https://review.openstack.org/#/c/274318 ??? | 20:50 |
*** shaohe_feng has joined #openstack-nova | 20:51 | |
markstur | mriedem, I know it is a hack that no one likes, but now should I abandon it or push for it as the only fix we have for attachments that existed before the other evacuate patch. | 20:51 |
*** lucas-dinner has quit IRC | 20:51 | |
*** baoli has joined #openstack-nova | 20:52 | |
markstur | mriedem, Unfortunately this one would need to land and backport and also get more Cinder-side fixes, too. | 20:52 |
cdent | jaypipes, dansmith: did you guys figure out the resource-related database changes or otherwise increase the degree of plan? | 20:53 |
*** baoli has quit IRC | 20:57 | |
*** lucasagomes has joined #openstack-nova | 20:58 | |
*** shaohe_feng has quit IRC | 20:58 | |
*** shaohe_feng has joined #openstack-nova | 20:59 | |
*** baoli has joined #openstack-nova | 21:00 | |
mriedem | markstur: i'm not crazy about adding random things to the connector dict that nova passes back to cinder, since that's really an api change, even though old cinder code won't know about it anyway so won't check for it, and new cinder code would have to know that it might not be there (if they don't have new enough nova to be sendingit) | 21:00 |
mriedem | and it really requires a workaround in each cinder volume driver | 21:01 |
*** baoli_ has joined #openstack-nova | 21:01 | |
mriedem | markstur: hemna and scottda have been looking at new APIs in cinder to handle cinder storing connector state in it's db, maybe this could fall into that somewhere | 21:02 |
*** thumpba has quit IRC | 21:02 | |
markstur | mriedem, Agreed. So I'm thinking abandon it. Unfortunately there isn't any better proposal for backports. It should be fixed in Newton with new API proposals. | 21:02 |
mriedem | although a new api in cinder wouldn't be backported to stable | 21:02 |
markstur | mriedem, Yep. That's my dilemma. So we can backport the hack and I can try to get Cinder folks to use it for backports... but moving forward everyone agrees there should be a better fix. | 21:03 |
*** aysyd has quit IRC | 21:03 | |
*** baoli has quit IRC | 21:04 | |
markstur | mriedem, or just abandon it and downstream can do the hacks instead of upstream | 21:05 |
*** rama_y has quit IRC | 21:05 | |
*** rama_y has joined #openstack-nova | 21:06 | |
jaypipes | cdent: yes. plan is this (copy/pasted from convo with dansmith): | 21:07 |
dansmith | dude | 21:07 |
dansmith | private logs man! | 21:07 |
*** abalutoiu has quit IRC | 21:07 | |
*** thorst_ has quit IRC | 21:08 | |
jaypipes | dansmith: there's no private content in my paste :) | 21:08 |
*** abalutoiu has joined #openstack-nova | 21:08 | |
jaypipes | 05-10 13:33:28 <dansmith> 1. Add the table to the api db | 21:08 |
jaypipes | 05-10 13:33:44 <dansmith> 2. make the inventory object write there instead | 21:08 |
jaypipes | 05-10 13:33:51 <dansmith> 3. make the compute node not update if they're zero, | 21:08 |
jaypipes | 05-10 13:34:22 <dansmith> then I think we don't need to do any major migration since we'll have a fresh api db set of inventories within a | 21:08 |
jaypipes | minute or two of restart | 21:08 |
jaypipes | 05-10 13:34:38 <dansmith> and then I'll let you refactor the compute code to actually do the inventory management itself | 21:08 |
*** thorst_ has joined #openstack-nova | 21:09 | |
*** shaohe_feng has quit IRC | 21:09 | |
cdent | jaypipes: is that EOP? | 21:09 |
jaypipes | cdent: yes. and I've been trying to unmess some of the resource tracker code this afternoon in prep for that. | 21:11 |
cdent | jaypipes: is there a plan, or does there need to be a plan, for moving resource_providers table to api db? | 21:11 |
