openstackgerrit | Matt Riedemann proposed openstack/nova: Fix the terminated_at field in the server query params schema https://review.openstack.org/427964 | 00:00 |
---|---|---|
*** erlon has quit IRC | 00:00 | |
*** krtaylor has quit IRC | 00:02 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: Handle NotImplementedError in _process_instance_vif_deleted_event https://review.openstack.org/427974 | 00:05 |
*** karimb has quit IRC | 00:06 | |
*** mriedem has quit IRC | 00:06 | |
*** xyang1 has quit IRC | 00:12 | |
*** hongbin has quit IRC | 00:16 | |
*** tbachman has quit IRC | 00:18 | |
*** karimb has joined #openstack-nova | 00:18 | |
*** sdague has quit IRC | 00:21 | |
*** jaosorior has quit IRC | 00:27 | |
*** unicell has quit IRC | 00:34 | |
*** dtp-afk has quit IRC | 00:34 | |
*** yamahata has quit IRC | 00:35 | |
*** owalsh-afk has quit IRC | 00:37 | |
*** chyka has quit IRC | 00:38 | |
*** dimtruck is now known as zz_dimtruck | 00:38 | |
*** ducttape_ has quit IRC | 00:39 | |
*** ducttape_ has joined #openstack-nova | 00:39 | |
*** ducttape_ has quit IRC | 00:41 | |
*** _ducttape_ has joined #openstack-nova | 00:41 | |
*** thorst_ has joined #openstack-nova | 00:41 | |
*** _ducttape_ has quit IRC | 00:42 | |
*** ducttape_ has joined #openstack-nova | 00:42 | |
*** nic has quit IRC | 00:43 | |
*** unicell has joined #openstack-nova | 00:48 | |
*** tbachman has joined #openstack-nova | 00:48 | |
*** ijw has quit IRC | 00:50 | |
*** jamesdenton has joined #openstack-nova | 00:53 | |
*** rfolco has quit IRC | 00:58 | |
*** openstackgerrit has quit IRC | 01:02 | |
*** thorst_ has quit IRC | 01:05 | |
*** dave-mccowan has quit IRC | 01:05 | |
*** thorst_ has joined #openstack-nova | 01:06 | |
*** john51 has joined #openstack-nova | 01:06 | |
*** rfolco has joined #openstack-nova | 01:08 | |
*** rfolco has quit IRC | 01:08 | |
*** thorst_ has quit IRC | 01:10 | |
*** john51_ has quit IRC | 01:11 | |
*** raunak has quit IRC | 01:13 | |
*** unicell has quit IRC | 01:14 | |
*** raunak has joined #openstack-nova | 01:15 | |
*** raunak has quit IRC | 01:15 | |
*** thorst_ has joined #openstack-nova | 01:16 | |
oomichi | mriedem: cool, nice catch | 01:22 |
*** krtaylor has joined #openstack-nova | 01:22 | |
*** thorst_ has quit IRC | 01:25 | |
*** rfolco has joined #openstack-nova | 01:27 | |
*** rfolco has quit IRC | 01:27 | |
*** rfolco has joined #openstack-nova | 01:27 | |
*** rfolco has quit IRC | 01:28 | |
*** karimb has quit IRC | 01:30 | |
*** openstackgerrit has joined #openstack-nova | 01:31 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 01:31 |
*** rfolco has joined #openstack-nova | 01:31 | |
*** rfolco has quit IRC | 01:31 | |
*** rfolco has joined #openstack-nova | 01:32 | |
*** unicell has joined #openstack-nova | 01:32 | |
*** rfolco has quit IRC | 01:32 | |
*** mriedem has joined #openstack-nova | 01:32 | |
*** tlian has quit IRC | 01:32 | |
*** thorst_ has joined #openstack-nova | 01:33 | |
*** thorst_ has quit IRC | 01:33 | |
*** zz_dimtruck is now known as dimtruck | 01:35 | |
mriedem | https://review.openstack.org/#/c/427782/ is merged | 01:37 |
mriedem | woot | 01:37 |
mriedem | hopefully we see that bug drop off | 01:37 |
mriedem | the postgres fix is merged too https://review.openstack.org/#/c/427667/ | 01:38 |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 01:40 |
*** tlian has joined #openstack-nova | 01:41 | |
*** Apoorva has quit IRC | 01:41 | |
*** sree has joined #openstack-nova | 01:44 | |
*** jose-phillips has quit IRC | 01:44 | |
*** hfu has joined #openstack-nova | 01:46 | |
melwitt | mriedem: I had forgotten I said I'd write up step-by-step examples for setting up cells v2, so I did it just now. do you think it's too late for that? | 01:46 |
*** thorst_ has joined #openstack-nova | 01:53 | |
*** mtanino has quit IRC | 01:53 | |
mriedem | no | 01:53 |
mriedem | link? | 01:53 |
mriedem | we really needed some upgrade steps | 01:53 |
mriedem | but i'll take anything | 01:53 |
melwitt | yay good. one sec, getting some sphinx warnings | 01:53 |
melwitt | "Title level inconsistent" I'm trying to figure out what level is one step smaller than ~~~~~ for a section | 01:54 |
*** thorst_ has quit IRC | 01:57 | |
mriedem | ----- ? | 01:59 |
mriedem | melwitt: ^ | 01:59 |
mriedem | Cells V2 ======== Manifesto ~~~~~~~~~ Proposal -------- | 02:00 |
melwitt | yeah, I think you're right. thanks | 02:00 |
*** Sukhdev has quit IRC | 02:02 | |
*** Apoorva has joined #openstack-nova | 02:02 | |
openstackgerrit | melanie witt proposed openstack/nova master: Add step-by-step examples for Cells V2 setup https://review.openstack.org/427981 | 02:03 |
*** jdillaman has quit IRC | 02:04 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 02:04 |
melwitt | mriedem: ^ | 02:04 |
gmann | Kevin_Zheng: looks like you are back, updating doc patch as per mriedem comments ? | 02:05 |
*** Apoorva has quit IRC | 02:07 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Trivial Fix typo in schema https://review.openstack.org/427983 | 02:08 |
*** zhurong has joined #openstack-nova | 02:08 | |
Kevin_Zheng | gmann: yeah I'm trying to | 02:08 |
*** tanee is now known as tanee_away | 02:08 | |
*** efried has quit IRC | 02:09 | |
gmann | Kevin_Zheng: thanks, please let me know any help. we should get those last bits in this week | 02:09 |
*** Jeffrey4l_ has quit IRC | 02:10 | |
Kevin_Zheng | gmann: sure | 02:10 |
Kevin_Zheng | Thanks | 02:10 |
*** tanee_away is now known as tanee | 02:10 | |
*** efried has joined #openstack-nova | 02:11 | |
*** dimtruck is now known as zz_dimtruck | 02:11 | |
*** unicell has quit IRC | 02:12 | |
gmann | Kevin_Zheng: terminated_at is fixed in https://review.openstack.org/#/c/427964/ | 02:15 |
Kevin_Zheng | Yeah just saw it | 02:15 |
gmann | ok | 02:16 |
*** jdillaman has joined #openstack-nova | 02:17 | |
*** cdelatte has quit IRC | 02:19 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova-specs master: Enable cold migration with target host https://review.openstack.org/426636 | 02:20 |
*** zhurong has quit IRC | 02:20 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova-specs master: Abort Cold Migration https://review.openstack.org/334732 | 02:21 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Fix deprecated proxy API parameters https://review.openstack.org/416882 | 02:21 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Use ListOfIntegersField in oslo.versionedobjects https://review.openstack.org/423912 | 02:22 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Fix parameters and examples in aggregate API https://review.openstack.org/425512 | 02:22 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Fix parameters in os-hosts.inc https://review.openstack.org/416444 | 02:22 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Fix doc generation warnings https://review.openstack.org/425549 | 02:22 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Refactor a test method including 3 test cases https://review.openstack.org/394738 | 02:22 |
mriedem | melwitt: comments inline | 02:24 |
mriedem | melwitt: i worry a little about duplicating some of the stuff that dan wrote above that in the first time setup section | 02:24 |
*** zz_dimtruck is now known as dimtruck | 02:25 | |
melwitt | thanks | 02:26 |
melwitt | mriedem: I thought about that a little too. I was thinking we should probably have short "do these steps" things for people to look at to help understand the different commands. well, based on the convos we had with operators in the channel about it | 02:28 |
mriedem | yeah, | 02:29 |
*** thorst_ has joined #openstack-nova | 02:29 | |
*** thorst_ has quit IRC | 02:29 | |
mriedem | i don't have a great solution to not be redundant but at the same time, have everything anyone would ever want or need to know in a single section and not make that so overly complicate with if/else conditions that no one can understand it | 02:29 |
mriedem | can't win really | 02:29 |
*** sree has quit IRC | 02:29 | |
mriedem | i've always failed at process/guide docs like that | 02:30 |
*** dimtruck is now known as zz_dimtruck | 02:30 | |
mriedem | the balance between too much and too little info, because no matter how you write it, someone won't be able to follow it | 02:30 |
mriedem | lived with that for years in my last project before openstack | 02:30 |
melwitt | yeah. what's there already is a thorough explanation of all the things, but for the sort of roll up or quick reference, we don't have anything | 02:31 |
melwitt | maybe that page isn't the best place for it but I wasn't sure where else we would put it | 02:31 |
openstackgerrit | Hidekazu Nakamura proposed openstack/nova master: Do not attempt volume swap when guest is stopped https://review.openstack.org/389798 | 02:33 |
*** raunak has joined #openstack-nova | 02:38 | |
*** raunak has quit IRC | 02:40 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Handle NotImplementedError in _process_instance_vif_deleted_event https://review.openstack.org/427974 | 02:41 |
*** raunak has joined #openstack-nova | 02:42 | |
*** raunak has quit IRC | 02:42 | |
*** hieulq has joined #openstack-nova | 02:48 | |
mriedem | melwitt: so i thought about this on the way home for some reason https://review.openstack.org/#/c/355091/ | 02:51 |
mriedem | i don't really want to open a can of worms, | 02:51 |
mriedem | but i'm not sure if what we have for placement right now resolves that with reporting 0 DISK_GB for boot from volume or not | 02:51 |
mriedem | i think it does | 02:52 |
*** thorst_ has joined #openstack-nova | 02:52 | |
*** thorst_ has quit IRC | 02:52 | |
*** kaisers_ has joined #openstack-nova | 02:53 | |
melwitt | mriedem: yeah, I skimmed some ML threads on that and it wasn't clear to me if the boot from volume disk reporting problem is solved yet with placement | 02:53 |
mriedem | "We're already reporting proper allocation information for instances with no disk (volume-backed and others) and as soon as we look to the placement API in the scheduler we'll stop suffering from this problem." | 02:53 |
mriedem | ^ as long as the scheduler is using placement, we're filtering on DISK_GB now | 02:53 |
melwitt | mriedem: is it now? I've been way out of the loop on what's merged | 02:54 |
*** hongbin has joined #openstack-nova | 02:55 | |
mriedem | melwitt: yeah https://review.openstack.org/#/c/417961/ | 02:56 |
mriedem | if the minimum nova-compute service version in the deployment is new enough, the filter scheduler will use placement now, | 02:56 |
mriedem | including filtering on DISK_GB | 02:56 |
*** kaisers has quit IRC | 02:56 | |
mriedem | https://review.openstack.org/#/c/417961/41/nova/scheduler/filter_scheduler.py@139 | 02:56 |
melwitt | okay, cool | 02:57 |
melwitt | well, that's a win, glad to see that | 02:58 |
mriedem | https://review.openstack.org/#/c/427367/ adds some upgrade notes to the placement devref | 02:58 |
mriedem | might help you catch up a bit | 02:58 |
melwitt | thanks | 02:59 |
mriedem | i missed pretty much all of the moving instance create to conductor, but after the last 24 hours of bugs i feel like i've caught up :) | 02:59 |
*** zerda3 has joined #openstack-nova | 02:59 | |
melwitt | I wonder how it makes sure volume backed instance doesn't request the flavor's amount of disk when rebuilding etc | 03:00 |
melwitt | haha, yeah really | 03:00 |
*** raunak has joined #openstack-nova | 03:01 | |
*** bkopilov has quit IRC | 03:02 | |
*** ssurana has quit IRC | 03:02 | |
*** tlian has quit IRC | 03:06 | |
*** jschlueter has quit IRC | 03:08 | |
*** jschlueter has joined #openstack-nova | 03:09 | |
*** hieulq has quit IRC | 03:09 | |
*** jamesdenton has quit IRC | 03:10 | |
*** jamesden_ has joined #openstack-nova | 03:10 | |
mriedem | got some other cells v2 related docs changes here, small ones https://review.openstack.org/#/q/topic:man+project:openstack/nova+status:open | 03:11 |
*** thorst_ has joined #openstack-nova | 03:13 | |
*** tlian has joined #openstack-nova | 03:18 | |
*** mtanino has joined #openstack-nova | 03:20 | |
*** thorst_ has quit IRC | 03:20 | |
*** gouthamr has quit IRC | 03:24 | |
*** mtanino has quit IRC | 03:24 | |
*** abalutoiu has quit IRC | 03:34 | |
*** armax has quit IRC | 03:40 | |
*** Jeffrey4l has joined #openstack-nova | 03:41 | |
*** jamesden_ has quit IRC | 03:41 | |
*** jamesdenton has joined #openstack-nova | 03:41 | |
*** jamesdenton has quit IRC | 03:50 | |
*** tlian has quit IRC | 03:52 | |
*** jamesdenton has joined #openstack-nova | 03:53 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix typo in help for discover_hosts_in_cells_interval https://review.openstack.org/427997 | 03:54 |
*** thorst_ has joined #openstack-nova | 03:55 | |
*** thorst_ has quit IRC | 03:55 | |
*** armax has joined #openstack-nova | 03:56 | |
*** baoli has joined #openstack-nova | 03:57 | |
*** tbachman has quit IRC | 03:57 | |
*** nicolasbock has quit IRC | 04:01 | |
*** raunak has quit IRC | 04:01 | |
*** baoli has quit IRC | 04:01 | |
*** tbachman has joined #openstack-nova | 04:02 | |
*** armax has quit IRC | 04:10 | |
*** diga has joined #openstack-nova | 04:14 | |
*** jamesdenton has quit IRC | 04:17 | |
*** tbachman has quit IRC | 04:17 | |
*** psachin has joined #openstack-nova | 04:18 | |
*** bkopilov has joined #openstack-nova | 04:19 | |
mriedem | gibi: https://review.openstack.org/#/c/427647/ apparently the cinder people have figured out how to short circuit sending notifications if they aren't enabled | 04:21 |
mriedem | https://review.openstack.org/#/c/426547 | 04:22 |
*** armax has joined #openstack-nova | 04:22 | |
*** dikonoor has joined #openstack-nova | 04:27 | |
*** mdnadeem has joined #openstack-nova | 04:29 | |
*** jamesdenton has joined #openstack-nova | 04:30 | |
*** guchihiro has joined #openstack-nova | 04:34 | |
*** armax has quit IRC | 04:34 | |
*** jamesdenton has quit IRC | 04:35 | |
*** jamesden_ has joined #openstack-nova | 04:35 | |
*** abhishek_k has joined #openstack-nova | 04:35 | |
*** abhishekk has quit IRC | 04:35 | |
*** ducttape_ has quit IRC | 04:36 | |
*** udesale has joined #openstack-nova | 04:36 | |
*** Kevin_Zheng has quit IRC | 04:37 | |
*** jamesdenton has joined #openstack-nova | 04:37 | |
*** jamesden_ has quit IRC | 04:39 | |
*** ssurana has joined #openstack-nova | 04:40 | |
*** jamesdenton has quit IRC | 04:42 | |
*** jamesden_ has joined #openstack-nova | 04:43 | |
*** ayogi has joined #openstack-nova | 04:48 | |
*** adisky_ has joined #openstack-nova | 04:50 | |
*** sree has joined #openstack-nova | 04:51 | |
*** hongbin_ has joined #openstack-nova | 04:53 | |
*** archies has joined #openstack-nova | 04:55 | |
*** hongbin has quit IRC | 04:56 | |
*** sree has quit IRC | 04:56 | |
*** sree has joined #openstack-nova | 04:58 | |
*** efried has quit IRC | 04:59 | |
*** hfu has quit IRC | 04:59 | |
*** hfu has joined #openstack-nova | 05:01 | |
*** hfu has quit IRC | 05:01 | |
*** pbandark has joined #openstack-nova | 05:01 | |
*** hfu has joined #openstack-nova | 05:01 | |
*** hfu has quit IRC | 05:02 | |
*** sree has quit IRC | 05:02 | |
*** hfu has joined #openstack-nova | 05:02 | |
*** hfu has quit IRC | 05:03 | |
archies | @team, I am facing an interesting issue while attaching a volume. In libvirt the error is "operation failed: open disk image file failed" | 05:03 |
archies | Any idea? | 05:03 |