*** rama_y has quit IRC | 21:11 | |
jaypipes | cdent: excellent question. dansmith? | 21:12 |
*** shaohe_feng has joined #openstack-nova | 21:12 | |
*** xyang1 has quit IRC | 21:12 | |
mriedem | heh, oh boy http://logs.openstack.org/44/314744/3/check/gate-tempest-dsvm-postgres-full/2098f4b/logs/screen-n-cpu.txt.gz?level=TRACE | 21:12 |
cdent | earlier mriedem and I thought "yes" because that's where aggregates be, and we have resource_provider_aggregates table | 21:12 |
jaypipes | cdent: right. | 21:12 |
jaypipes | cdent: there's just the small question of who gets to do that :) | 21:13 |
mikal | What is people's feeling about copyright statements in the nova/conf directory? A lot of the refactorings have been stripping peoples copyright statements, which makes me very uncomfortable. | 21:13 |
dansmith | the only providers we have in there now are the compute nodes, right? so the assumption is that in the process above, the table goes at the same time and we just do the same auto-create migration step again, but there | 21:13 |
mikal | There's at least two patches held up on that. | 21:13 |
openstackgerrit | Julian proposed openstack/nova: Add unit tests for nova.virt.firewall.IpTablesFirewallDriver https://review.openstack.org/311758 | 21:13 |
openstackgerrit | Clenimar Filemon proposed openstack/nova: Use Identity v3 params when connecting to Ironic https://review.openstack.org/300154 | 21:13 |
jaypipes | mikal: I've bitched at folks who *add* their own copyright when just moving code around. | 21:13 |
*** unicell has quit IRC | 21:14 | |
mikal | jaypipes: you +2'ed the one I am looking at at the moment, which effectively does that | 21:14 |
mikal | jaypipes: removes someone elses, and adds Copyright 2016 OpenStack Foundation | 21:14 |
mikal | jaypipes: which makes me uncomfortable | 21:14 |
jaypipes | mikal: I don't consider "Copyright 2016 OpenStack Foundation" to be their *own* copyright. | 21:15 |
mikal | jaypipes: well, maybe. Its probably invalid too, you can't gift copyright like that. | 21:15 |
mikal | jaypipes: either way, it removed the original author's copyright statement | 21:15 |
mikal | I think its a nit, but its something which bothers me | 21:16 |
mikal | I'm gonna upload a fix and then +W this thing | 21:16 |
jaypipes | go for ti. | 21:16 |
jaypipes | it | 21:16 |
openstackgerrit | Michael Still proposed openstack/nova: config options: move s3 related options https://review.openstack.org/314146 | 21:16 |
mikal | Cause no one has shouted at me in the last two minutes and therefore no one else cares | 21:16 |
*** ametts has quit IRC | 21:17 | |
*** thorst_ has quit IRC | 21:17 | |
hemna | markstur, mriedem if our work ends up getting in, it will solve the evacuate and forced detach issues | 21:18 |
hemna | because Cinder will have the connector | 21:18 |
*** shaohe_feng has quit IRC | 21:19 | |
openstackgerrit | Karen Bradshaw proposed openstack/nova: [WIP] api-ref, evacuate action https://review.openstack.org/314776 | 21:19 |
*** shaohe_feng has joined #openstack-nova | 21:20 | |
*** rtheis has quit IRC | 21:20 | |
*** unicell has joined #openstack-nova | 21:20 | |
*** unicell has quit IRC | 21:20 | |
*** unicell has joined #openstack-nova | 21:20 | |
mriedem | oh fun http://logs.openstack.org/44/314744/3/check/gate-tempest-dsvm-postgres-full/2098f4b/logs/screen-n-cpu.txt.gz?level=TRACE#_2016-05-10_20_51_24_982 | 21:23 |
*** yamahata has quit IRC | 21:23 | |
mriedem | so i have an expected exception for rpc, but oslo.messaging logs the error anyway | 21:23 |
*** smatzek has quit IRC | 21:23 | |
markstur | hemna, Yes for Newton that's the plan. Unfortunately that won't be backported. So the hack for backports is still under consideration -- to backport or to abandon. | 21:23 |