*** sree has joined #openstack-nova | 05:04 | |
*** ratailor has joined #openstack-nova | 05:08 | |
*** guchihiro has quit IRC | 05:09 | |
*** efried has joined #openstack-nova | 05:12 | |
*** prateek has joined #openstack-nova | 05:12 | |
*** jamesden_ has quit IRC | 05:24 | |
*** bkopilov_ has joined #openstack-nova | 05:25 | |
*** amotoki has joined #openstack-nova | 05:28 | |
*** thorst_ has joined #openstack-nova | 05:33 | |
*** satyar has joined #openstack-nova | 05:37 | |
*** ducttape_ has joined #openstack-nova | 05:37 | |
*** thorst_ has quit IRC | 05:38 | |
*** ducttape_ has quit IRC | 05:42 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 05:42 |
*** ayogi has quit IRC | 05:46 | |
*** Sukhdev has joined #openstack-nova | 05:46 | |
*** trinaths has joined #openstack-nova | 05:47 | |
*** ayogi has joined #openstack-nova | 05:49 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 05:53 |
*** hongbin_ has quit IRC | 05:53 | |
*** unicell has joined #openstack-nova | 05:55 | |
*** unicell has quit IRC | 05:57 | |
*** unicell has joined #openstack-nova | 06:01 | |
trinaths | Hi- While enabling VM with huge pages, I get the following error. Insufficient compute resources: Requested instance NUMA topology cannot fit the given host NUMA topology. I’m using devstack (stable/newton) and nova (6.0.0). I have configured hugepages and flavors following the guide at http://docs.openstack.org/admin-guide/compute-huge-pages.html . Here is the configuration of the host. http://paste.openstack.org/sh | 06:01 |
*** hfu has joined #openstack-nova | 06:04 | |
*** mdnadeem has quit IRC | 06:06 | |
*** hfu has quit IRC | 06:06 | |
*** efried has quit IRC | 06:07 | |
*** sree has quit IRC | 06:08 | |
*** kaisers_ has quit IRC | 06:09 | |
*** artom_ has joined #openstack-nova | 06:09 | |
*** markvoelker has joined #openstack-nova | 06:10 | |
*** satyar_ has joined #openstack-nova | 06:10 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 06:11 |
*** efried has joined #openstack-nova | 06:11 | |
*** rnoriega has quit IRC | 06:11 | |
*** pkoniszewski has quit IRC | 06:11 | |
*** mdnadeem has joined #openstack-nova | 06:12 | |
*** artom has quit IRC | 06:12 | |
*** timello has quit IRC | 06:12 | |
*** markvoelker_ has quit IRC | 06:12 | |
*** egarbade- has quit IRC | 06:12 | |
*** John341_ has quit IRC | 06:12 | |
*** mnaser has quit IRC | 06:12 | |
*** satyar has quit IRC | 06:13 | |
*** abhishek_k has quit IRC | 06:13 | |
*** abhishek_k has joined #openstack-nova | 06:13 | |
*** John341_ has joined #openstack-nova | 06:14 | |
*** pkoniszewski has joined #openstack-nova | 06:15 | |
*** rnoriega has joined #openstack-nova | 06:18 | |
*** lpetrut has joined #openstack-nova | 06:18 | |
*** egarbade- has joined #openstack-nova | 06:19 | |
*** jdurgin1 has quit IRC | 06:19 | |
*** unicell1 has joined #openstack-nova | 06:20 | |
*** unicell has quit IRC | 06:21 | |
*** sree has joined #openstack-nova | 06:21 | |
*** mnaser has joined #openstack-nova | 06:23 | |
*** timello has joined #openstack-nova | 06:24 | |
*** Jeffrey4l has quit IRC | 06:25 | |
*** avolkov has joined #openstack-nova | 06:26 | |
*** Jeffrey4l has joined #openstack-nova | 06:26 | |
*** kaisers has joined #openstack-nova | 06:30 | |
*** mjura has joined #openstack-nova | 06:31 | |
*** vsaienko has joined #openstack-nova | 06:32 | |
*** Hosam has joined #openstack-nova | 06:35 | |
*** vsaienko has quit IRC | 06:37 | |
*** sudipto_ has joined #openstack-nova | 06:38 | |
*** sudipto has joined #openstack-nova | 06:38 | |
*** psachin has quit IRC | 06:42 | |
*** clayton has quit IRC | 06:46 | |
*** markus_z has joined #openstack-nova | 06:49 | |
*** raunak has joined #openstack-nova | 06:56 | |
*** psachin has joined #openstack-nova | 06:58 | |
*** thorst_ has joined #openstack-nova | 07:00 | |
*** rcernin has joined #openstack-nova | 07:06 | |
*** sridharg has joined #openstack-nova | 07:06 | |
*** thorst_ has quit IRC | 07:06 | |
*** lpetrut has quit IRC | 07:07 | |
*** ducttape_ has joined #openstack-nova | 07:08 | |
*** zhurong has joined #openstack-nova | 07:13 | |
*** ducttape_ has quit IRC | 07:13 | |
*** andreas_s has joined #openstack-nova | 07:15 | |
*** tesseract has joined #openstack-nova | 07:16 | |
*** kaisers has quit IRC | 07:16 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/427978 | 07:16 |
*** kaisers has joined #openstack-nova | 07:17 | |
*** lpetrut has joined #openstack-nova | 07:29 | |
*** Apoorva has joined #openstack-nova | 07:29 | |
*** psachin has quit IRC | 07:30 | |
markus_z | mikal: Would you maybe take a look at this newton backport? https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/newton+topic:%22bug+1455252%22 | 07:30 |
*** mptacekx has joined #openstack-nova | 07:30 | |
markus_z | mikal: ^ lyarwood already checked it. | 07:31 |
*** unicell1 has quit IRC | 07:36 | |
*** Apoorva has quit IRC | 07:36 | |
*** unicell has joined #openstack-nova | 07:36 | |
*** pcaruana has joined #openstack-nova | 07:39 | |
*** raunak has quit IRC | 07:40 | |
*** pcaruana has quit IRC | 07:42 | |
*** psachin has joined #openstack-nova | 07:42 | |
*** pcaruana has joined #openstack-nova | 07:43 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Enable cold migration with target host(1/2) https://review.openstack.org/408955 | 07:43 |
*** nkrinner_afk is now known as nkrinner | 07:45 | |
*** yamahata has joined #openstack-nova | 07:47 | |
*** Hosam has quit IRC | 07:47 | |
*** Hosam has joined #openstack-nova | 07:47 | |
*** unicell has quit IRC | 07:49 | |
*** efried has quit IRC | 07:51 | |
*** Hosam has quit IRC | 07:52 | |
*** efried has joined #openstack-nova | 07:55 | |
*** karimb has joined #openstack-nova | 07:56 | |
*** hfu has joined #openstack-nova | 08:06 | |
*** lpetrut has quit IRC | 08:07 | |
*** Sukhdev has quit IRC | 08:15 | |
*** karthiks has quit IRC | 08:16 | |
*** lpetrut has joined #openstack-nova | 08:18 | |
*** ygl has joined #openstack-nova | 08:26 | |
*** Hosam has joined #openstack-nova | 08:27 | |
*** zhurong has quit IRC | 08:28 | |
*** hfu has quit IRC | 08:28 | |
*** rcernin has quit IRC | 08:28 | |
*** ygl has quit IRC | 08:31 | |
*** slaweq has joined #openstack-nova | 08:33 | |
*** karthiks has joined #openstack-nova | 08:35 | |
gibi | Q~. | 08:35 |
gibi | sorry ^^ | 08:35 |
gibi | mriedem: thanks for the link, I will check the cinder solution | 08:35 |
*** ducttape_ has joined #openstack-nova | 08:38 | |
*** Hosam has quit IRC | 08:40 | |
*** Hosam has joined #openstack-nova | 08:41 | |
*** ducttape_ has quit IRC | 08:43 | |
*** edmondsw has joined #openstack-nova | 08:43 | |
*** hfu has joined #openstack-nova | 08:45 | |
*** edmondsw has quit IRC | 08:48 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Enable cold migration with target host(2/2) https://review.openstack.org/408964 | 08:49 |
*** rcernin has joined #openstack-nova | 08:52 | |
*** karimb has quit IRC | 08:53 | |
*** lpetrut has quit IRC | 08:57 | |
*** pbandark has quit IRC | 08:58 | |
*** gema has joined #openstack-nova | 08:59 | |
gema | can someone point me in the right direction regarding how to update this page: http://docs.openstack.org/developer/nova/feature_classification.html | 08:59 |
gema | are there any requirements in terms of testing that need to be met or is this manually maintained? | 08:59 |
*** zzzeek has quit IRC | 09:00 | |
*** lpetrut has joined #openstack-nova | 09:01 | |
*** zzzeek has joined #openstack-nova | 09:01 | |
*** pbandark has joined #openstack-nova | 09:02 | |
*** thorst_ has joined #openstack-nova | 09:02 | |
BobBall | johnthetubaguy: OK - so the XenServer CI problem is that https://review.openstack.org/#/c/396269/ merged breaking support for the Ubuntu we're using. | 09:05 |
*** thorst_ has quit IRC | 09:08 | |
*** karimb has joined #openstack-nova | 09:13 | |
*** karthiks has quit IRC | 09:13 | |
openstackgerrit | Lee Yarwood proposed openstack/nova master: libvirt: Remove redundant bdm serial mangling and saving during swap_volume https://review.openstack.org/427364 | 09:15 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Add functional tests for cold migration to same host https://review.openstack.org/414926 | 09:16 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Add functional tests for cold migration to same host https://review.openstack.org/414926 | 09:16 |
*** lpetrut has quit IRC | 09:19 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Add parameters in cold migrate action https://review.openstack.org/410042 | 09:19 |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Add parameters in cold migrate action https://review.openstack.org/410042 | 09:19 |
*** gszasz has joined #openstack-nova | 09:20 | |
johnthetubaguy | BobBall: really :S | 09:21 |
BobBall | Indeed. | 09:22 |
johnthetubaguy | BobBall: how did that break the uploads from inside the XenAPI plugin? | 09:23 |
BobBall | It broke Cinder which meant that the glance upload couldn't create the volume it was going to upload to | 09:23 |
johnthetubaguy | but I didn't think the xenapi plugin is used for any cinder volumes? | 09:23 |
johnthetubaguy | oh, it breaks all glance uploads in the destination sense? | 09:24 |
BobBall | uhhhh ok now you've got me confused... | 09:24 |
*** lucas-afk is now known as lucasagomes | 09:25 | |
*** gszasz has quit IRC | 09:27 | |
*** ralonsoh has joined #openstack-nova | 09:27 | |
*** karthiks has joined #openstack-nova | 09:27 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: api-ref: Add parameters in cold migrate action https://review.openstack.org/410042 | 09:29 |
*** Jeffrey4l has quit IRC | 09:30 | |
*** Jeffrey4l has joined #openstack-nova | 09:31 | |
*** ssurana has quit IRC | 09:37 | |
*** panda|off is now known as panda | 09:47 | |
*** yamahata has quit IRC | 09:48 | |
*** sambetts_ is now known as sambetts | 09:57 | |
mdbooth | lyarwood: Thinking about https://review.openstack.org/#/c/427364/ | 10:04 |
mdbooth | Specifically about the chunk in libvirt driver which you delete that we discussed yesterday | 10:05 |
*** owalsh-afk has joined #openstack-nova | 10:05 | |
*** owalsh-afk is now known as owalsh | 10:05 | |
mdbooth | I think the fundamental issue there is that libvirt volume drivers mutate connection_info | 10:05 |
mdbooth | The example which is specifically called out is LibvirtISCSIVolumeDriver. This adds the local device path that created to connect to the remote iscsi volume to connection_info. | 10:06 |
lyarwood | mdbooth: that's still added after my change | 10:07 |
mdbooth | But this isn't strictly related to the volume connection. | 10:07 |
lyarwood | mdbooth: we pass the modified connection_info back to the compute layer and update the BDM with it | 10:07 |
mdbooth | lyarwood: Yeah, I'm being expansive here :) | 10:07 |
lyarwood | ah | 10:07 |
lyarwood | :D | 10:07 |
mdbooth | So, anyway, I think the bug here is in LibvirtISCSIVolumeDriver and all other places which mutate connection_info | 10:08 |
openstackgerrit | Tomi Juvonen proposed openstack/nova-specs master: Add Host Details https://review.openstack.org/428070 | 10:08 |
mdbooth | They need to stash that metadata somewhere else. | 10:08 |
mdbooth | I also disapprove of even looking inside connection_info, tbh | 10:08 |
lyarwood | mdbooth: yeah that's all going to change with cinder v3 anyway | 10:09 |
*** ducttape_ has joined #openstack-nova | 10:09 | |
mdbooth | If we could have cinder pass it to us as an opaque packed C struct, that would be awesome | 10:09 |
mdbooth | The only thing you can do with this data is give it back to cinder | 10:09 |
lyarwood | mdbooth: but I can't see how that relates to this issue tbh | 10:09 |
*** sree has quit IRC | 10:09 | |
mdbooth | It relates because it means we *can't* refetch connection info from cinder after cinder also mutates it | 10:09 |
*** sree has joined #openstack-nova | 10:10 | |
mdbooth | Because Nova futzes with it, we're now stuck trying to second guess one or the other. | 10:10 |
lyarwood | mdbooth: ah right you're talking about the cinder retype situation | 10:11 |
mdbooth | cinder volume migration | 10:11 |
lyarwood | mdbooth: yeah same thing | 10:11 |
* mdbooth has never worked out what retype means | 10:11 | |
mdbooth | volume migration is meaningful to me :) | 10:11 |
lyarwood | mdbooth: I need to drop now, can we pick this up when I get back? | 10:12 |
mdbooth | lyarwood: Sure thing. | 10:12 |
lyarwood | mdbooth: cool | 10:12 |
*** satyar_ has quit IRC | 10:12 | |
openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Fix access_ip_v4/6 filters params for servers filter https://review.openstack.org/428071 | 10:13 |
gmann | johnthetubaguy: mriedem alex_xu seems like we missed another bit in filter/sort thing - https://bugs.launchpad.net/nova/+bug/1661195 | 10:13 |
openstack | Launchpad bug 1661195 in OpenStack Compute (nova) "Servers filter by access_ip_v4 does not filter servers" [Undecided,In progress] - Assigned to Ghanshyam Mann (ghanshyammann) | 10:13 |
gmann | johnthetubaguy: alex_xu mriedem this is fix - https://review.openstack.org/428071 | 10:14 |
*** ducttape_ has quit IRC | 10:14 | |
*** sree has quit IRC | 10:14 | |
johnthetubaguy | gmann: oh, you mean we should allow them, because we used to allow them | 10:14 |
gmann | but different filed name in filter and sort than what API represent is odd to me | 10:15 |
gmann | johnthetubaguy: i think so, or no ? | 10:15 |
johnthetubaguy | gmann: don't they now match what the API has? | 10:15 |
gmann | i may be missed the discussion on that if any | 10:15 |
gmann | johnthetubaguy: oh so you mean new list will match the what API has either it used to work before or not | 10:16 |
johnthetubaguy | gmann: the whole plan is the filter name matches what the API names are now, rather than matching the DB | 10:17 |
gmann | johnthetubaguy: yea | 10:17 |
gmann | johnthetubaguy: ok, i think m going in opposite way :) | 10:18 |
gmann | johnthetubaguy: i agree on that point | 10:18 |
gmann | johnthetubaguy: but bug seems now accessIPv4 work on filter or not as DB fileds are different for those? | 10:18 |
gmann | johnthetubaguy: and sort has 'access_ip_v4' not accessIPv4 so it is issue on sort list then | 10:19 |
johnthetubaguy | gmann: so we must have accessipv4 and accessipv6 as thats what in the API, in both filter and sort | 10:19 |
gmann | johnthetubaguy: yea and those should work as filter. because i doubt currently those works as we did not add translation layer for those | 10:20 |
johnthetubaguy | gmann: but... for backwards compatibility, we should probably also allow access_ip_v4 and access_ip_v6, and mark them for removal in a future microversion | 10:20 |
mdbooth | Can anybody confirm that schema migrations are intended to be idempotent? Specifically that I should be able to run db_sync multiple times? | 10:20 |
johnthetubaguy | gmann: yeah, we should add a test for both | 10:20 |
johnthetubaguy | mdbooth: thats correct | 10:20 |
gmann | johnthetubaguy: yea, that was actually i was thinking initially | 10:21 |
gmann | johnthetubaguy: cool, | 10:21 |
gmann | johnthetubaguy: ll update | 10:21 |
johnthetubaguy | gmann: cool | 10:21 |
johnthetubaguy | gmann: we should give that the tag, and get it on mriedem's etherpad I think | 10:22 |
johnthetubaguy | RC potential thing | 10:22 |
mdbooth | johnthetubaguy: Thought so, thanks. I'm seeing an error in 216_havana creating the instances table, specifically line 1046 where it calls instances.create() raises an error containing: (1050, u"Table 'instances' already exists") | 10:22 |
gmann | johnthetubaguy: but last doubt, is it ok to introduced new param without version up. i mean accessIPv4 are kind of new addition as filter | 10:22 |
gmann | johnthetubaguy: sure, ll do on etherpad | 10:22 |
mdbooth | It does already exist. Any idea how that's supposed to work? | 10:22 |
* mdbooth suspects a higher level failure, but isn't sure what to look for. | 10:23 | |
*** udesale has quit IRC | 10:24 | |
mdbooth | Or maybe it's a lower level failure... Perhaps sqlalchemy is supposed to transmute it into gold. | 10:24 |
*** hfu has quit IRC | 10:25 | |
johnthetubaguy | mriedem: hmm thats odd, thats the our base one that everyone has run again on upgrade for years, its curious that has broken | 10:25 |
gmann | johnthetubaguy: i remember we did microversion when adding ipv6 as filter | 10:25 |
mdbooth | johnthetubaguy: Yeah, suspect it's something weird in this environment. | 10:26 |
*** Hosam has quit IRC | 10:26 | |
mdbooth | File hasn't changed in ages. | 10:26 |
mdbooth | It could be an interesting candidate for robustification, though. | 10:26 |
mdbooth | If I can work out why it's broken. | 10:27 |
*** Hosam has joined #openstack-nova | 10:27 | |
johnthetubaguy | gmann: yes, thats a good point, we also have these: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/schemas/servers.py#L354 :S | 10:27 |
johnthetubaguy | mdbooth: generally we try never to change those files, but yeah | 10:27 |
johnthetubaguy | flavors was the exception in there, I think | 10:27 |
gmann | johnthetubaguy: i thought of adding accessIPv4 with version up fir interop things | 10:28 |
* mdbooth is suspecting sqla weirdness, tbh. | 10:28 | |
johnthetubaguy | gmann: we haven't yet released the last fix, so I am OK with that for right now | 10:28 |
johnthetubaguy | mdbooth: maybe you died half way through the migration, and somehow that got committed into the DB? That might confuse things | 10:30 |
gmann | johnthetubaguy: but accessIPv4 are always new even before filter/sort fix. i means they are totally new to users after Ocata | 10:31 |
*** Hosam has quit IRC | 10:31 | |
johnthetubaguy | gmann: yeah, I am not sure why we added new ones without a microversion, now you meantion it, I got distracted with killing bad ones | 10:32 |
*** hfu has joined #openstack-nova | 10:33 | |
gmann | johnthetubaguy: yea, list is too huge we did not notice those | 10:33 |
*** owalsh is now known as owalsh-brb | 10:34 | |
gmann | johnthetubaguy: i have added that in etherpad also (so mriedem can mark that rc potential), ll check from home. | 10:35 |
johnthetubaguy | gmann: thanks | 10:36 |
johnthetubaguy | gmann: it was such a massive list, there were always going to be niggles, would love to squash then before release, if possible | 10:36 |
gmann | +1 | 10:36 |
*** owalsh-brb is now known as owalsh | 10:41 | |
*** erlon has joined #openstack-nova | 10:45 | |
sfinucan | bauzas: Can I +W this? https://review.openstack.org/#/c/426864/6 | 10:48 |
sfinucan | I assume it won't do anything silly like tag the release or whatnot | 10:48 |
*** rfolco has joined #openstack-nova | 10:51 | |
johnthetubaguy | sfinucan: that sounds fine to me, we have added lots of preludes by accident in the past | 10:51 |
sfinucan | johnthetubaguy: Righto - +W it is | 10:51 |
* sfinucan just noticed there's a unapproved dependency anyway so it won't merge yet | 10:52 | |
johnthetubaguy | oh year, I should review that one again | 10:52 |
bauzas | sfinucan: cool with me | 10:52 |
bauzas | johnthetubaguy: I'm already on it | 10:52 |
*** derekh has joined #openstack-nova | 10:52 | |
sfinucan | bauzas: Could you also take a look at https://review.openstack.org/#/c/427499/, seeing as you're far more familiar with that feature | 10:53 |
bauzas | sfinucan: mmm, I had a concern on the bottom patch | 10:55 |
bauzas | at least a typo | 10:55 |
sfinucan | bauzas: Hmm, if you want to fix it now we can just re-approve, right? | 10:56 |
sfinucan | Also, do you know what "barber pole sorting" is? :D | 10:56 |
*** tbachman has joined #openstack-nova | 10:56 | |
*** efried has quit IRC | 10:57 | |
sfinucan | I'm guessing it's Python's 'zip', but that's a guess | 10:57 |
bauzas | sfinucan: https://en.wikipedia.org/wiki/Barber's_pole#Computer_science | 10:58 |
sfinucan | bauzas: I don't think that's what he's talking about though | 10:58 |
sfinucan | That describes a loading bar - he's on about sorting something | 10:59 |
bauzas | sfinucan: that would mean it takes time IIUC | 10:59 |
bauzas | sfinucan: oh, you removed +W | 10:59 |
bauzas | sfinucan: but that doesn't mean the gate jobs will stop | 11:00 |
bauzas | sfinucan: we have to trigger a new PS | 11:00 |
sfinucan | bauzas: it doesn't? | 11:00 |
sfinucan | Oh | 11:00 |
bauzas | nope | 11:00 |
sfinucan | Well, should we fix that typo? | 11:00 |
bauzas | we could do that in a follow-up | 11:00 |
bauzas | now it's in the gate | 11:00 |
sfinucan | That sounds fair | 11:00 |
bauzas | and given we have 3 patches that could be trampled if we provide a new rev | 11:00 |
* sfinucan is disgusted that he missed a typo :( | 11:01 | |
bauzas | so, adding the fix at a top patch | 11:01 |
*** lpetrut has joined #openstack-nova | 11:01 | |
mdbooth | johnthetubaguy: In case you're interested, looks like the db state was probably corrupt, perhaps due to a previous incomplete cleanup. This was in a CI system. So robustification likely not relevant to anybody. | 11:01 |
bauzas | sfinucan: restored the +W | 11:02 |
sfinucan | (y) | 11:02 |
*** pbandark has quit IRC | 11:03 | |
*** shaner has quit IRC | 11:03 | |
*** thorst_ has joined #openstack-nova | 11:04 | |
*** shaner has joined #openstack-nova | 11:04 | |
*** aarefiev_afk is now known as aarefiev | 11:08 | |
*** thorst_ has quit IRC | 11:08 | |
sfinucan | Are we still merging bugfixes? If so, could I get some eyes on this (to be backported) change? https://review.openstack.org/#/c/396184/ | 11:09 |
*** efried has joined #openstack-nova | 11:09 | |
openstackgerrit | Stephen Finucane proposed openstack/nova master: Fix backwards compatibility for InstanceNUMACell https://review.openstack.org/396184 | 11:09 |
*** ducttape_ has joined #openstack-nova | 11:10 | |
*** pbandark has joined #openstack-nova | 11:12 | |
johnthetubaguy | sfinucan: I think so, although we should avoid high risk ones till tomorrow | 11:13 |
johnthetubaguy | sfinucan: we usually try to keep master not too different to stable/ocata so any backports we need are likely to be easier | 11:13 |
*** ducttape_ has quit IRC | 11:14 | |
openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Make 'get_domain' private https://review.openstack.org/417378 | 11:15 |
*** ociuhandu has quit IRC | 11:17 | |
johnthetubaguy | mdbooth: ah, good context, in complete cleanup would do it | 11:17 |
*** nicolasbock has joined #openstack-nova | 11:18 | |
sfinucan | Gerrit's "Add Reviewer" dialog is truly awful | 11:18 |
*** sudipto_ has quit IRC | 11:19 | |
*** sudipto has quit IRC | 11:19 | |
openstackgerrit | John Garbutt proposed openstack/nova-specs master: WIP: write up policy ideas for pike https://review.openstack.org/427872 | 11:22 |
*** ducttape_ has joined #openstack-nova | 11:30 | |
*** aloga has quit IRC | 11:35 | |
*** aloga has joined #openstack-nova | 11:36 | |
openstackgerrit | John Garbutt proposed openstack/nova-specs master: WIP: write up policy ideas for pike https://review.openstack.org/427872 | 11:37 |
*** tbachman has quit IRC | 11:39 | |
panda | any hints on what may cause this ? http://logs.openstack.org/periodic/periodic-tripleo-ci-centos-7-ovb-nonha/632c7f2/logs/overcloud-novacompute-0/var/log/nova/nova-compute.txt.gz#_2017-02-02_08_13_29_759 | 11:40 |
*** psachin has quit IRC | 11:43 | |
*** sree has joined #openstack-nova | 11:43 | |
kashyap | mdbooth: or johnthetubaguy: When you have a moment, need an opinon | 11:45 |
kashyap | Where do you think is appropriate for these kinds of tests to live? https://review.openstack.org/#/c/315786/ | 11:46 |
kashyap | They are trying to add the NMI Watchdog test for a Nova instance in Tempest. But Tempest upstream, rightfully rejects it, as there are no API operations, and it is too Nova-specific | 11:46 |
johnthetubaguy | kashyap: sounds like we need a nova tempest plugin | 11:47 |
johnthetubaguy | I would say nova functional tests, but they don't normally need images setup up, AFAIK, I could be wrong | 11:47 |
johnthetubaguy | kashyap: you really want to be talking to sdague about all that | 11:47 |
kashyap | johnthetubaguy: Does such a thing exists (or is it exist)? | 11:47 |
*** catintheroof has joined #openstack-nova | 11:47 | |
kashyap | johnthetubaguy: Yeah, maybe sdague is still getting ready to wake up | 11:48 |
johnthetubaguy | kashyap: there are some details here: http://docs.openstack.org/developer/nova/test_strategy.html#types-of-tests | 11:48 |
*** amoralej has joined #openstack-nova | 11:48 | |
johnthetubaguy | but looks like it might need updating | 11:48 |
* kashyap clicks | 11:48 | |
kashyap | johnthetubaguy: So, one place you are suggesting is - nova/nova/tests/functional | 11:48 |
*** ftersin has joined #openstack-nova | 11:52 | |
mdbooth | johnthetubaguy: More context on the db_sync thing, looks like we had a puppet job which retried on failure, which included exceeding its timeout. Seems db sync was taking too long and we ended up running 2 simultaneously. | 11:53 |
mdbooth | So perhaps robustification might be in order. | 11:53 |
mdbooth | Perhaps the migration should hold an exclusive lock while it runs. | 11:53 |
johnthetubaguy | mdbooth: it should be all done in a single transaction, for each version, I thought? | 11:53 |
johnthetubaguy | maybe we never added that stuff in db sync | 11:54 |
johnthetubaguy | maybe thats not possible... | 11:54 |
mdbooth | Does msql honour transactions for ddl? | 11:54 |
johnthetubaguy | I was just thinking the same question, I don't know | 11:55 |
johnthetubaguy | I had always assume that, but you might be right | 11:55 |
johnthetubaguy | nothing like an olso.concurrency lock to stop that being an issue, like you say | 11:55 |
*** psachin has joined #openstack-nova | 11:55 | |
mdbooth | Interwebs say no transactional ddl | 11:55 |
mdbooth | Think it would have to be a db lock, because we can't be sure where db sync is running. | 11:56 |
mdbooth | Also, think change would have to be in sqla migrations | 11:57 |
*** ducttape_ has quit IRC | 11:57 | |
*** bkopilov_ has quit IRC | 11:57 | |
*** bkopilov has quit IRC | 11:58 | |
johnthetubaguy | mdbooth: true, DB lock is safer | 11:58 |
*** archies has quit IRC | 11:59 | |
*** ducttape_ has joined #openstack-nova | 12:00 | |
*** prateek has quit IRC | 12:00 | |
*** Hosam has joined #openstack-nova | 12:01 | |
*** ratailor has quit IRC | 12:02 | |
*** openstackgerrit has quit IRC | 12:03 | |
*** edmondsw has joined #openstack-nova | 12:04 | |
*** hfu has quit IRC | 12:09 | |
*** hfu has joined #openstack-nova | 12:09 | |
*** hfu has quit IRC | 12:09 | |
*** sdague has joined #openstack-nova | 12:09 | |
*** hfu has joined #openstack-nova | 12:10 | |
owalsh | mdbooth: DB lock might be overkill... | 12:10 |
*** hfu has quit IRC | 12:10 | |
*** jpena is now known as jpena|lunch | 12:10 | |
*** karimb has quit IRC | 12:10 | |
owalsh | the first db sync succeeded IIRC | 12:11 |
mdbooth | owalsh: I don't think so. It would only have to used for exclusion of the migration itself. | 12:11 |
mdbooth | I think it could be pretty simple to implement, tbh. | 12:11 |
*** hfu has joined #openstack-nova | 12:11 | |
*** openstackgerrit has joined #openstack-nova | 12:12 | |
openstackgerrit | Derek Higgins proposed openstack/nova master: Simplify uses of assert_has_calls https://review.openstack.org/428108 | 12:12 |
*** hfu has quit IRC | 12:12 | |
derekh | johnthetubaguy: sfinucan ^^ sorry forgot to push it up yesterday | 12:12 |
mdbooth | When you initially read the current version from the migrate_version table you'd take an exclusive lock on it. | 12:12 |
*** hfu has joined #openstack-nova | 12:12 | |
*** hfu has quit IRC | 12:12 | |
mdbooth | Then you'd exclude any other migration from starting until this once completed, but other db operations can continue. | 12:12 |
*** clayton has joined #openstack-nova | 12:13 | |
sfinucan | derekh: Sweet. Cheers :) Think it can/should wait until Ocata is cut though | 12:13 |
derekh | sfinucan: sounds fine by me, just letting ye know its there | 12:13 |
*** dave-mccowan has joined #openstack-nova | 12:14 | |
*** Hosam has quit IRC | 12:16 | |
*** bvanhav has joined #openstack-nova | 12:18 | |
*** panda is now known as panda|lch | 12:18 | |
owalsh | mdbooth: hmm the oslo.db tests used to lock - https://review.openstack.org/58830 | 12:20 |
mdbooth | owalsh: A good reason to use a db lock | 12:20 |
*** ociuhandu has joined #openstack-nova | 12:21 | |
owalsh | but why protect the tests with a lock instead of the migration? | 12:21 |
* owalsh wonders if there was a good reason for this | 12:21 | |
mdbooth | owalsh: I think every test runs the migrations. | 12:21 |
*** Hosam has joined #openstack-nova | 12:21 | |
mdbooth | Unless... optimisation. | 12:22 |
mdbooth | But mostly. | 12:22 |
*** ducttape_ has quit IRC | 12:22 | |
owalsh | forget that, (should have looked at the diff), it's just locking the engine reset | 12:26 |
mdbooth | owalsh: I've emailed zzzeek for an opinion. cc'd you. | 12:27 |
*** READ10 has quit IRC | 12:31 | |
*** thorst_ has joined #openstack-nova | 12:33 | |
*** vsaienko has joined #openstack-nova | 12:34 | |
*** thorst_ has quit IRC | 12:34 | |
*** ayogi has quit IRC | 12:40 | |
*** ducttape_ has joined #openstack-nova | 12:42 | |
lyarwood | hopefully a stupid question, does the gate rebase changes? I keep hitting Bug 1660878 in https://review.openstack.org/#/c/427267/ and wonder if I need to manually rebase it to pick up https://review.openstack.org/#/c/427782/ | 12:42 |
openstack | bug 1660878 in OpenStack Compute (nova) "test_reboot_deleted_server fails with 409 "Cannot 'reboot' instance while it is in vm_state building"" [High,In progress] https://launchpad.net/bugs/1660878 - Assigned to Matt Riedemann (mriedem) | 12:42 |
*** sree has quit IRC | 12:46 | |
lyarwood | http://logs.openstack.org/67/427267/2/gate/gate-tempest-dsvm-neutron-full-ubuntu-xenial/e4e67b3/logs/devstacklog.txt.gz#_2017-02-02_10_50_57_699 - okay, so we merge the change, so I should have https://review.openstack.org/#/c/427782/ | 12:46 |
*** takashin has left #openstack-nova | 12:47 | |
*** jpena|lunch is now known as jpena | 12:50 | |
*** satyar has joined #openstack-nova | 12:51 | |
*** ducttape_ has quit IRC | 12:53 | |
openstackgerrit | John Garbutt proposed openstack/nova master: Report reserved_host_disk_mb in GB not KB https://review.openstack.org/428120 | 12:54 |
*** vsaienko has quit IRC | 12:55 | |
*** mvk has quit IRC | 12:55 | |
johnthetubaguy | bauzas: sfinucan: a nasty placement issue the openstack ansible guys were hitting ^ | 12:55 |
johnthetubaguy | just want to make sure I am not reading that incorrectly | 12:55 |
* johnthetubaguy goes for some lunch | 12:56 | |
*** panda|lch is now known as panda | 12:57 | |
*** tbachman has joined #openstack-nova | 12:58 | |
*** gcb has joined #openstack-nova | 13:00 | |
*** thorst_ has joined #openstack-nova | 13:01 | |
*** ducttape_ has joined #openstack-nova | 13:02 | |
mdbooth | lyarwood: Yep | 13:04 |
lyarwood | mdbooth: thanks, I'll wait for dansmith and mriedem to come online before I recheck https://review.openstack.org/#/c/427267/ again | 13:05 |
mdbooth | johnthetubaguy: Do you think you'll get around to https://review.openstack.org/#/c/383859/ before release? Fixes NFS, and makes doing the same for other filesystem-backed drivers trivial. See glusterfs follow-on for eg: https://review.openstack.org/#/c/423402/ | 13:08 |
gcb | sfinucan, mriedem: As I know we have been preparing for final octata release, Can we still merge python 3 unit tests patches now ? | 13:08 |
mdbooth | johnthetubaguy: So the NFS change is somewhat complex because it includes infrastructure applicable to all filesystem-backed volume backends. | 13:09 |
sfinucan | gcb: They're pretty low risk (unit test only) but I think we're only accepting critical bugfixes and doc patches at the moment | 13:09 |
mdbooth | It's pretty well tested, though. | 13:09 |
sfinucan | The Pike window will be open tomorrow though | 13:09 |
mdbooth | It's a bug. Are we passed the bugfix window? | 13:10 |
mdbooth | past | 13:10 |
mdbooth | That said, if I can get it into Pike I'll take it and propose for backport. | 13:10 |
mdbooth | I just want to land the damned thing. | 13:10 |
*** karimb has joined #openstack-nova | 13:10 | |
gcb | sfinucan: thanks, got it. The only two patches to make all unit tests work on Python 3.5 https://review.openstack.org/#/q/topic:bug/1482633 | 13:11 |
gcb | it's okay to make them merge into Pike :) | 13:11 |