hemna | :( | 21:24 |
* hemna is not a fan of the hack at all | 21:24 | |
mriedem | dims: i'm having trouble finding how nova even gets here http://logs.openstack.org/44/314744/3/check/gate-tempest-dsvm-postgres-full/2098f4b/logs/screen-n-cpu.txt.gz?level=TRACE#_2016-05-10_20_51_24_982 | 21:26 |
mriedem | dims: since it looks like that comes from https://github.com/openstack/oslo.messaging/blob/master/oslo_messaging/_drivers/common.py#L174 which i don't see nova using | 21:27 |
mriedem | or rabbit in oslo.messaging using | 21:27 |
* dims peeks | 21:27 | |
*** ivase has quit IRC | 21:28 | |
*** rlrossit has quit IRC | 21:28 | |
mriedem | err i guess we're using oslo_messaging._drivers.amqpdriver now? | 21:29 |
*** yamahata has joined #openstack-nova | 21:29 | |
*** shaohe_feng has quit IRC | 21:29 | |
mriedem | http://logs.openstack.org/44/314744/3/check/gate-tempest-dsvm-postgres-full/2098f4b/logs/screen-n-cpu.txt.gz#_2016-05-10_20_29_40_344' | 21:30 |
dims | mriedem : http://codesearch.openstack.org/?q=%5C.serialize_remote_exception&i=nope&files=&repos=oslo.messaging first hit line 59 | 21:31 |
mriedem | yeah but i'm wondering what is oslo_messaging/_drivers/amqpdriver.py being used for? | 21:32 |
*** ivase has joined #openstack-nova | 21:32 | |
*** lpetrut has joined #openstack-nova | 21:33 | |
dims | hang on | 21:33 |
mriedem | ah class RabbitDriver(amqpdriver.AMQPDriverBase): | 21:34 |
*** jamesden_ has quit IRC | 21:34 | |
dims | mriedem : if you see....yep | 21:34 |
*** shaohe_feng has joined #openstack-nova | 21:34 | |
*** daemontool_ has joined #openstack-nova | 21:35 | |
dims | mriedem : what's the concern, garbage in logs? | 21:35 |
mriedem | yes | 21:35 |
mriedem | nova has it as an expected exception but it's logged as an error | 21:35 |
*** ivase has quit IRC | 21:36 | |
openstackgerrit | Mark Doffman proposed openstack/nova: Make Aggregate metadata functions work with API db https://review.openstack.org/297350 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Add aggregates tables to the API db. https://review.openstack.org/294806 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Make Aggregate.save work with the API db https://review.openstack.org/305466 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Make Aggregate host operations work against API db https://review.openstack.org/296597 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Make AggregateList.get_ return API & cell db items https://review.openstack.org/296062 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Make Aggregate.get operation favor the API db https://review.openstack.org/296014 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Aggregate create and destroy work against API db https://review.openstack.org/314781 | 21:37 |
openstackgerrit | Mark Doffman proposed openstack/nova: Add data migration methods for Aggregate. https://review.openstack.org/314782 | 21:37 |
*** thorst_ has joined #openstack-nova | 21:38 | |
dims | mriedem : we'll need sileht to peel this one, oslo.messaging bug please in launchpad? | 21:39 |
mriedem | sure | 21:39 |
*** shaohe_feng has quit IRC | 21:39 | |
dims | mriedem : we have flags to control that all over the place.... :) | 21:39 |
*** ctrath has quit IRC | 21:39 | |
*** shaohe_feng has joined #openstack-nova | 21:40 | |
melwitt | dansmith: do you recall if ComputeNode.save ever used to update/save the associated Service record in the past? I know there used to be service_id in ComputeNode but it doesn't look like that would cause a save to result in a save of the Service record too | 21:40 |