sfinucan | gcb: Perfect. I'll take a look tomorrow or Monday :) Will be good to get that all closed off | 13:11 |
*** gouthamr has joined #openstack-nova | 13:11 | |
gcb | sfinucan: cool, thanks | 13:12 |
*** Hosam has quit IRC | 13:15 | |
sfinucan | johnthetubaguy: Took a gawk. Makes sense to me | 13:18 |
*** esberglu has joined #openstack-nova | 13:21 | |
*** jheroux has joined #openstack-nova | 13:22 | |
*** sree has joined #openstack-nova | 13:22 | |
jaypipes | morning folks | 13:24 |
bauzas | mriedem: FWIW, we have another related problem | 13:24 |
bauzas | johnthetubaguy: https://bugs.launchpad.net/nova/+bug/1659266 | 13:24 |
openstack | Launchpad bug 1659266 in OpenStack Compute (nova) "Disk allocation for instance is not good with swap" [High,Confirmed] - Assigned to Tanvir Talukder (tanvirt16) | 13:24 |
bauzas | but fixing that could be difficult given we already have allocations | 13:25 |
sfinucan | trinaths: Not sure if you got a response to your question or not. Have you tried building an instance with a NUMA topology but without hugepages? | 13:25 |
* sfinucan is reading through scrollback to find patches to review | 13:25 | |
mriedem | gcb: sfinucan: mdbooth: we shouldn't be putting things into the gate or check queue that don't need to be, py3 unit test cleanup included | 13:28 |
sfinucan | mriedem: Morning. That's what we thought | 13:28 |
gcb | mriedem: ack | 13:28 |
*** eharney has quit IRC | 13:29 | |
*** mvk has joined #openstack-nova | 13:29 | |
*** trinaths has quit IRC | 13:30 | |
mdbooth | lyarwood: I'm just going to grab coffee. Want to talk about swap_volume in a few minutes? | 13:31 |
lyarwood | mdbooth: sure | 13:31 |
mriedem | bauzas: is that related to the thing that johnthetubaguy has a patch up for? | 13:31 |
*** cdent has joined #openstack-nova | 13:32 | |
mriedem | https://review.openstack.org/#/c/428120/ | 13:32 |
sfinucan | mriedem, bauzas: That's the one, yes | 13:32 |
bauzas | mriedem: not really but the same problme | 13:33 |
bauzas | mriedem: honestly, MHO is that we should not use GBs for disk resources but rather bytes | 13:33 |
bauzas | or KB | 13:33 |
bauzas | if that's too big | 13:33 |
bauzas | but, we have lots of ceilings like the one I wrote | 13:33 |
mriedem | dansmith: melwitt: hmm, lyarwood's patch failed in the gate on the cells inst delete/create race thing http://logs.openstack.org/67/427267/2/gate/gate-tempest-dsvm-neutron-full-ubuntu-xenial/e4e67b3/console.html#_2017-02-02_11_48_07_755243 | 13:34 |
bauzas | mriedem: https://github.com/openstack/nova/blob/master/nova/scheduler/filter_scheduler.py#L148 | 13:34 |
mriedem | https://review.openstack.org/#/c/427782/ | 13:34 |
mriedem | which merged at 7:34pm | 13:35 |
mriedem | and lyarwood's change failed at 2017-02-02 11:48:07.755243 | 13:35 |
lyarwood | mriedem: twice, I think the previous run also hit this a few hours before that | 13:35 |
* lyarwood gets the link | 13:35 | |
mriedem | lyarwood: well i could see it hitting before the other fix was merged, | 13:36 |
mriedem | but i'm worried if it hit after the fix was merged | 13:36 |
mriedem | i think ES is backed up too so we probably can't tell if the hits have dropped off after the fix merged | 13:36 |
mriedem | yeah we can't tell http://status.openstack.org//elastic-recheck/index.html | 13:36 |
mriedem | everything is backed up | 13:36 |
lyarwood | mriedem: http://logs.openstack.org/67/427267/2/gate/gate-tempest-dsvm-neutron-full-ubuntu-xenial/cc0ed47/console.html#_2017-02-02_05_45_00_341186 was the other hit | 13:37 |
*** Hosam has joined #openstack-nova | 13:38 | |
*** diga has quit IRC | 13:38 | |
*** nmathew- has joined #openstack-nova | 13:38 | |
mriedem | yeah that's before the fix was merged | 13:39 |
mriedem | so not a surprise | 13:39 |
mriedem | but failing 3 hours after the merge worries me | 13:39 |
lyarwood | mriedem: ah did it land at 7:34am? | 13:39 |
mriedem | pm | 13:39 |
lyarwood | mriedem: you said pm above | 13:39 |
lyarwood | mriedem: that's 05:45am | 13:39 |
mriedem | oh yeah | 13:40 |
lyarwood | anyway | 13:40 |
nmathew- | hi all | 13:41 |
nmathew- | can anybody point to a material on flavors | 13:41 |
mriedem | lyarwood: apparently the times in the zuul log don't match up with what's in the gerrit ui | 13:41 |
mriedem | nmathew-: http://docs.openstack.org/admin-guide/compute-flavors.html | 13:41 |
mriedem | docs.o.o has a search box | 13:42 |
nmathew- | mriedem, say I not enabling hugepages in flavors, I can stil enable that by kernel parameters to that vm kernel | 13:44 |
nmathew- | mriedem, what is the difference between specifying in flavors and not specifying in flavors? | 13:44 |
lyarwood | nmathew-: see the topic, if you have questions like that #openstack is a better place to ask :) | 13:44 |
nmathew- | ok | 13:45 |
nmathew- | oh sorry | 13:45 |
lyarwood | nmathew-: np at all | 13:45 |
mriedem | so looking at my fix for the delete/create race bug, | 13:46 |
mriedem | https://review.openstack.org/#/c/427775/1/nova/conductor/manager.py | 13:46 |
mriedem | i see that debug log message hit 3 times in the n-cond logs on a run of that patch http://logs.openstack.org/75/427775/1/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/c906c15/logs/screen-n-cond.txt.gz#_2017-02-01_16_14_03_273 | 13:46 |
mriedem | so it's obviously catching the issue | 13:46 |
mriedem | dansmith: bauzas: ^ so i'm thinking we should probably just get that in given the time | 13:46 |
*** annegentle has joined #openstack-nova | 13:47 | |
*** zhurong has joined #openstack-nova | 13:48 | |
*** owalsh is now known as owalsh-brb | 13:51 | |
*** satyar has quit IRC | 13:51 | |
*** pbandark has quit IRC | 13:52 | |
cdent | mriedem: unless you're already on it, I'll fix the nits in https://review.openstack.org/#/c/427367 | 13:53 |
mriedem | i'll get it quick | 13:53 |
*** burt has joined #openstack-nova | 13:54 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: add upgrade notes to the placement devref https://review.openstack.org/427367 | 13:55 |
mriedem | sfinucan: bauzas: ^ fixed the typo | 13:55 |
*** psachin has quit IRC | 13:55 | |
*** sree has quit IRC | 13:55 | |
*** sree has joined #openstack-nova | 13:56 | |
bauzas | mriedem: okay, let's merge https://review.openstack.org/#/c/427775/1 then | 13:56 |
*** sree has quit IRC | 13:56 | |
*** crushil has quit IRC | 13:56 | |
*** ducttape_ has quit IRC | 13:57 | |
sfinucan | mriedem: Job. Done. | 13:57 |
mriedem | i wonder if it would be better to destroy the build request right before creating the instance? | 13:57 |
mriedem | but, i don't really want to ponder that today | 13:57 |
bauzas | mriedem: possibly | 13:58 |
bauzas | mriedem: the main issue I see here is timing | 13:58 |
bauzas | so let's merge your change is quite impactless and think about the best approach for Pike | 13:58 |
johnthetubaguy | mriedem: FWIW, I think its best to create the instance first, so we don't "loose" the instance, although marking the build request in someway that stops deletes might make sense, maybe thats the cell mapping entry? | 14:00 |
johnthetubaguy | hmm, I can't 100% remember the ordering now | 14:01 |
johnthetubaguy | mriedem: this one the openstack ansible folks I spotted feels nasty, but we totally could back port it later, I believe | 14:02 |
*** eharney has joined #openstack-nova | 14:02 | |
mriedem | johnthetubaguy: https://review.openstack.org/#/c/428120/ right? | 14:02 |
mriedem | i left some comments in there | 14:02 |
mriedem | there is a bug report i think that fits the same thing already | 14:03 |
mriedem | i've marked it rc1 | 14:03 |
*** owalsh-brb is now known as owalsh | 14:03 | |
*** annegentle has quit IRC | 14:04 | |
*** amoralej is now known as amoralej|lunch | 14:05 | |
*** bvanhav_ has joined #openstack-nova | 14:05 | |
*** gcb has quit IRC | 14:05 | |
*** bvanhav has quit IRC | 14:05 | |
alex_xu | johnthetubaguy: I think gmann is right on https://review.openstack.org/#/c/428071, we agreed on we have filter which match the name in the API representation, but that will be done in the future with microversion. | 14:06 |
*** zhurong has quit IRC | 14:06 | |
*** eharney has quit IRC | 14:07 | |
*** mdrabe has joined #openstack-nova | 14:08 | |
*** Hosam has quit IRC | 14:10 | |
*** Hosam has joined #openstack-nova | 14:10 | |
johnthetubaguy | mriedem: the bug is different I think, the bug talks about the allocation, this my bug is about the inventory (due to the reservation) | 14:11 |
jroll | mriedem: fyi this one looks fun https://bugs.launchpad.net/nova/+bug/1661258 | 14:12 |
openstack | Launchpad bug 1661258 in OpenStack Compute (nova) "Deleted ironic node has an inventory in nova_api database" [Undecided,New] | 14:12 |
mriedem | jroll: damn | 14:13 |
jroll | it might actually be easy | 14:13 |
mriedem | jroll: will need cdent, edleafe and/or jaypipes to look at that one | 14:13 |
jroll | mriedem: thank you for pinging them :P | 14:14 |
mriedem | i was thinking the periodic update would remove the inventory for the removed node but vdrok said in the bug that it didn't | 14:14 |
cdent | joy | 14:14 |
*** Hosam has quit IRC | 14:14 | |
*** sree has joined #openstack-nova | 14:15 | |
*** READ10 has joined #openstack-nova | 14:15 | |
edleafe | mriedem: I can take a look at that | 14:15 |
jroll | right, I assume we're missing something like this for inventory https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6577 | 14:15 |
*** ducttape_ has joined #openstack-nova | 14:16 | |
*** zhurong has joined #openstack-nova | 14:17 | |
mriedem | jroll: yeah that looks reasonable | 14:18 |
mriedem | jaypipes refactored a ton of that code so he'll know best | 14:19 |
*** crushil has joined #openstack-nova | 14:19 | |
jaypipes | jroll, vdrok, mriedem, edleafe, cdent: I have a suspicion on cause of that bug. | 14:19 |
*** eharney has joined #openstack-nova | 14:19 | |
*** bkopilov_ has joined #openstack-nova | 14:19 | |
edleafe | jaypipes: do tell | 14:19 |
*** bkopilov has joined #openstack-nova | 14:19 | |
*** tbachman has quit IRC | 14:20 | |
*** tbachman_ has joined #openstack-nova | 14:20 | |
*** nmathew has joined #openstack-nova | 14:20 | |
jaypipes | jroll, vdrok, mriedem, edleafe, cdent: Ironic virt driver is probably returning False for node_is_available() and therefore when update_available_resource() runs, we never even get to the call for InstanceList.get_by_host_and_node() and therefore never clean up the now-deleted instance. | 14:20 |
mriedem | sdague: how do you feel about this? https://review.openstack.org/#/c/426926/ | 14:20 |
jaypipes | mriedem: remember when you asked me to change RT.disabled(nodename) to account for the Ironic-only node_is_available() call? :) | 14:20 |
mriedem | sdague: because i'd like to get the dependent nova change in for ocata | 14:20 |
mriedem | jaypipes: vaguely | 14:21 |
*** nmathew- has quit IRC | 14:21 | |
mriedem | jaypipes: i'm pretty sure i was drunk reviewing most of that refactor series :) | 14:21 |
jroll | jaypipes: this is about deleted nodes, not deleted instances, to be clear | 14:21 |
jroll | jaypipes: so it wouldn't even be returned in get_available_nodes() | 14:22 |
jaypipes | vdrok: I asked for you to do a flash-patch in the bug above. | 14:22 |
cdent | is running two nova-computes on the same machine normal and/or an important consideration in this bug? | 14:22 |
jaypipes | jroll: right, but the inventory won't get cleaned up in the placement DB if the node_is_available() returns False here: | 14:23 |
vdrok | jaypipes: ok, thanks, will do after lunch | 14:23 |
jaypipes | https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L414 | 14:23 |
*** esberglu has quit IRC | 14:23 | |
jaypipes | cdent: no. | 14:23 |
sdague | mriedem: yeh, that seems fine. | 14:23 |
sdague | mriedem: under what circumstances does that error out? | 14:23 |
jroll | jaypipes: yeah, it'll definitely return False there if the node is deleted | 14:24 |
mriedem | sdague: would have to dig into nova-status code for specifics, but things like not being able to talk to the placement api | 14:24 |
*** Jeffrey4l has quit IRC | 14:24 | |
sdague | mriedem: yeh, seems fine | 14:24 |
mriedem | no cell mappings, no cell0, no host mappings when there are computes, etc | 14:25 |
mriedem | thanks | 14:25 |
edleafe | jroll: So if the node is deleted, the problem is just orphaned invntory, right? IOW, it isn't in the DB anymore, so it won't get returned from a placement GET | 14:26 |
jroll | edleafe: AIUI it *is* being returned from the placement API | 14:26 |
*** psachin has joined #openstack-nova | 14:26 | |
mriedem | the error in the bug report was missing some details, like what the actual build failure was | 14:27 |
*** amoralej|lunch is now known as amoralej | 14:27 | |
jroll | mriedem: fault | {"message": "Node 6cc8803d-4e77-4948-b653-663d8d5e52b7 could not be found. (HTTP 404)", "code": 500, "details": " File \"/opt/stack/nova/....... | 14:27 |
*** ducttape_ has quit IRC | 14:27 | |
*** mlavalle has joined #openstack-nova | 14:27 | |
jroll | right at the top | 14:27 |
mriedem | yeah, but | 14:27 |
mriedem | that doesn't really have anything to do with placement... | 14:28 |
mriedem | it's in _build_and_run_instance | 14:28 |
*** satyar has joined #openstack-nova | 14:28 | |
jroll | well | 14:28 |
jroll | it selected a node that doesn't exist in ironic | 14:28 |
jroll | it also doesn't exist in the compute_nodes table, but does exist in inventory | 14:29 |
mriedem | unless it's blowing up here https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L1905 | 14:29 |
*** udesale has joined #openstack-nova | 14:29 | |
*** liverpooler has joined #openstack-nova | 14:30 | |
*** nmathew- has joined #openstack-nova | 14:30 | |
edleafe | jroll: so it's gone from the compute_nodes table in the cell, but not in the resource_providers table in the API/placement db | 14:30 |
edleafe | jroll: So placement still thinks it's there | 14:30 |
jroll | edleafe: AIUI, yes | 14:30 |
*** liverpooler has quit IRC | 14:30 | |
jroll | keep in mind I did not file this bug | 14:30 |
edleafe | jroll: ack | 14:30 |
jroll | I have not independently reproduced it | 14:30 |
*** liverpooler has joined #openstack-nova | 14:30 | |
jroll | I'm just trying to answer question while vdrok is at lunch :) | 14:31 |
vdrok | jaypipes: so, I tried that, return (nodename not in self.compute_nodes) only in the disabled, restarted both computes, still the same | 14:31 |
jroll | mriedem: nah, I think all the placement/allocations stuff is working fine, except that the node being selected doesn't exist, so spawn() explodes | 14:31 |
*** prateek has joined #openstack-nova | 14:32 | |
mriedem | what is selecting the node? the scheduler? | 14:32 |
jroll | I assume so | 14:33 |
jroll | that's what selects things, right | 14:33 |
edleafe | jroll: jaypipes: so maybe we need to also add a call to the placement API after https://github.com/openstack/nova/blob/master/nova/objects/compute_node.py#L324 to delete the compute node as a resource provider? | 14:33 |
vdrok | mriedem: I added a bit more log from the n-cpu | 14:33 |
mriedem | i guess if everything is fully master then the filter scheduler will use placement | 14:33 |
*** sdague has quit IRC | 14:34 | |
*** nmathew has quit IRC | 14:34 | |
*** baoli has joined #openstack-nova | 14:34 | |
jroll | yeah, looks recent enough for placement api, that's why I bring it up :) | 14:34 |
*** sandanar has joined #openstack-nova | 14:35 | |
jroll | edleafe: that seems sensible to my feeble brain | 14:35 |
mriedem | edleafe: then the compute node object has to use the scheduler client... | 14:35 |
vdrok | now going to lunch for real :) | 14:35 |
mriedem | i'd rather we just call the scheduler client to delete the RP after deleting the CN here https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6585 | 14:35 |
mriedem | since the compute manager already has it's grubby hands into that code | 14:35 |
*** Jeffrey4l has joined #openstack-nova | 14:36 | |
edleafe | mriedem: yeah, makes sense | 14:36 |