dansmith | melwitt: no I wouldn't expect it would | 21:41 |
openstackgerrit | Sujitha proposed openstack/nova: Complete method verification of os-cloudpipe.inc https://review.openstack.org/314783 | 21:41 |
*** ijw_ has quit IRC | 21:41 | |
*** ducttape_ has quit IRC | 21:42 | |
*** lpetrut has quit IRC | 21:42 | |
*** lpetrut has joined #openstack-nova | 21:42 | |
*** ducttape_ has joined #openstack-nova | 21:42 | |
*** thorst_ has quit IRC | 21:42 | |
mriedem | dims: https://bugs.launchpad.net/oslo.messaging/+bug/1580352 | 21:44 |
openstack | Launchpad bug 1580352 in oslo.messaging "excepted exception is still logged as an error" [Undecided,New] | 21:44 |
*** ducttape_ has quit IRC | 21:44 | |
dims | ty mriedem | 21:44 |
*** ducttape_ has joined #openstack-nova | 21:44 | |
melwitt | dansmith: thanks for the sanity check | 21:44 |
*** gagehugo has quit IRC | 21:45 | |
*** neophy has quit IRC | 21:45 | |
openstackgerrit | Sarafraj Singh proposed openstack/nova: api-ref: image.inc - Update method validation https://review.openstack.org/314268 | 21:46 |
sneti | Hi oomichi...I have updated my patch as per your comments. Can you please review it: https://review.openstack.org/#/c/290156/ | 21:46 |
*** _ducttape_ has joined #openstack-nova | 21:47 | |
bauzas | melwitt: no | 21:48 |
bauzas | melwitt: we did cut the relationship between the Service object and the CN one | 21:49 |
*** ccarmack has quit IRC | 21:49 | |
bauzas | melwitt: and then, the service_id field is deprecated | 21:49 |
*** edmondsw has quit IRC | 21:49 | |
*** shaohe_feng has quit IRC | 21:50 | |
* bauzas is back from a meetup where he heard "AWS is open" | 21:50 | |
melwitt | bauzas: yeah, I remember that. I'm trying to figure out what's going on with this bug opened against Kilo https://bugs.launchpad.net/nova/+bug/1579213 | 21:50 |
openstack | Launchpad bug 1579213 in OpenStack Compute (nova) "ComputeFilter fails because compute node has not been heard from in a while" [Undecided,New] | 21:50 |
*** shaohe_feng has joined #openstack-nova | 21:50 | |
*** ijw has joined #openstack-nova | 21:51 | |
melwitt | bauzas: the reporter said he worked around by forcing a save of the compute node every time, but that's not connected to how service up/down is determined | 21:51 |
*** efried has quit IRC | 21:51 | |
melwitt | even in Kilo, IIUC | 21:51 |
*** sacharya has joined #openstack-nova | 21:52 | |
*** mriedem has quit IRC | 21:52 | |
*** ducttape_ has quit IRC | 21:52 | |
bauzas | melwitt: mmmm | 21:52 |
bauzas | melwitt: so the cut was done in Kilo IIRC | 21:52 |
*** cdelatte has joined #openstack-nova | 21:52 | |
melwitt | bauzas: yeah, I was looking at that. but even before the cut, it didn't look like any save of ComputeNode could cascade into a save of Service | 21:52 |
*** amotoki has joined #openstack-nova | 21:53 | |
*** sacharya_ has quit IRC | 21:54 | |
*** delattec has quit IRC | 21:54 | |
*** cdelatte has quit IRC | 21:55 | |
bauzas | melwitt: well, the ComputeFilter is checking the servicegroup AFAIR | 21:55 |
bauzas | melwitt: which means it depends on the backend | 21:55 |
*** daemontool_ has quit IRC | 21:55 | |
melwitt | bauzas: yeah, it does. and the servicegroup uses the Service last_seen_up, updated_at, created_at to determine up/down in the db driver case | 21:55 |
bauzas | melwitt: right | 21:55 |
melwitt | bauzas: but I was thinking, while the scheduler is working on a set of compute nodes, would a save of any compute node cause it to get a freshened view of the associated service record? | 21:56 |
bauzas | melwitt: sec, verifying | 21:57 |