*** psachin has quit IRC | 14:37 | |
*** zzzeek has quit IRC | 14:40 | |
*** zzzeek has joined #openstack-nova | 14:40 | |
mriedem | as far as i can tell, we only ever delete inventories or allocations, but never the actual resource provider | 14:40 |
mriedem | and since the CN is the RP, and we delete the CN, i guess we need to also delete the RP | 14:40 |
*** jamesdenton has joined #openstack-nova | 14:41 | |
mriedem | i'm going to head into the office, bbiab | 14:41 |
*** mriedem has quit IRC | 14:41 | |
cdent | yes | 14:41 |
cdent | (the delete the rp) | 14:42 |
*** Jeffrey4l has quit IRC | 14:42 | |
*** tblakes has joined #openstack-nova | 14:43 | |
*** crushil has quit IRC | 14:43 | |
*** crushil has joined #openstack-nova | 14:44 | |
*** Jeffrey4l has joined #openstack-nova | 14:44 | |
*** m1dev has joined #openstack-nova | 14:44 | |
*** esberglu has joined #openstack-nova | 14:45 | |
*** nmathew- has quit IRC | 14:46 | |
*** nmathew has joined #openstack-nova | 14:46 | |
*** krtaylor has quit IRC | 14:48 | |
*** nmathew has quit IRC | 14:52 | |
*** hongbin has joined #openstack-nova | 14:52 | |
mdbooth | lyarwood: I've reviewed the swap_volume change. Haven't looked at the tests yet, btw. Doing that now. | 14:56 |
*** m1dev has quit IRC | 14:58 | |
lyarwood | mdbooth: thanks again, working on it now. | 14:59 |
*** udesale has quit IRC | 15:00 | |
*** prateek has quit IRC | 15:02 | |
*** sudipto_ has joined #openstack-nova | 15:03 | |
*** sudipto has joined #openstack-nova | 15:03 | |
*** sree_ has joined #openstack-nova | 15:04 | |
*** nmathew has joined #openstack-nova | 15:04 | |
*** sree_ is now known as Guest15586 | 15:04 | |
*** baoli has quit IRC | 15:05 | |
*** sree has quit IRC | 15:06 | |
*** karts has joined #openstack-nova | 15:06 | |
*** jamesden_ has joined #openstack-nova | 15:06 | |
*** mriedem has joined #openstack-nova | 15:07 | |
*** sridharg has quit IRC | 15:07 | |
*** karthiks has quit IRC | 15:07 | |
mriedem | please all be fixed, please all be fixed... | 15:07 |
mriedem | no whammies | 15:07 |
*** jamesdenton has quit IRC | 15:07 | |
lyarwood | mriedem: my fix is failing in check, looking now | 15:07 |
*** jamesden_ has quit IRC | 15:08 | |
*** zz_dimtruck is now known as dimtruck | 15:09 | |
*** zhurong has quit IRC | 15:09 | |
*** jamesdenton has joined #openstack-nova | 15:10 | |
*** dikonoor has quit IRC | 15:10 | |
*** karts has quit IRC | 15:10 | |
lyarwood | mriedem: http://logs.openstack.org/67/427267/2/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/0f048f1/console.html#_2017-02-02_15_03_33_671655 - sorry if I've missed a previous discussion about this still being broken btw | 15:10 |
*** sridhargaddam has joined #openstack-nova | 15:11 | |
mriedem | lyarwood: another fix for that is in the gate | 15:12 |
lyarwood | mriedem: ah great, sorry I missed that fly past | 15:13 |
*** mtanino has joined #openstack-nova | 15:13 | |
mriedem | lyarwood: this one https://review.openstack.org/#/c/427775/ | 15:13 |
mriedem | 2 in the gate | 15:13 |
*** ducttape_ has joined #openstack-nova | 15:14 | |
* lyarwood fires up openstack-gerrit-dashboard/dash.py to watch | 15:14 | |
*** johndperkins has quit IRC | 15:17 | |
*** baoli has joined #openstack-nova | 15:17 | |
mriedem | do we have a definition of barber-pole sorting anywhere? | 15:17 |
*** dave-mccowan has quit IRC | 15:18 | |
*** mptacekx has quit IRC | 15:18 | |
mriedem | just means it wraps around from cell to cell right? | 15:18 |
*** Guest15586 has quit IRC | 15:19 | |
mriedem | "striped across the cell boundaries" | 15:19 |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix typo in cells v2 ocata reno https://review.openstack.org/428198 | 15:19 |
*** sdague has joined #openstack-nova | 15:21 | |
*** sridhargaddam has quit IRC | 15:21 | |
*** sridharg has joined #openstack-nova | 15:21 | |
jaypipes | edleafe: when (if ever) do compute_nodes table records get deleted? | 15:22 |
*** karts has joined #openstack-nova | 15:22 | |
*** kfarr has joined #openstack-nova | 15:22 | |
jaypipes | mriedem: more generally, I wonder when we ever delete compute nodes... | 15:24 |
*** tbachman_ is now known as tbachman | 15:24 | |
mriedem | in the case of ironic, i guess when they stop reporting available nodes | 15:24 |
mriedem | jaypipes: also when the nova-compute service is deleted via the api i think | 15:25 |
mriedem | will have to look | 15:25 |
openstackgerrit | Balazs Gibizer proposed openstack/nova master: Improve assertJsonEqual error reporting https://review.openstack.org/428199 | 15:25 |
*** baoli has quit IRC | 15:25 | |
mriedem | i remember fixinga 404 in the api around that recently | 15:25 |
jaypipes | mriedem: that would be a big mistake in the case of Ironic then... (since service != compute_node for Ironic) | 15:25 |
mriedem | https://review.openstack.org/#/c/406627/ was the 404 i was thinking of | 15:26 |
mriedem | "Compute node resources must currently be deleted manually in the database" | 15:26 |
*** baoli has joined #openstack-nova | 15:27 | |
edleafe | jaypipes: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6585 | 15:27 |
mriedem | but we do have this https://github.com/openstack/nova/blob/13.1.2/nova/compute/manager.py#L6513 | 15:27 |
mriedem | yeah | 15:27 |
edleafe | jaypipes: which in turn calls https://github.com/openstack/nova/blob/master/nova/objects/compute_node.py#L324 | 15:27 |
jaypipes | mriedem: yes, and I had some strong words on that patch :) | 15:27 |
diana_clarke | sfinucan: I heard you like docs ;) https://review.openstack.org/#/c/411441/ | 15:27 |
diana_clarke | sfinucan: (but only if you're not otherwise busy) | 15:28 |
jaypipes | edleafe: doink. | 15:28 |
mriedem | http://www.thereluctanttecchie.com/openstack-removing-a-compute-node-in-icehouse/ | 15:28 |
jaypipes | edleafe: right on, brother, nice detective work. | 15:28 |
jaypipes | edleafe: we need to have the resource tracker do that, not the compute manager. | 15:28 |
jaypipes | edleafe: I can work on a patch. | 15:28 |
edleafe | jaypipes: already working on it | 15:29 |
jaypipes | edleafe: nice find. | 15:29 |
mriedem | " The fact that the crap hypervisor-list/hypervisor-show REST APIs are still manually attempting to conflate a ComputeNode with a Service is a problem that should be fixed entirely in the REST API, since it is the REST API that is borked." | 15:29 |
jaypipes | edleafe: oh? even better :) | 15:29 |
mriedem | those strong words? :) | 15:29 |
jaypipes | mriedem: heh, yes sir. | 15:29 |
*** sridharg has quit IRC | 15:29 | |
edleafe | jaypipes: but it's the manager that has to make the call. I don't see the benefit in adding the RT as an indirection layer | 15:29 |
*** sree has joined #openstack-nova | 15:29 | |
jaypipes | edleafe: because the RT has the placement client, not the manager. | 15:29 |
jaypipes | edleafe: and the RT has all the compute node objects now, not the manager. | 15:30 |
mriedem | the compute manager is the one looking for orphaned computes | 15:30 |
mriedem | not the RT | 15:30 |
edleafe | jaypipes: the manager has the scheduler client | 15:30 |
*** prateek has joined #openstack-nova | 15:30 | |
*** prateek has quit IRC | 15:30 | |
*** sdague has quit IRC | 15:30 | |
mriedem | unless you just move that into RT.update_available_resource | 15:30 |
jaypipes | mriedem: all of that update_available_resource() code in the manager should be in the RT, IMHO. | 15:30 |
sfinucan | diana_clarke: ack | 15:31 |
edleafe | jaypipes: we can do scheduler_client.report_client like we added to RT | 15:31 |
*** gabor_antal_ has joined #openstack-nova | 15:32 | |
*** gabor_antal has quit IRC | 15:32 | |
jaypipes | mriedem: jinx. | 15:32 |
mriedem | jaypipes: well, | 15:32 |
mriedem | it's a chicken/egg, | 15:32 |
mriedem | because we only call rt.update_available_resource(context, nodename) | 15:32 |
mriedem | for an available node | 15:32 |
mriedem | and in this case, it's a not available node | 15:32 |
mriedem | so orphaned and we delete the CN record | 15:32 |
mriedem | we could build a new method into the RT to remove it if that's cleaner | 15:32 |
edleafe | mriedem: that sounds clunky to me | 15:32 |
edleafe | there's nothing magic about the client | 15:32 |
*** artom_ has quit IRC | 15:32 | |
mriedem | either way, i'm totally fine with a quick rt.report_client.delete_resource_provider from the compute manager in ocata if it gets the job done and saves time | 15:32 |
*** edmondsw_ has joined #openstack-nova | 15:32 | |
jaypipes | edleafe, mriedem: what I'm saying is that all the code in the compute manager's update_available_resource() method should be moved to the RT now. There's no reason to have that in the compute manager. | 15:32 |
mriedem | and a TODO to move that into RT later | 15:32 |
*** cleong has joined #openstack-nova | 15:32 | |
*** artom_ has joined #openstack-nova | 15:32 | |
mriedem | jaypipes: *now* is not the day to do it | 15:32 |
jaypipes | mriedem: sure, I'm also fine with a quick fix. | 15:32 |
jaypipes | mriedem: understood. | 15:32 |
edleafe | mriedem: gimme a few to get something done | 15:32 |
jaypipes | edleafe: cheers | 15:33 |
edleafe | all this IRC chat is slowing me down :) | 15:33 |
*** sdague has joined #openstack-nova | 15:33 | |
*** prateek has joined #openstack-nova | 15:33 | |
jaypipes | mriedem: in that future cleanup, I will also change the InstanceList.get_all_by_host_and_nodename() to be an InstanceList.get_all_by_host() and fix the performance issues with the Ironic RT once and for all. | 15:33 |
*** breitz has quit IRC | 15:33 | |
mriedem | yeah i guess compute manager has self.scheduler_client = scheduler_client.SchedulerClient() | 15:33 |
jaypipes | mriedem: that was the end goal of moving all the compute_node -> compute_nodes handling in the RT. | 15:34 |
*** dave-mccowan has joined #openstack-nova | 15:34 | |
mriedem | jaypipes: ok | 15:34 |
*** breitz has joined #openstack-nova | 15:34 | |
*** edmondsw has quit IRC | 15:34 | |
*** konan_abhi has joined #openstack-nova | 15:34 | |
jaypipes | mriedem: in the meantime, lemme look at the latest patch from edleafe on the Ironic inventory management. | 15:34 |
*** abhishek_k has quit IRC | 15:35 | |
*** baoli has quit IRC | 15:35 | |
mriedem | the harder part of this change is probably going to be the method in the scheduler to remove the RP b/c we'll also have to make sure it doesn't have allocations and inventory in order to delete the RP, else it will 409 | 15:35 |
jaypipes | mriedem: yup | 15:36 |
*** sandanar has quit IRC | 15:36 | |
mriedem | diana_clarke: wow nice catch L( | 15:36 |
mriedem | :) | 15:36 |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix typo in cells v2 ocata reno https://review.openstack.org/428198 | 15:36 |
edleafe | mriedem: add 'force=True'? | 15:36 |
mriedem | edleafe: is that a thing in the REST API? | 15:37 |
jaypipes | edleafe: I assigned that bug to you, FYI | 15:37 |
edleafe | mriedem: nope - I was joking | 15:37 |
*** jaosorior has joined #openstack-nova | 15:37 | |
jaypipes | mriedem: no. | 15:37 |
edleafe | forgot the smiley | 15:37 |
*** baoli has joined #openstack-nova | 15:37 | |
jaypipes | edleafe: nah, it's easy enough to write code that looks at existence of allocations and deletes them, then inventory, then resource provider. | 15:37 |
jaypipes | heh | 15:37 |
mriedem | we already have methods to delete the allocations and inventory, so probably can just re-use those | 15:38 |
* mriedem crosses fingers | 15:38 | |
edleafe | jaypipes: it's the same code; doing it in the client just makes more HTTP calls | 15:38 |
edleafe | mriedem: yep, that's the plan | 15:38 |
jaypipes | mriedem: indeed. in fact, https://review.openstack.org/#/c/404472 has examples of using those exact calls :) | 15:38 |
mriedem | i'm going to need to see johnthetubaguy be happy with that before i can look at it probably | 15:39 |
jaypipes | edleafe: understood. I'd prefer to keep a single HTTP call to a single thing unless there's a strong argument not to. | 15:39 |
mriedem | jaypipes: we also have 2 bugs for disk inventory/allocation in the report client | 15:39 |
jaypipes | s/to a single thing/to do a single thing/ | 15:39 |
jaypipes | mriedem: links pls? | 15:39 |
mriedem | https://review.openstack.org/#/c/428120/ | 15:40 |
mriedem | https://bugs.launchpad.net/nova/+bug/1659266 | 15:40 |
openstack | Launchpad bug 1659266 in OpenStack Compute (nova) "Disk allocation for instance is not good with swap" [High,Confirmed] - Assigned to Tanvir Talukder (tanvirt16) | 15:40 |
edleafe | jaypipes: there is no HTTP method that will remove a RP and all of its associated invetory and allocations | 15:40 |
*** mdnadeem has quit IRC | 15:40 | |
*** dimtruck is now known as zz_dimtruck | 15:40 | |
edleafe | jaypipes: so adding one will mean a new microversion :( | 15:40 |
jaypipes | edleafe: I know, and I think that's appropriate. | 15:40 |
edleafe | I would certainly prefer to add it that way, but with RC1... | 15:41 |
jaypipes | edleafe: meaning DELETE /resource-providers/<UUID> should return a 409 if there are existing allocations and inventories. The user should be required to DELETE /allocations|inventories in order to perform the DELETE /r-p | 15:41 |
edleafe | jaypipes: of course | 15:41 |
*** zz_dimtruck is now known as dimtruck | 15:41 | |
jaypipes | edleafe: I don't care for implementing DELETE CASCADE in a single HTTP call. | 15:41 |
mriedem | let us rathole on that in ATL | 15:42 |
jaypipes | yup | 15:42 |
edleafe | jaypipes: well, that's why I said several HTTP calls | 15:42 |
edleafe | jaypipes: one to delete allocation, one to delete inventory, and the final to delete the RP | 15:42 |
jaypipes | edleafe: right, which is how things currently work. | 15:42 |
mriedem | think if we actually had hard constraints on the instances table and nova didn't auto delete all related things in code... | 15:43 |
*** annegentle has joined #openstack-nova | 15:43 | |
mriedem | *auto soft-delete... | 15:43 |
dims_ | mriedem : dansmith : melwitt : should ServersNegativeTestJSON have been fixed? (with https://review.openstack.org/#/c/427782)? | 15:43 |
dansmith | dims_: yes, but we have another patch in the queue for it | 15:43 |
mriedem | dims_: yes, but https://review.openstack.org/#/c/427775/ | 15:43 |
*** krtaylor has joined #openstack-nova | 15:44 | |
mriedem | sfinucan: bauzas: want to send it on https://review.openstack.org/#/c/428198/ ? | 15:44 |
dims_ | mriedem : dansmith : melwitt : gotcha, saw it pop back up. will track 427775. thanks (http://logs.openstack.org/43/408143/9/gate/gate-tempest-dsvm-neutron-full-ubuntu-xenial/eb7a75d/logs/testr_results.html.gz) | 15:44 |
*** nmathew has quit IRC | 15:45 | |
*** baoli has quit IRC | 15:45 | |
jroll | jaypipes: I'm putting this on a plaque "in the future I will fix the performance issues with the Ironic RT once and for all." | 15:45 |
dims_ | oh boy! | 15:46 |
jaypipes | jroll: heh. | 15:46 |
*** sree has quit IRC | 15:46 | |
*** tbachman has quit IRC | 15:47 | |
*** abalutoiu has joined #openstack-nova | 15:47 | |
*** prateek has quit IRC | 15:48 | |
*** baoli has joined #openstack-nova | 15:48 | |
openstackgerrit | Pushkar Umaranikar proposed openstack/nova master: [WIP] Placement api: Add json_error_formatter to defaults https://review.openstack.org/395194 | 15:49 |
*** annegentle has quit IRC | 15:51 | |
*** sree has joined #openstack-nova | 15:52 | |
*** jamesdenton has quit IRC | 15:52 | |
*** baoli has quit IRC | 15:54 | |
johnthetubaguy | mriedem: jaypipes: I can take a look at that in a few mins | 15:54 |
jaypipes | johnthetubaguy: actually, just reviewing you patch to fix the MB -> GB thing. running some tests on it locally, but overall the code looks fine (but also bauzas' identified issue with the allocation side also is a good find) | 15:55 |
openstackgerrit | Lance Bragstad proposed openstack/nova-specs master: Cleanup pike policy spec https://review.openstack.org/428222 | 15:56 |
bauzas | jaypipes: mriedem: FWIW I'm about to add a PTG bullet about discussing on modifying the disk resources so it's bytes | 15:57 |
lbragstad | johnthetubaguy ^ feel free to squash that into your original version (it's just cleanup to get your draft passing) | 15:57 |
johnthetubaguy | lbragstad: oh, cool, thanks | 15:57 |
*** marst has joined #openstack-nova | 15:58 | |
jaypipes | bauzas: I support the general idea. but would be a pain in the ass to migrate all the MEMORY_MB and DISK_GB inventory and allocation records to bytes... | 15:58 |
johnthetubaguy | lbragstad: feel free to just push to my change if you want, thats cool | 15:58 |
johnthetubaguy | jaypipes: great, thank you | 15:58 |
lbragstad | johnthetubaguy yep! also - you have a todo in there for the keystone model | 15:58 |
jaypipes | bauzas: in addition to the fact that MEMORY_MB and DISK_GB are standard resource classes. | 15:58 |
lbragstad | johnthetubaguy I can take that action item - but where do you want that documented? I felt weird putting keystone things in a nova specific spec | 15:58 |
*** armax has joined #openstack-nova | 15:58 | |
bauzas | jaypipes: exactly | 15:59 |
*** burgerk has joined #openstack-nova | 15:59 | |
bauzas | jaypipes: that's why I think we should be discussing about it during the PTG | 15:59 |
bauzas | jaypipes: or during a hangout | 15:59 |
bauzas | because yes, changing that could be a PITA | 15:59 |
jaypipes | cool with me. | 15:59 |
jaypipes | (the discussions) :) | 16:00 |
*** _ducttape_ has joined #openstack-nova | 16:00 | |
*** jamesdenton has joined #openstack-nova | 16:01 | |
*** _ducttape_ has quit IRC | 16:01 | |
edleafe | jaypipes: cdent just found a flaw in this approach | 16:02 |
edleafe | jaypipes: we can't delete allocations by RP or ID | 16:02 |
edleafe | jaypipes: only by consumer_id | 16:02 |
edleafe | jaypipes: and a consumer might have allocations across multiple RPs | 16:02 |
jaypipes | edleafe: yes, and? | 16:03 |
*** ducttape_ has quit IRC | 16:03 | |
dansmith | so after an evacuate, | 16:04 |
dansmith | we wouldn't want to delete allocations for not-our-instance-anymore | 16:04 |
johnthetubaguy | lbragstad: ah, sorry, I missundertood, got you now | 16:04 |
jaypipes | edleafe: should still be able to delete the resource provider if the deletion of the consumers on that resource provider (identified with an InstanceList.get_all_by_host()) is called to DELETE /allocations/<instance_uuid> | 16:05 |
lbragstad | johnthetubaguy yep - line 102 here https://review.openstack.org/#/c/427872/3/specs/pike/approved/policy-cleanup-pike.rst | 16:05 |
jaypipes | dansmith: yeah, we would. the evacuate would result in a *new* allocation on a new resource provider. we would definitely want to delete the allocation on the old one. | 16:06 |
*** stvnoyes has quit IRC | 16:06 | |
edleafe | jaypipes: yeah, that could work. | 16:06 |
*** ducttape_ has joined #openstack-nova | 16:06 | |
dansmith | jaypipes: edleafe just said you can only delete by consumer id, which is instance uuid right? | 16:06 |
*** stvnoyes has joined #openstack-nova | 16:06 | |
openstackgerrit | Balazs Gibizer proposed openstack/nova master: Improve assertJsonEqual error reporting https://review.openstack.org/428199 | 16:07 |
edleafe | dansmith: yeah, so to delete by compute node, you'd have to get all the instances for that node, and delete their allocations first | 16:08 |
jaypipes | dansmith: yes, what edleafe said. | 16:08 |
dansmith | edleafe: right, but after an evacuation we have allocations on two compute nodes for that instance, right? | 16:08 |
edleafe | dansmith: wouldn't the evacuation delete allocations? | 16:09 |
jaypipes | dansmith: I think what we're saying is that the evacuation process would delete the allocs on the source node before adding them in the dest node. | 16:09 |
*** mdrabe has quit IRC | 16:09 | |
dansmith | edleafe: how? allocations are only managed from the compute node's RT AFAIK | 16:09 |
dansmith | if we add that, then maybe that's okay | 16:10 |
dansmith | but I don't think it happens today | 16:10 |
sfinucan | diana_clarke: https://review.openstack.org/#/c/411441/2/nova/tests/unit/api/openstack/compute/test_cells.py | 16:10 |
* edleafe wonders why he started this without a new pot of coffee... | 16:10 | |
jaypipes | dansmith: no, does not happen today. | 16:10 |
dansmith | jaypipes: so today we probably already corrupt the allocations after an evacuate anyway I guess | 16:11 |
edleafe | so let me summarize to make sure I'm clear on the evacuation scenario | 16:11 |
jaypipes | dansmith: sounds like it. | 16:11 |
dansmith | jaypipes: we just allocate again on the new node, and then maybe when the old node comes up, we delete the new and old ones | 16:11 |
jaypipes | dansmith: sounds right. | 16:12 |
edleafe | when we evacuate an instance to a new host, the instance ID stays the same, so the allocations for that instance remain | 16:12 |
*** Oku_OS is now known as Oku_OS-away | 16:12 | |
dansmith | jaypipes: sounds wrong.. but accurate, unfortunately. | 16:12 |
jaypipes | dansmith: I mean, not right, but sounds like what currently would happen. | 16:12 |
dansmith | heh | 16:12 |
jaypipes | jinx :) | 16:12 |
edleafe | the new host for that instance would periodically update, see the new instance, and then re-allocate against the new RP | 16:12 |
jaypipes | dansmith: want to add a bug? or should I? | 16:12 |
dansmith | jaypipes: I have done nothing useful so far today, so let me | 16:12 |
jaypipes | edleafe: yes, eventually it would resolve itself... but still not ideal | 16:13 |
edleafe | If we then delete all allocations for that instance / consumer, there will be a window where there are no allocations for that instance | 16:13 |
dansmith | jaypipes: oh yeah I guess it will heal, although not after cdent's proposed lessening of the healing action I think | 16:13 |
dansmith | or at least, heal slower | 16:13 |
dansmith | jaypipes: but opens us up for a poor decision because we think that host is less loaded than it is | 16:13 |
cdent | dansmith: i've punted on that until we figure out how that stuff even works | 16:14 |
dansmith | cdent: yeah, I know, I'm just saying | 16:14 |
dansmith | cdent: either (a) this is a good reason to take the hit on the healing or (b) need to consider this kind of thing if/when we proceed with that | 16:14 |
cdent | I'd vote for b | 16:14 |
cdent | so we can get rid of some of the implictness and confusion when we talk about this stuff :) | 16:14 |
*** tbachman has joined #openstack-nova | 16:15 | |
dansmith | well, things like this being papered over because we heal aggressively is problematic I think, | 16:15 |
dansmith | because before we heal, we're making bad decisions | 16:15 |
*** ftersin has quit IRC | 16:15 | |
dansmith | which I think is an argument for (b), but.. | 16:15 |
cdent | ( )|( ) | 16:16 |
dansmith | um | 16:17 |
* dansmith squints to see the real picture | 16:17 | |
dansmith | a butt? | 16:17 |
cdent | a big ol butt | 16:17 |
dansmith | classy, cdent, classy | 16:18 |
*** annegentle has joined #openstack-nova | 16:18 | |
openstackgerrit | Maciej Szankin proposed openstack/nova master: conf: mark glance.api_servers opt as required https://review.openstack.org/428241 | 16:19 |
*** mdrabe has joined #openstack-nova | 16:19 | |
*** xinli has joined #openstack-nova | 16:20 | |
dansmith | edleafe: https://bugs.launchpad.net/nova/+bug/1661312 | 16:21 |
openstack | Launchpad bug 1661312 in OpenStack Compute (nova) "Evacuation will corrupt instance allocations" [Undecided,New] | 16:21 |
edleafe | dansmith: Thanks. | 16:22 |
lbragstad | johnthetubaguy i'm going to propose a keystone spec for policy in code and I'll get a follow on to that to address your keystone TODOs in https://review.openstack.org/#/c/427872 (then we can reassess) | 16:24 |
*** kberger has joined #openstack-nova | 16:25 | |
*** jamesden_ has joined #openstack-nova | 16:26 | |
*** jamesdenton has quit IRC | 16:28 | |
*** tlian has joined #openstack-nova | 16:28 | |
*** rcernin has quit IRC | 16:29 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix access_ip_v4/6 filters params for servers filter https://review.openstack.org/428071 | 16:30 |
mriedem | gmann: any idea if Kevin had an update for this https://review.openstack.org/#/c/421760/? | 16:35 |
mriedem | or if the release note was going to be split out? | 16:35 |
mriedem | b/c we probably don't have time to wait for china to wake up | 16:36 |
*** ducttape_ has quit IRC | 16:38 | |
*** ducttape_ has joined #openstack-nova | 16:39 | |
*** sree has quit IRC | 16:39 | |
*** baoli has joined #openstack-nova | 16:40 | |
*** chyka has joined #openstack-nova | 16:41 | |
*** tesseract has quit IRC | 16:41 | |
*** markus_z has quit IRC | 16:43 | |
*** browne has joined #openstack-nova | 16:47 | |
*** artom_ is now known as artom | 16:47 | |
*** baoli has quit IRC | 16:48 | |
*** annegentle has quit IRC | 16:48 | |
*** andreas_s has quit IRC | 16:49 | |
*** annegentle has joined #openstack-nova | 16:50 | |
openstackgerrit | Balazs Gibizer proposed openstack/nova master: Short circuit notifiations when not enabled https://review.openstack.org/428260 | 16:52 |
*** dikonoor has joined #openstack-nova | 16:52 | |
openstackgerrit | Balazs Gibizer proposed openstack/nova master: handle uninited fields in notification payload https://review.openstack.org/415857 | 16:54 |
*** tblakes has quit IRC | 16:55 | |
jaypipes | sdague, dims: new box, trying to run some py 3.5-specific things. getting this: http://paste.openstack.org/show/597397/. Any ideas? | 16:55 |
*** bvanhav__ has joined #openstack-nova | 16:56 | |
*** bvanhav_ has quit IRC | 16:56 | |
*** tlian has quit IRC | 16:58 | |
openstackgerrit | Artom Lifshitz proposed openstack/nova master: Pass APIVersionRequest to extensions https://review.openstack.org/425876 | 16:58 |
mriedem | jaypipes: read the note in tox.ini | 16:59 |
mriedem | you need to delete .testrepository | 16:59 |
jaypipes | ah. thx mriedem. | 16:59 |
*** dimtruck is now known as zz_dimtruck | 16:59 | |
alex_xu | mriedem: let me try to take care of https://review.openstack.org/#/c/421760 | 17:01 |
openstackgerrit | John Garbutt proposed openstack/nova master: [WIP] Placement api: Add json_error_formatter to defaults https://review.openstack.org/395194 | 17:02 |
*** liverpooler has quit IRC | 17:02 | |
ildikov | johnthetubaguy: mriedem: can you come to the Cinder-NOva meeting? | 17:02 |
*** liverpooler has joined #openstack-nova | 17:03 | |
*** hoonetorg has quit IRC | 17:03 | |
*** Apoorva has joined #openstack-nova | 17:03 | |
*** Apoorva has quit IRC | 17:03 | |
johnthetubaguy | ildikov: ah, thats moved to today then I guess | 17:04 |
*** hoonetorg has joined #openstack-nova | 17:04 | |
johnthetubaguy | ildikov: I have updated my calendar, but honestly I have to chase a few of these RC related patches | 17:04 |
mriedem | ildikov: busy with rc1 things | 17:05 |
*** tblakes has joined #openstack-nova | 17:05 | |
ildikov | johnthetubaguy: mriedem: yep, it got moved, but will ping you directly if something pops up then | 17:05 |
mriedem | alex_xu: thanks, i think at this point i'd rather just make the reno a separate patch | 17:06 |
mriedem | the api-ref change is too large and needs a much more detailed review | 17:07 |
alex_xu | mriedem: yea, agree with that, just working on separating the patch | 17:07 |
*** raunak has joined #openstack-nova | 17:08 | |
*** kfarr has quit IRC | 17:10 | |
*** mriedem1 has joined #openstack-nova | 17:17 | |
*** mriedem has quit IRC | 17:17 | |
*** mtanino_ has joined #openstack-nova | 17:18 | |
*** nic has joined #openstack-nova | 17:18 | |
*** kfarr has joined #openstack-nova | 17:19 | |
*** mtanino__ has joined #openstack-nova | 17:19 | |
*** mtanino has quit IRC | 17:19 | |
*** sree has joined #openstack-nova | 17:20 | |
*** mtanino_ has quit IRC | 17:22 | |
*** panda is now known as panda|bbl | 17:26 | |
*** raunak has quit IRC | 17:27 | |
*** raunak has joined #openstack-nova | 17:28 | |
openstackgerrit | Diana Clarke proposed openstack/nova master: Add offset & limit docs & tests https://review.openstack.org/411441 | 17:29 |
openstackgerrit | John Garbutt proposed openstack/nova master: Report reserved_host_disk_mb in GB not KB https://review.openstack.org/428120 | 17:33 |
*** alexpilo_ has joined #openstack-nova | 17:35 | |
*** dtp has joined #openstack-nova | 17:35 | |
*** alexpilotti has quit IRC | 17:37 | |
*** hoonetorg has quit IRC | 17:38 | |
*** ducttape_ has quit IRC | 17:39 | |
jaypipes | johnthetubaguy: lol between 2.7 and 3.5 float handling, eh? | 17:39 |
*** hoonetorg has joined #openstack-nova | 17:40 | |
*** nkrinner is now known as nkrinner_afk | 17:41 | |
*** karimb has quit IRC | 17:41 | |
*** annegentle has quit IRC | 17:43 | |
melwitt | jaypipes: mriedem1 and I were discussing yesterday, is everything already merged that handles the local disk for volume backed instances to treat it as 0? does it also request 0 disk during a rebuild, etc? | 17:43 |
*** mriedem1 is now known as mriedem | 17:43 | |
*** aarefiev is now known as aarefiev_afk | 17:44 | |
*** liangy has joined #openstack-nova | 17:45 | |
*** zz_dimtruck is now known as dimtruck | 17:45 | |
mriedem | mdbooth: i'm going through your nfs mount patch now | 17:46 |
mriedem | mdbooth: so don't drop off yet | 17:46 |
openstackgerrit | Alex Xu proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/428285 | 17:46 |
alex_xu | mriedem: ^ done | 17:46 |
mriedem | thanks | 17:46 |
openstackgerrit | Maciej Szankin proposed openstack/nova master: conf: mark glance.api_servers opt as required https://review.openstack.org/428241 | 17:46 |
alex_xu | np | 17:46 |
dansmith | melwitt: yeah, I think we should be good there | 17:46 |
*** hoonetorg has quit IRC | 17:47 | |
melwitt | sweet | 17:47 |
*** satyar_ has joined #openstack-nova | 17:48 | |
*** baoli has joined #openstack-nova | 17:48 | |
*** satyar has quit IRC | 17:49 | |
*** annegentle has joined #openstack-nova | 17:50 | |
*** lucasagomes is now known as lucas-afk | 17:51 | |
*** cdent has quit IRC | 17:52 | |
*** sudipto_ has quit IRC | 17:53 | |
*** sudipto has quit IRC | 17:53 | |
*** vsaienko has joined #openstack-nova | 17:53 | |
johnthetubaguy | jaypipes: the joys! | 17:53 |
johnthetubaguy | jaypipes: I have a concern on this ironic patch still: https://review.openstack.org/#/c/404472 | 17:54 |
johnthetubaguy | jaypipes: but something tells me I am miss understanding the resource tracker sequence | 17:54 |
*** Apoorva has joined #openstack-nova | 17:54 | |
*** mjura has quit IRC | 17:54 | |
diana_clarke | sfinucan: thanks a bunch for reviewing that! | 17:55 |
*** jose-phillips has joined #openstack-nova | 17:55 | |
*** lpetrut has quit IRC | 17:56 | |
*** liangy has quit IRC | 17:57 | |
*** owalsh is now known as owalsh-afk | 17:57 | |
*** baoli has quit IRC | 17:58 | |
*** hoonetorg has joined #openstack-nova | 17:59 | |
*** baoli has joined #openstack-nova | 17:59 | |
*** vsaienko has quit IRC | 17:59 | |
*** vsaienko has joined #openstack-nova | 18:00 | |
*** raunak has quit IRC | 18:01 | |
*** ralonsoh has quit IRC | 18:02 | |
sfinucan | diana_clarke: np :) | 18:02 |
mriedem | mdbooth: lyarwood: before i go further, some comments/questions inline https://review.openstack.org/#/c/383859/13/nova/virt/libvirt/volume/mount.py | 18:02 |
*** kfarr has quit IRC | 18:02 | |
*** owalsh-afk has quit IRC | 18:03 | |
*** derekh has quit IRC | 18:03 | |
*** raunak has joined #openstack-nova | 18:03 | |
*** vsaienko has quit IRC | 18:04 | |
*** vsaienko has joined #openstack-nova | 18:05 | |
*** sree has quit IRC | 18:05 | |
BobBall | johnthetubaguy / mriedem FYI I believe https://review.openstack.org/#/c/428180/ will fix the Citrix XenServer CI | 18:09 |
*** Swami has joined #openstack-nova | 18:09 | |
mriedem | cool | 18:10 |