*** fragatina has joined #openstack-nova | 21:57 | |
*** rama_y has joined #openstack-nova | 21:57 | |
melwitt | bauzas: or would it see nothing different because it has already pulled the compute node list and is seeing a potentially outdated view depending on how long scheduling takes | 21:57 |
bauzas | melwitt: so the SG backend verifies the Service object, not the CN object | 21:57 |
melwitt | right | 21:57 |
bauzas | melwitt: but that object is updated elsewhere but the RT AFAIR | 21:58 |
bauzas | sec, verifying where | 21:58 |
melwitt | bauzas: it's a timer task in the servicegroup, it increments report_count and does a save | 21:58 |
melwitt | bauzas: I was just wondering how saving compute nodes could possibly have made anything better in this case. the only thing I can think is if saving a compute node somehow refreshes the object and service object that the scheduler is looking at as it iterates over the nods | 21:59 |
melwitt | *nodes | 21:59 |
*** cdelatte has joined #openstack-nova | 21:59 | |
*** shaohe_feng has quit IRC | 22:00 | |
bauzas | melwitt: the problem is that you can have multiple RTs per compute service | 22:00 |
*** shaohe_feng has joined #openstack-nova | 22:00 | |
bauzas | melwitt: which means that each RT (one per compute node) would then update the related service | 22:01 |
bauzas | in a synchronous call | 22:01 |
melwitt | bauzas: this bug is using Ironic, which means there are many nodes to one compute service | 22:02 |
bauzas | melwitt: yeah, which means we have one RT per Ironic node | 22:02 |
melwitt | right | 22:02 |
bauzas | I'm surprised to see that bug popping up now TBH | 22:03 |
bauzas | I'll need to go to bed, but I'll try to look at that tomorrow | 22:05 |
bauzas | melwitt: ^ | 22:05 |
bauzas | FWIW, I'm pretty busy those days trying to finish some implementation for a non-priority BP | 22:05 |
melwitt | bauzas: okay. I'm going to comment on the bug with a question for the reporter | 22:05 |
melwitt | bauzas: no worries, thanks for letting me pick your scheduler brain | 22:06 |
bauzas | (which makes me pretty off the API doc effort btw. :( ) | 22:06 |
*** ccarmack has joined #openstack-nova | 22:06 | |
bauzas | that makes me think of sprints being in first milestones as something difficult if we work on non-prio BPs | 22:07 |
*** salv-orl_ has quit IRC | 22:07 | |
*** mageshgv has quit IRC | 22:07 | |
bauzas | anyway, \o | 22:08 |
*** cdelatte has quit IRC | 22:08 | |
melwitt | o/ | 22:08 |
*** burt has quit IRC | 22:09 | |
*** markvoelker_ has joined #openstack-nova | 22:10 | |
*** shaohe_feng has quit IRC | 22:10 | |
oomichi | sneti: thanks, +2 | 22:11 |
*** ccarmack has quit IRC | 22:11 | |
*** shaohe_feng has joined #openstack-nova | 22:11 | |
sneti | oomichi, thank you!! | 22:11 |
*** cdelatte has joined #openstack-nova | 22:11 | |
*** thorst_ has joined #openstack-nova | 22:12 | |
*** kfarr has quit IRC | 22:12 | |
*** _ducttape_ has quit IRC | 22:13 | |
*** sheel has quit IRC | 22:15 | |
*** thorst_ has quit IRC | 22:16 | |
*** thorst_ has joined #openstack-nova | 22:17 | |
openstackgerrit | Pushkar Umaranikar proposed openstack/nova: api-ref: Method verification for os-floating-ips https://review.openstack.org/314794 | 22:18 |
*** thorst_ has quit IRC | 22:19 | |
*** roeyc has joined #openstack-nova | 22:19 | |
*** svenkat has quit IRC | 22:19 | |
*** shaohe_feng has quit IRC | 22:20 | |
*** shaohe_feng has joined #openstack-nova | 22:21 | |
*** sdake has joined #openstack-nova | 22:22 | |
*** roeyc has quit IRC | 22:23 | |
*** haplo37 has quit IRC | 22:24 | |
*** delattec has joined #openstack-nova | 22:24 | |
*** cdelatte has quit IRC | 22:27 | |
*** whenry has joined #openstack-nova | 22:27 | |
*** dimtruck is now known as zz_dimtruck | 22:27 | |
*** lpetrut has quit IRC | 22:28 | |