BobBall | johnthetubaguy / mriedem: A dodgy Cinder commit meant that "older" versions of pvs would all fail (as they don't have the newly required option) - the check for the version was broken. | 18:10 |
BobBall | We will monitor the CI after this cinder commit is merged and let you know the results and when we'd like to re-start voting. | 18:11 |
mriedem | sure | 18:11 |
openstackgerrit | Maciej Szankin proposed openstack/nova master: conf: resolved TODOs in compute configuration https://review.openstack.org/426793 | 18:11 |
mriedem | i guess if LVM.supports_pvs_ignoreskippedcluster was just set to False | 18:11 |
mriedem | because of the old pvs | 18:11 |
mriedem | *lvm | 18:12 |
BobBall | Actually it was always returning true... LVM.supports_pvs_ignoreskippedcluster was *meant* to be reading a property with a method behind it (using self.supportsXXX) but LVM.supportsXXX is the method itself so that was not None. | 18:13 |
openstackgerrit | melanie witt proposed openstack/nova master: Add step-by-step examples for Cells V2 setup https://review.openstack.org/427981 | 18:13 |
BobBall | oh sorr I mis-read. | 18:13 |
BobBall | Couldn't "set" it because it's just a method on the class | 18:13 |
*** alexpilotti has joined #openstack-nova | 18:15 | |
*** cdent has joined #openstack-nova | 18:17 | |
*** dikonoor has quit IRC | 18:18 | |
*** alexpilo_ has quit IRC | 18:18 | |
*** alexpilotti has quit IRC | 18:20 | |
mriedem | mdbooth: lyarwood: done with that review, only real questions are in that mount module | 18:20 |
*** alexpilotti has joined #openstack-nova | 18:20 | |
jaypipes | johnthetubaguy: +W from me on you disk_gb patch. | 18:21 |
* mriedem gets lunch | 18:21 | |
*** unicell has joined #openstack-nova | 18:26 | |
*** unicell has quit IRC | 18:26 | |
*** unicell has joined #openstack-nova | 18:26 | |
*** yamahata has joined #openstack-nova | 18:27 | |
mriedem | is anyone working a fix for this? https://bugs.launchpad.net/nova/+bug/1659266 | 18:28 |
openstack | Launchpad bug 1659266 in OpenStack Compute (nova) "Disk allocation for instance is not good with swap" [High,Confirmed] - Assigned to Tanvir Talukder (tanvirt16) | 18:28 |
*** hieulq has joined #openstack-nova | 18:29 | |
mriedem | bauzas: ^ you reported it | 18:29 |
EmilienM | 2 questions : | 18:29 |
EmilienM | 1. Where [cells] section needs to be configured (on which services, eg: nova-api? etc) | 18:29 |
mriedem | (1) who is your daddy and (2) what does he do? | 18:29 |
mriedem | EmilienM: cells config is only for cells v1 | 18:30 |
mriedem | you don't care about it | 18:30 |
EmilienM | 2. Do we *need* cells/enable= true in Ocata, | 18:30 |
mriedem | no | 18:30 |
EmilienM | amoralej: ^ | 18:30 |
mriedem | [cells] is all about cells v1 | 18:30 |
EmilienM | mriedem: ok | 18:30 |
amoralej | thanks mriedem | 18:30 |
EmilienM | yeah thx | 18:30 |
* EmilienM goes back in debugging | 18:31 | |
mriedem | i'll update the config option help to make that more clear | 18:31 |
EmilienM | mriedem: we're unable to spawn VMs anymore | 18:31 |
EmilienM | maybe you'll know why | 18:31 |
EmilienM | http://logs.openstack.org/15/424915/8/check/gate-puppet-openstack-integration-4-scenario001-tempest-centos-7/b29f35b/logs/nova/nova-compute.txt.gz#_2017-02-02_12_57_00_754 | 18:31 |
amoralej | mriedem, we getg "Instance 6a26cec0-3769-44b6-851c-b77359897e07 could not be found." kind of errors in certain tempest tests | 18:32 |
EmilienM | mriedem: we're deploying trunk | 18:32 |
EmilienM | with UC | 18:32 |
johnthetubaguy | jaypipes: thanks! | 18:33 |
*** dharinic is now known as dharinic|lunch | 18:34 | |
*** mvk has quit IRC | 18:35 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Clarify the [cells] config option help https://review.openstack.org/428344 | 18:35 |
mriedem | melwitt: dansmith: ^ | 18:35 |
*** annegentle has quit IRC | 18:36 | |
mriedem | EmilienM: that trace says the bdm is invalid | 18:37 |
EmilienM | bdm? | 18:37 |
mriedem | EmilienM: because the volume isn't found http://logs.openstack.org/15/424915/8/check/gate-puppet-openstack-integration-4-scenario001-tempest-centos-7/b29f35b/logs/nova/nova-compute.txt.gz#_2017-02-02_12_56_57_274 | 18:37 |
mriedem | bdm == block_device_mapping | 18:37 |
EmilienM | Block Device Mapping | 18:37 |
EmilienM | ok | 18:37 |
EmilienM | well, it worked yesterday | 18:38 |
mriedem | EmilienM: did it work with devstack? :P | 18:38 |
johnthetubaguy | mriedem: I can put something up for that bug 1659266 | 18:38 |
openstack | bug 1659266 in OpenStack Compute (nova) "Disk allocation for instance is not good with swap" [High,Confirmed] https://launchpad.net/bugs/1659266 - Assigned to Tanvir Talukder (tanvirt16) | 18:38 |
* mriedem ducks | 18:38 | |
mriedem | johnthetubaguy: thanks | 18:38 |
EmilienM | mriedem: I was about to say it but well... | 18:38 |
johnthetubaguy | I have to run, soon, but lets see what can be done | 18:38 |
mriedem | johnthetubaguy: you're not allowed to leave | 18:38 |
EmilienM | mriedem: FYI we're delaying TripleO release because upgrades are not working currently from Newton to Ocata | 18:38 |
mriedem | sally will understand | 18:38 |
EmilienM | mriedem: we hope to solve blockers by next week :/ | 18:39 |
johnthetubaguy | mriedem: you haven't met the conductor, she is scary | 18:39 |
*** ducttape_ has joined #openstack-nova | 18:40 | |
mriedem | EmilienM: my wild guess would be something related to https://review.openstack.org/#/c/427782/ | 18:41 |
mriedem | which merged last night | 18:41 |
*** salv-orlando has joined #openstack-nova | 18:41 | |
mriedem | EmilienM: but would also mean you don't have things setup for cells v2 | 18:41 |
EmilienM | mriedem: do you run experimental job sometimes? | 18:42 |
EmilienM | it seems like you're not | 18:42 |
EmilienM | mriedem: I think we are pretty well covered for cells v2 but I'll need to confirm | 18:42 |
diana_clarke | EmilienM: I've run it, but I've yet to see it pass ;( | 18:43 |
*** vsaienko has quit IRC | 18:43 | |
EmilienM | diana_clarke: where? | 18:43 |
diana_clarke | EmilienM: for example: https://review.openstack.org/#/c/420186/ | 18:43 |
*** vsaienko has joined #openstack-nova | 18:44 | |
diana_clarke | (it's not recent, but people are running it) | 18:44 |
*** ducttape_ has quit IRC | 18:44 | |
EmilienM | diana_clarke: cool | 18:44 |
*** READ10 has quit IRC | 18:45 | |
dansmith | I run it too on occasion, but it takes a long time, fails often, and the logs are harder to sift through | 18:46 |
*** vsaienko has quit IRC | 18:46 | |
EmilienM | dansmith: ping me when it fails | 18:47 |
EmilienM | I'm happy to help | 18:47 |
openstackgerrit | John Garbutt proposed openstack/nova master: Stop swap allocations being wrong due to MB vs GB https://review.openstack.org/428352 | 18:47 |
EmilienM | dansmith: any idea about my bug mentioned above ^ | 18:48 |
dansmith | EmilienM: I looked a bit earlier, but I saw no debug logs to help track anything down | 18:48 |
EmilienM | dansmith: no debug on http://logs.openstack.org/15/424915/8/check/gate-puppet-openstack-integration-4-scenario001-tempest-centos-7/b29f35b/logs/nova/nova-compute.txt.gz#_2017-02-02_12_56_57_274 ? | 18:49 |
johnthetubaguy | jaypipes: mriedem: its a quick cut and refactor from the previous fix, hopefully, at least thats what the unit tests suggested: https://review.openstack.org/#/c/428352 | 18:49 |
EmilienM | dansmith: could it be related to cells again? | 18:49 |
dansmith | EmilienM: no, I was looking at another run that had no debug, periodic maybe? | 18:50 |
EmilienM | dansmith: yeah periodic but this one has debug ^ | 18:50 |
*** vsaienko has joined #openstack-nova | 18:50 | |
dansmith | EmilienM: ah, I hadn't seen that volume error, and no that's not cells related AFAIK | 18:51 |
openstackgerrit | John Garbutt proposed openstack/nova master: Stop swap allocations being wrong due to MB vs GB https://review.openstack.org/428352 | 18:51 |
EmilienM | dansmith: could it be related to https://review.openstack.org/#/c/427782/ ? | 18:52 |
*** karimb has joined #openstack-nova | 18:52 | |
*** annegentle has joined #openstack-nova | 18:52 | |
openstackgerrit | John Garbutt proposed openstack/nova master: DNM: hack ironic with resource providers https://review.openstack.org/426296 | 18:52 |
* johnthetubaguy runs to rehersal | 18:53 | |
dansmith | EmilienM: not the volume error for sure no | 18:53 |
EmilienM | well, if you have any idea so far, please let me know. We're stuck. | 18:54 |
*** vsaienko has quit IRC | 18:55 | |
openstackgerrit | Maciej Szankin proposed openstack/nova master: conf: add min parameter to scheduler opts https://review.openstack.org/428355 | 18:56 |
mriedem | EmilienM: so the instance not found you were talking about is like this http://logs.openstack.org/15/424915/8/check/gate-puppet-openstack-integration-4-scenario001-tempest-centos-7/b29f35b/console.html#_2017-02-02_13_04_48_204185 | 18:56 |
mriedem | in a bunch of the tempest tests | 18:57 |
*** vsaienko has joined #openstack-nova | 18:57 | |
mriedem | it looks like it's waiting for the server to be active, it's gone | 18:57 |
amoralej | EmilienM, dansmith, reported in https://bugs.launchpad.net/nova/+bug/1661360 | 18:58 |
openstack | Launchpad bug 1661360 in OpenStack Compute (nova) "tempest test fails with "Instance not found" error" [Undecided,New] | 18:58 |
jroll | johnthetubaguy: there's an experimental job for resource class stuff, so I was going to abandon https://review.openstack.org/#/c/426296/ | 18:58 |
jroll | johnthetubaguy: but can keep it around if you rather | 18:58 |
mriedem | EmilienM: i can put up a debug patch if you guys can depend on it | 18:58 |
amoralej | mriedem, i've reproduced locally, nova server returns "instance not found" but it's found in "nova list" command | 18:58 |
mriedem | amoralej: that's probably a race between when we have a build request and the actual instance, | 18:59 |
mriedem | the api returns either | 18:59 |
mriedem | since they both represent a server at different stages of the lifecycle | 18:59 |
EmilienM | mriedem: yes I can | 18:59 |
EmilienM | mriedem: you can run check experimental | 18:59 |
*** carthaca_ has quit IRC | 19:00 | |
*** sapcc-bot has quit IRC | 19:00 | |
dansmith | mriedem: what race? | 19:00 |
*** eharney has quit IRC | 19:00 | |
*** carthaca_ has joined #openstack-nova | 19:00 | |
*** dgonzalez_ has joined #openstack-nova | 19:00 | |
*** sapcc-bot has joined #openstack-nova | 19:00 | |
*** tpatzig_ has joined #openstack-nova | 19:00 | |
melwitt | mriedem: isn't that issue fixed by https://review.openstack.org/#/c/427775 this morning? | 19:01 |
dansmith | melwitt: there's no delete going on here, which is what ^ is for right? | 19:01 |
melwitt | oh. yeah the above is for delete | 19:01 |
*** ociuhandu has quit IRC | 19:01 | |
*** dgonzalez_ has quit IRC | 19:02 | |
*** tpatzig_ has quit IRC | 19:02 | |
openstackgerrit | Matt Riedemann proposed openstack/nova master: WIP: debugging for instance not found in tripleo https://review.openstack.org/428360 | 19:04 |
mriedem | EmilienM: amoralej: ^ | 19:04 |
cdent | edleafe: do you recall if this allocation checking is happening anywhere else because this seems important but is stuck behind something else: https://review.openstack.org/#/c/416752/ | 19:04 |
mriedem | melwitt: dansmith: even with my fix, | 19:05 |
mriedem | couldn't we have a window of time where the build request exists, the instance exists, but the instance mapping doesn't? | 19:05 |
dansmith | mriedem: sure, but then we return the buildrequest, that's the point | 19:05 |
dansmith | mriedem: in their logs, I do not see "Found 1 cells" | 19:05 |
edleafe | cdent: it's not really "stuck"; just hasn't gotten core attention | 19:05 |
*** trevormc has joined #openstack-nova | 19:06 | |
*** raunak has quit IRC | 19:06 | |
dansmith | mriedem: I wonder if it's the change from melwitt that skips looking in all cells -- maybe incorrectly following that path? | 19:06 |
cdent | edleafe: what I meant was that fix is +W and something we ought to release, but the other thing less so | 19:06 |
*** Sukhdev has joined #openstack-nova | 19:06 | |
mriedem | dansmith: so maybe it hits a window when the api code sees there is no instance mapping, but then by the time it pulls the build request, that's gone | 19:06 |
edleafe | but yeah, it's not technically dependent on the one below it | 19:06 |
mriedem | that would be tight | 19:06 |
dansmith | mriedem: the api isn't even looking at the cells, though | 19:07 |
melwitt | that's only if CONF.cells.enable is True though | 19:07 |
edleafe | cdent: the other thing is as important | 19:07 |
dansmith | melwitt: I know, but | 19:07 |
edleafe | the create() method allows all sorts of bad stuff in | 19:07 |
EmilienM | mriedem: "check experimental" and we'll see what happens | 19:07 |
amoralej | I'm testing locally | 19:07 |
dansmith | melwitt: where is this in their log? https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2518 | 19:07 |
*** _gryf has quit IRC | 19:08 | |
*** _gryf has joined #openstack-nova | 19:08 | |
*** lpetrut has joined #openstack-nova | 19:08 | |
*** dharinic|lunch is now known as dharinic | 19:08 | |
cdent | edleafe: but it's not wired into the api is it? that's what I'm trying to get at. the patch above prevents the api being used badly, the patch below removes something that was never exposed | 19:08 |
dansmith | cells.enable=False in their startup dump | 19:08 |
melwitt | hm | 19:09 |
dansmith | melwitt: see what I mean though? | 19:09 |
melwitt | yeah | 19:09 |
dansmith | compare to a devstack run: http://logs.openstack.org/03/427903/1/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/fb34df9/logs/screen-n-api.txt.gz#_2017-02-01_21_00_47_552 | 19:10 |
*** trevormc_ has joined #openstack-nova | 19:10 | |
edleafe | cdent: true | 19:10 |
*** trevormc_ has left #openstack-nova | 19:10 | |
edleafe | cdent: but rebasing on master will lose the +W | 19:10 |
dansmith | now it's possible that they have never had it right, and removing those pre-cellsv2 short circuits are why they're failing now, but that's exactly why those have to be gone | 19:10 |
melwitt | dansmith: I'm also wondering about your change that changed the fallbacks into raises of InstanceNotFound | 19:10 |
cdent | edleafe: so the question is: how much of a bug is it that it is possible to write negative allocations? | 19:10 |
*** trevormc has quit IRC | 19:10 | |
* cdent tries it | 19:10 | |
*** raunak has joined #openstack-nova | 19:11 | |
dansmith | melwitt: right, that's what I just mentioned, but we *have* to remove those else we'll paper over not having things | 19:11 |
melwitt | yeah | 19:11 |
edleafe | cdent: negative not so much. Zero allocations was the issue | 19:11 |
*** trevormc has joined #openstack-nova | 19:11 | |
cdent | edleafe: by "tries it", I mean "will they get all the way through" | 19:11 |
cdent | if they get through, that seems a fair risk | 19:12 |
mriedem | dansmith: i'll throw this into the get instance flow in my debug patch https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2515-L2522 | 19:13 |
melwitt | dansmith: the 'Found cells' message is only for the get_all so that's why we're not seeing it I think. they're seeing a fail on a get of a single instance and said they see it in the list of all instances if they do a 'nova list' | 19:14 |
mriedem | right | 19:14 |
dansmith | melwitt: which is a get all | 19:14 |
mriedem | so i'll throw that into the debug patch | 19:14 |
mriedem | the test is failing on a single GET | 19:14 |
mriedem | not list | 19:14 |
melwitt | dansmith: yeah, it shows up in get_all but not in get | 19:14 |
mriedem | that's why we don't see it | 19:14 |
melwitt | is the problem they described | 19:14 |
melwitt | yeah | 19:14 |
dansmith | melwitt: I know, but they've done at least one get_all by that point right? | 19:14 |
mriedem | but, we should see it in the logs somewhere, | 19:14 |
mriedem | because tempest will list instances at some point | 19:14 |
dansmith | right | 19:14 |
melwitt | hm | 19:15 |