*** thorst_ has joined #openstack-nova | 22:30 | |
*** shaohe_feng has quit IRC | 22:31 | |
*** thorst_ has quit IRC | 22:31 | |
*** thorst_ has joined #openstack-nova | 22:31 | |
*** roeyc has joined #openstack-nova | 22:31 | |
*** shaohe_feng has joined #openstack-nova | 22:31 | |
*** markvoelker_ has quit IRC | 22:32 | |
*** dedery__ has quit IRC | 22:32 | |
*** jdurgin has quit IRC | 22:33 | |
*** dedery has joined #openstack-nova | 22:33 | |
*** jdurgin has joined #openstack-nova | 22:34 | |
mgagne | melwitt: tyvm for the follow up on bug #1579213. like I mentioned in the bug report, I will reverify my first assumption and see they still hold with the new knowledges you provided. | 22:34 |
openstack | bug 1579213 in OpenStack Compute (nova) "ComputeFilter fails because compute node has not been heard from in a while" [Undecided,New] https://launchpad.net/bugs/1579213 | 22:34 |
*** mrkz has joined #openstack-nova | 22:35 | |
melwitt | mgagne: great. I'm very interested in uncovering what's going on there | 22:37 |
*** cdent has quit IRC | 22:37 | |
openstackgerrit | Anusha Unnam proposed openstack/nova: api-ref: os-certificates.inc method verification https://review.openstack.org/314796 | 22:37 |
*** raildo is now known as raildo-afk | 22:38 | |
mgagne | melwitt: I think we are hitting a scalability issue with the scheduler where scheduling takes more than 60s | 22:38 |
*** claudiub has quit IRC | 22:40 | |
melwitt | mgagne: I have seen that before in a 700+ node deployment. we "fixed it" by moving ComputeFilter to the front of the filters list. also, we were running Juno which had some db accesses in some filters, which were very slow once compute nodes got more numerous | 22:40 |
mgagne | melwitt: I found that our dev team added some custom filters to the scheduler which could explain why ComputeFilter (and servicegroup) thinks nova-compute is down. | 22:41 |
melwitt | mgagne: so I think we backported some changes from Liberty (?) that removed db access from filters | 22:41 |
*** shaohe_feng has quit IRC | 22:41 | |
mgagne | melwitt: yes, that's the conclusion I come to by talking to my colleagues atm | 22:41 |
mgagne | came* | 22:41 |
openstackgerrit | Pushkar Umaranikar proposed openstack/nova: api-ref: Method verification for floating-ip-pools https://review.openstack.org/314798 | 22:41 |
*** shaohe_feng has joined #openstack-nova | 22:41 | |
mgagne | melwitt: yes, it makes perfect sense. | 22:42 |
*** vladikr has quit IRC | 22:42 | |
mgagne | I will update the bug report with those findings so people can at least look at the filter order if they think they have the same problem | 22:42 |
*** zenoway has quit IRC | 22:43 | |
*** mdrabe has quit IRC | 22:44 | |
melwitt | mgagne: sounds good. the only thing that remains unexplained is how forcing a ComputeNode save helped your case. from what I looked around in the Kilo code I don't find connections | 22:44 |
*** zenoway has joined #openstack-nova | 22:44 | |
openstackgerrit | Pushkar Umaranikar proposed openstack/nova: api-ref: Method verification for floating-ip-pools https://review.openstack.org/314798 | 22:44 |
mgagne | yea, if service state is cached in scheduler once scheduling process begins, I don't see how sending scheduler updates from compute fixes things. Unless I'm missing some complex details about the resource tracker inner working | 22:45 |
*** jwcroppe_ has joined #openstack-nova | 22:45 | |
*** jwcroppe has quit IRC | 22:47 | |
melwitt | mgagne: yeah. the twist is you're using Ironic which means you have one resource tracker per Ironic node but only one nova-compute Service, which is much different than the usual vm case of one resource tracker per nova-compute | 22:47 |
mgagne | yes. This further increases my interest toward all those nova-scheduler performance works people were talking about at the summit | 22:48 |
*** zenoway has quit IRC | 22:48 | |
*** mriedem has joined #openstack-nova | 22:50 | |
*** shaohe_feng has quit IRC | 22:51 | |
*** shaohe_feng has joined #openstack-nova | 22:51 | |
melwitt | mgagne: same here. you should check out the nova-specs for generic resource pools and dynamic resources classes if you haven't already. there's some stuff in there about allocating nova-compute processes to aggregates, which AFAICT would solve the bottleneck problem we have with Ironic | 22:51 |
mgagne | melwitt: this is definitely in my area of interests. I'm still trying to sell the idea to upper management that reviewing and contributing to those specs would be beneficial to us. mainly looking for time resources =) | 22:53 |
melwitt | mgagne: yeah, I know what you mean :) | 22:54 |
*** abalutoiu has quit IRC | 22:54 | |
mgagne | https://review.openstack.org/#/c/313519/ right? | 22:54 |
melwitt | mgagne: no, there's this one https://review.openstack.org/#/c/300176/ and I'm looking for the second one | 22:56 |
melwitt | https://review.openstack.org/#/c/312696/ | 22:57 |
mgagne | tyvm | 22:57 |
*** baoli_ has quit IRC | 23:01 | |
*** shaohe_feng has quit IRC | 23:01 | |
openstackgerrit | Karen Bradshaw proposed openstack/nova: [WIP] servers-action-evacuate.inc https://review.openstack.org/314776 | 23:03 |
*** tonytan4ever has quit IRC | 23:03 | |
*** shaohe_feng has joined #openstack-nova | 23:04 | |
openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: method verification of os-assisted-volume-snapshots https://review.openstack.org/314802 | 23:05 |
mriedem | jaypipes: have you and cdent come to some sort of understanding on the generic-resource-pools spec? | 23:06 |
mriedem | i saw cdent was pushing changes | 23:06 |
*** zz_dimtruck is now known as dimtruck | 23:09 | |
*** efried has joined #openstack-nova | 23:10 | |
*** efried has quit IRC | 23:11 | |
*** shaohe_feng has quit IRC | 23:12 | |
*** shaohe_feng has joined #openstack-nova | 23:12 | |
*** jamielennox is now known as jamielennox|away | 23:13 | |
*** salv-orlando has joined #openstack-nova | 23:16 | |
*** roeyc has quit IRC | 23:17 | |
*** delatte has joined #openstack-nova | 23:18 | |
*** zenoway has joined #openstack-nova | 23:19 | |
*** hemna is now known as hemnafk | 23:20 | |
*** svenkat has joined #openstack-nova | 23:21 | |
*** delattec has quit IRC | 23:21 | |
*** salv-orlando has quit IRC | 23:21 | |
*** sdague has quit IRC | 23:21 | |
*** jamielennox|away is now known as jamielennox | 23:21 | |
*** shaohe_feng has quit IRC | 23:22 | |
*** diogogmt has quit IRC | 23:23 | |
mriedem | dansmith: is my understanding here correct? https://review.openstack.org/#/c/312751/7/nova/compute/api.py@978 | 23:23 |
mgagne | melwitt: it's strange because when comparing nova-scheduler log entries with updated_at field of the service, there is a different of 8 minutes :O (2016-05-06 17:43:51+00:00 vs 2016-05-06 17:51:40.002) that seems very strange. Could accessing some host_state field trigger an implicit database access? | 23:23 |
dansmith | mriedem: indeed | 23:24 |
mriedem | f yeah | 23:24 |
*** zenoway has quit IRC | 23:25 | |
*** shaohe_feng has joined #openstack-nova | 23:25 | |
*** rcernin has quit IRC | 23:28 | |
*** nic has quit IRC | 23:29 | |
*** sdake has quit IRC | 23:30 | |
*** nic has joined #openstack-nova | 23:30 | |
*** nic has quit IRC | 23:31 | |
melwitt | mgagne: I don't think so. the thing I was thinking of was what this commit fixed https://github.com/openstack/nova/commit/3aff2d7bff7f6e9edb5fa8b688287265722c27fb | 23:31 |
*** shaohe_feng has quit IRC | 23:32 | |
melwitt | mgagne: that commit is part of 2015.1.2 so I think you shouldn't have any db access in your filters. you'll also want to take a look at what you custom filters are doing | 23:32 |
mgagne | yea, I guess I can post them somewhere | 23:32 |
melwitt | mgagne: for an easy check I would do something like grep for "db\." under nova/scheduler and see if anything comes up, assuming your custom filters are also in that dir | 23:34 |
*** sacharya has quit IRC | 23:34 | |
*** shaohe_feng has joined #openstack-nova | 23:35 | |
*** sacharya has joined #openstack-nova | 23:35 | |
melwitt | mgagne: not that actually but maybe just "db" or "objects" as I notice the db accesses weren't using the db api directly in the commit I linked | 23:35 |
mgagne | there is no db module included in our filters nor objects usage. I guess I'll have to profile those filters and see where the time is spent | 23:36 |
melwitt | mgagne: you can find out how long each filter is taking if you turn on debug logging on the scheduler and compare timestamps. to zero in on the slow filter(s) | 23:36 |
melwitt | I turned it on briefly and did a few requests to collect something to look at | 23:38 |
*** sdake has joined #openstack-nova | 23:38 | |
dansmith | LOL https://wiki.openstack.org/wiki/ConfigureInstanceTypesDynamically | 23:38 |
dansmith | glad we have that "proposal" on the wiki | 23:39 |
melwitt | mgagne: based on how long each filter takes and how many hosts are eliminated by each filter, you can reorder your filters to place the most restrictive ones earlier to get rid of more hosts sooner and speed up scheduling | 23:39 |
*** rooooook has quit IRC | 23:39 | |
mgagne | not all filters are logging on our side. will have to add traces I guess | 23:40 |
mgagne | melwitt: yea, will have to do that work I guess now that it's a problem | 23:40 |
melwitt | mgagne: oh, on debug even not logging? there should be at least a message about how many hosts passed each filter after each one runs and you can tell how long they took by time deltas on those | 23:41 |
*** shaohe_feng has quit IRC | 23:42 | |
mgagne | melwitt: I'll do my homework and double checks. Looks like DEBUG wasn't enabled in the end, only WARNING were showing from some filters :-/ | 23:43 |
*** zenoway has joined #openstack-nova | 23:43 | |
mgagne | but yes, you are right, profile, reorder and update code if needed | 23:43 |
melwitt | dansmith: wait, is that how instance_types went into a db migration? | 23:43 |
*** rooooook has joined #openstack-nova | 23:43 | |
*** rooooook has quit IRC | 23:43 | |
*** shaohe_feng has joined #openstack-nova | 23:44 | |
dansmith | melwitt: well, I guess, but it's actually "don't just hard code 7 flavors in python and actually let people *create* flavors" :) | 23:44 |
melwitt | mgagne: yeah, you don't want to have DEBUG on in general, turn on temporarily to collect data and then turn off. it will slow things down | 23:44 |
mgagne | yes | 23:44 |
dansmith | melwitt: if you look at the initial checkin, some file was like "flavors = {'m1.small': ..." | 23:44 |
*** rooooook has joined #openstack-nova | 23:44 | |
melwitt | dansmith: whoa | 23:45 |
*** zenoway has quit IRC | 23:47 | |
*** shaohe_feng has quit IRC | 23:53 | |
*** shaohe_feng has joined #openstack-nova | 23:54 | |
*** crinkle has quit IRC | 23:55 | |
*** psargent has left #openstack-nova | 23:55 | |
*** thorst_ has quit IRC | 23:57 | |
*** pumarani__ has joined #openstack-nova | 23:57 | |
*** thorst_ has joined #openstack-nova | 23:58 | |
*** pushkaru has quit IRC | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!