*** mlavalle has quit IRC | 19:15 | |
*** bvanhav__ has quit IRC | 19:15 | |
*** andrewbogott has quit IRC | 19:16 | |
*** bvanhav__ has joined #openstack-nova | 19:16 | |
melwitt | it's possible it didn't yet, but let me see if I can locate a list all in the log | 19:16 |
*** karimb has quit IRC | 19:16 | |
dansmith | am I missing where our requests are logged in n-api log? | 19:17 |
*** karlamrhein has quit IRC | 19:17 | |
melwitt | I'm not seeing it them yet. this doesn't look usual to me | 19:17 |
*** karimb has joined #openstack-nova | 19:17 | |
dansmith | melwitt: http://logs.openstack.org/15/424915/8/check/gate-puppet-openstack-integration-4-scenario001-tempest-centos-7/b29f35b/logs/tempest.txt.gz#_2017-02-02_12_53_52_735 | 19:18 |
dansmith | melwitt: yeah, something seems wrong about it | 19:18 |
dansmith | melwitt: but the above should have triggered a list all right? | 19:18 |
jaypipes | johnthetubaguy: +2 | 19:18 |
melwitt | dansmith: no, I think that's a create. it will return the one instance, or the first instance if it's a multi create | 19:18 |
melwitt | that's a POST | 19:19 |
dansmith | oh you're right | 19:19 |
openstackgerrit | Ed Leafe proposed openstack/nova master: Delete a compute node's resource provider when node is deleted https://review.openstack.org/428375 | 19:19 |
edleafe | jaypipes: cdent: ^^ please review | 19:20 |
cdent | edleafe: yessir | 19:20 |
mriedem | there is no "200 GET https://[::1]:8774/v2.1/servers" in that log | 19:20 |
dansmith | how is that possible | 19:20 |
melwitt | this is very odd | 19:21 |
*** karlamrhein has joined #openstack-nova | 19:22 | |
*** andrewbogott has joined #openstack-nova | 19:22 | |
dansmith | mriedem: so your race... | 19:22 |
dansmith | mriedem: we look up the map, if we fail, we look up the buildreq, if that fails we re-check for the map | 19:22 |
dansmith | by the time we've deleted the buildreq, we've finished up the instance map and saved it | 19:23 |
dansmith | so I don't see that there is any window there | 19:23 |
openstackgerrit | Matt Riedemann proposed openstack/nova master: WIP: debugging for instance not found in tripleo https://review.openstack.org/428360 | 19:23 |
dansmith | are you thinking of a different place? | 19:23 |
mriedem | dansmith: i was thinking it's like this | 19:23 |
mriedem | 1. https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2271 we don't have an instance mappingyet | 19:24 |
mriedem | well, shit that's it | 19:24 |
mriedem | if there is no instance mapping, that doesn't lookup the build req | 19:24 |
mriedem | at all, right? | 19:24 |
mriedem | the conditionals in there are only caring if there is an instance mapping AND it has a cell mapping | 19:25 |
dansmith | but we create the instance map early | 19:25 |
melwitt | uh oh, looks tha tway | 19:25 |
dansmith | lemme look | 19:25 |
mriedem | we create the instance mapping after the instance is created, but before the build request is deleted | 19:25 |
dansmith | no | 19:25 |
dansmith | we update it | 19:25 |
dansmith | we create the instance mapping at almost the same time as the buildreq | 19:25 |
dansmith | although the buildreq is first by a few python cycles | 19:26 |
dansmith | this would be nanoseconds of a window | 19:26 |
mriedem | oh right | 19:26 |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L997 | 19:26 |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1010 | 19:26 |
dansmith | picoseconds even | 19:26 |
dansmith | yeah | 19:26 |
mriedem | i thought you just started making up small time units now | 19:26 |
mriedem | danoseconds | 19:26 |
dansmith | femtoseconds | 19:27 |
melwitt | smithoseconds | 19:27 |
mriedem | only Vek can say femto-anything | 19:27 |
dansmith | mriedem: so I can swap the order of those in a test patch to run experimental on, but I'm highly dubious | 19:27 |
dansmith | mriedem: Vek gave me the femtotorch | 19:27 |
mriedem | ha | 19:27 |
dansmith | it was a small and quick handoff | 19:27 |
mriedem | how appropriate | 19:28 |
melwitt | /facepalm | 19:28 |
mriedem | rimshot? | 19:28 |
*** ociuhandu has joined #openstack-nova | 19:28 | |
openstackgerrit | Dan Smith proposed openstack/nova master: DNM: Reorder two electrons to see if it fixes a race https://review.openstack.org/428385 | 19:30 |
melwitt | haha, like that commit message | 19:30 |
dansmith | so this probably comes from the logic in get_instance needing to be defensive against pre-cellsv2 data, so we could also just rearrange that and probably simplify | 19:30 |
*** vsaienko has quit IRC | 19:31 | |
*** ducttape_ has joined #openstack-nova | 19:32 | |
*** eharney has joined #openstack-nova | 19:37 | |
*** mtanino__ has quit IRC | 19:39 | |
*** kfarr has joined #openstack-nova | 19:52 | |
amoralej | mriedem, dansmith, i've been debugging by printing value of service_version right before https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2268 | 19:54 |
amoralej | and i'm getting allways 0 | 19:54 |
amoralej | in fact if y remove the "service_version < 15" part of the if, tempest test passes | 19:55 |
amoralej | how can i see the versions of each service? | 19:55 |
amoralej | EmilienM ^ | 19:55 |
dansmith | amoralej: orly | 19:55 |
dansmith | amoralej: maybe we're grabbing a cached version of zero before we start computes | 19:56 |
dansmith | not sure what would have changed around that specific thing recently though | 19:56 |
amoralej | i can try to remove the compute node and register it again | 19:56 |
dansmith | we really should never cache version=0, which I think we're probably doing right now | 19:58 |
*** raunak has quit IRC | 20:00 | |
*** annegentle has quit IRC | 20:00 | |
*** rfolco has quit IRC | 20:01 | |
*** panda|bbl is now known as panda | 20:02 | |
*** annegentle has joined #openstack-nova | 20:02 | |
*** vsaienko has joined #openstack-nova | 20:06 | |
*** salv-orlando has quit IRC | 20:09 | |
amoralej | dansmith, in nova database all services have version 16 | 20:10 |
amoralej | but nova_cell0 has no services | 20:10 |
mriedem | there are no hosts for nova_cell0 | 20:10 |
mriedem | cell0 is just a place to dump dead instances that didn't get built | 20:11 |
cdent | edleafe: my testing suggests there's no urgency, but it does suggest that checking the allocations is not necessary if we change (and validate) the default min_unit when creating inventory | 20:11 |
cdent | if that's done, then there's no need to check the allocations manually | 20:11 |
dansmith | amoralej: the problem is it may have 16 at the end, but might've had zero early such that it cached zero and never saw the 16 | 20:11 |
amoralej | dansmith, in that case it should work if i restart nova api? | 20:12 |
amoralej | i can try | 20:12 |
dansmith | amoralej: yeah | 20:12 |
openstackgerrit | Ed Leafe proposed openstack/nova master: Delete a compute node's resource provider when node is deleted https://review.openstack.org/428375 | 20:13 |
mriedem | vdrok: want to test this? https://review.openstack.org/#/c/428375/ | 20:14 |
openstackgerrit | Dan Smith proposed openstack/nova master: Don't ever cache service versions at zero https://review.openstack.org/428404 | 20:14 |
*** lpetrut has quit IRC | 20:15 | |
amoralej | dansmith, reboot didn't work | 20:16 |
amoralej | but | 20:16 |
*** kfarr has quit IRC | 20:16 | |
dansmith | amoralej: did you get all api workers on all nodes? | 20:16 |
amoralej | replacing nova-osapi_compute by nova-compute worked | 20:16 |
dansmith | oh jeez | 20:16 |
amoralej | so | 20:16 |
dansmith | where? | 20:16 |
amoralej | https://github.com/openstack/nova/blame/master/nova/compute/api.py#L2268 | 20:16 |
dansmith | cripes I see | 20:16 |
dansmith | jeezus | 20:16 |
amoralej | i mean https://github.com/openstack/nova/blame/master/nova/compute/api.py#L2264 | 20:17 |
dansmith | ffs | 20:17 |
dansmith | how. | 20:17 |
dansmith | many. | 20:17 |
dansmith | times. | 20:17 |
amoralej | is service-name configurable? | 20:17 |
dansmith | no | 20:17 |
dansmith | this is a bug | 20:17 |
dansmith | I dunno why we're not hitting it elsewhere | 20:17 |
amoralej | EmilienM ^ | 20:17 |
EmilienM | amoralej: nice catch! | 20:18 |
amoralej | dansmith, and https://github.com/openstack/nova/blame/master/nova/compute/api.py#L1707 | 20:18 |
dansmith | wait wait wai | 20:18 |
*** kfarr has joined #openstack-nova | 20:18 | |
dansmith | I think that's actually looking for the api service not the compute service | 20:19 |
dansmith | so it should be right | 20:19 |
melwitt | yeah, that should be right. it used to be wrong as 'nova-api' in the past, somewhere in compute/api I recall | 20:19 |
cdent | edleafe: I left more about the min_unit stuff on https://review.openstack.org/#/c/416752/7 | 20:19 |
dansmith | right, and I was thinking it was looking for nova-compute, which is what the change was | 20:19 |
dansmith | so I jumped | 20:19 |
amoralej | and where api services are registered? | 20:19 |
dansmith | amoralej: you changing it just means you're looking up a different service | 20:19 |
amoralej | i don't see it in services table | 20:19 |
amoralej | we are running it as wsgi | 20:20 |
dansmith | oh, hmm, I bet that is a problem | 20:20 |
edleafe | cdent: I can get to those changes today | 20:20 |
amoralej | dansmith, this is my services table http://paste.openstack.org/show/597426/ | 20:20 |
dansmith | amoralej: you see no service records in the services table for nova-osapi_compute? | 20:20 |
dansmith | yeah | 20:21 |
dansmith | because you don't use our service infrastructure | 20:21 |
dansmith | mriedem: we've never said we're okay to run under wsgi did we? | 20:21 |
mriedem | dansmith: we don't have anything official | 20:21 |
amoralej | EmilienM ^ | 20:21 |
mriedem | but i know people run it | 20:21 |
mriedem | i think Fuel did too | 20:21 |
cdent | edleafe: feel free, but I don't think there's any rush: because the RT sets min_unit correctly, any "proper" use of the placement api will work okay and a bad admin can't send 0 allocations to the inventory that's been written (because min_unit is 1) | 20:21 |
cdent | (if you do do it, add me to the review so it is on my radar so I know not to do it) | 20:22 |
dansmith | the thing is, I'm not sure what in the last day would have broken | 20:22 |
edleafe | cdent: Understood, but I hate building those checks into things external to placement | 20:22 |
mriedem | dansmith: even with your change, we would have hit this i think https://review.openstack.org/#/c/427782/3/nova/compute/api.py@2269 | 20:23 |
cdent | edleafe: oh yeah, totally agree | 20:23 |
EmilienM | nova gate is not running api in wsgi with apache? | 20:23 |
edleafe | cdent: because the next time some service tries to use placement, we have to copy/paste all that external stuff there, too | 20:23 |
mriedem | EmilienM: no | 20:23 |
* cdent nods | 20:23 | |
dansmith | mriedem: on wsgi, yeah | 20:23 |
EmilienM | why? | 20:23 |
mriedem | EmilienM: no one has made that happen | 20:23 |
dansmith | and we have things like this that hook into our service framework | 20:24 |
mriedem | apache support was hacked in several releases ago as a kind of 'i want to see if i can do this' but it wasn't like a supported new feature or anything | 20:24 |
amoralej | i'm trying to test it with nova-api | 20:24 |
mriedem | it was a little toy patch | 20:24 |
EmilienM | tripleo has been running it in productince since Newton | 20:25 |
EmilienM | production* | 20:25 |
*** crushil has quit IRC | 20:26 | |
dansmith | so few real-world users at this point I guess | 20:26 |
*** crushil has joined #openstack-nova | 20:26 | |
dansmith | dims_: since you were originally involved, any idea how many real people running fuel might be running nova api under apache? | 20:26 |
EmilienM | dansmith: few real-world users? It's in OSP10 | 20:27 |
dansmith | EmilienM: yeah, I dunno what the footprint of that looks like | 20:27 |
*** mlavalle has joined #openstack-nova | 20:28 | |
mriedem | jaypipes: "any idea how many real people running fuel might be running nova api under apache?" | 20:28 |
*** Apoorva has quit IRC | 20:28 | |
mwhahaha | mriedem: a bunch, it's in 9 believe | 20:29 |
* mwhahaha goes looking | 20:29 | |
mriedem | https://github.com/openstack/nova/blob/ed55dcad83d5db2fa7e43fc3d5465df1550b554c/nova/wsgi/nova-api.py#L15 | 20:30 |
mriedem | EXPERIMENTAL support script for running Nova API under Apache2 etc. | 20:30 |
mriedem | anywho | 20:30 |
jroll | experimental means battle-tested production grade software, right | 20:30 |
jroll | like nova-bm | 20:31 |
mriedem | carrier grade | 20:31 |
jroll | ++ | 20:31 |
cdent | I'm unclear why exporting deploy() as application is so fraught with risk? | 20:31 |
EmilienM | mriedem: yeah a lot of operators are doing that | 20:32 |
EmilienM | mriedem: I don't have numbers but it's something common | 20:32 |
*** kaisers_ has joined #openstack-nova | 20:32 | |
mriedem | cdent: like i said earlier, it was just a 'lets see if this can happen' thing long ago that we never built CI around | 20:32 |
jroll | cdent: I think the current theory is that it doesn't get any of this code: https://github.com/openstack/nova/blob/ed55dcad83d5db2fa7e43fc3d5465df1550b554c/nova/service.py#L139 | 20:32 |
melwitt | yeah, we didn't do it but thought we should have since things like the api rate limiting can't work as-is | 20:33 |
jroll | specifically line 154 | 20:33 |
mriedem | so i think we provide a workaround hack in ocata | 20:33 |
cdent | thank jroll that's what I was after | 20:33 |
mriedem | and work on running nova-api under apache in one of the gate jobs in pike | 20:33 |
jroll | cdent: it's overridden for the wsgi thing, I guess, but same idea: https://github.com/openstack/nova/blob/ed55dcad83d5db2fa7e43fc3d5465df1550b554c/nova/service.py#L376 | 20:33 |
jroll | no problem | 20:33 |
*** baoli has quit IRC | 20:34 | |
mwhahaha | so to answer, fuel doesn't use apache for the nova api. it's haproxy fronting the standalone service | 20:35 |
EmilienM | I think Nova team should consider moving the experimental tripleo job to the gate. We have seen over the last weeks (more probably) that it brings very useful feedback on how it works outside devstack | 20:35 |
*** salv-orlando has joined #openstack-nova | 20:35 | |
dansmith | mwhahaha: ah, good to know | 20:35 |
*** bvanhav__ has quit IRC | 20:35 | |
*** kaisers has quit IRC | 20:35 | |
*** baoli has joined #openstack-nova | 20:37 | |
EmilienM | mriedem: I think testing api service with apache is part of the goal for Pike | 20:37 |
mriedem | which goal? | 20:37 |
mriedem | how many goals are we up to now? | 20:37 |
mriedem | i said we can start running it in one gating job for nova in pike as a start | 20:37 |
EmilienM | in https://governance.openstack.org/tc/goals/pike/deploy-api-in-wsgi.html#completion-criteria | 20:38 |
*** mvk has joined #openstack-nova | 20:38 | |
mriedem | because we have a unicorn snowflake job for stuff like that already | 20:38 |
EmilienM | mriedem: 2 goals | 20:38 |
*** amoralej is now known as amoralej|off | 20:38 | |
mriedem | ok, so i guess it's a goal for us then anyway | 20:39 |
mriedem | i'll throw it in our ptg etherpad | 20:39 |
*** annegentle has quit IRC | 20:40 | |
*** Sukhdev has quit IRC | 20:40 | |
dims_ | dansmith : i don't think the fuel team ever picked it up from where we left off here in nova. so probably not very many (http://git.openstack.org/cgit/openstack/fuel-library/tree/deployment/puppet/openstack_tasks/manifests/openstack_controller/openstack_controller.pp) | 20:40 |
dims_ | jaypipes : amirite? | 20:41 |
dansmith | dims_: gotcha | 20:41 |
EmilienM | mriedem: I will propose to run tripleo job in the gate for nova | 20:42 |
mriedem | EmilienM: i will -1 | 20:42 |
EmilienM | of course you will | 20:42 |
EmilienM | but you'll have to explain why | 20:42 |
mriedem | because it's not even non-voting in the nova check queue for starters | 20:42 |
mriedem | we aren't going to gate on something right off the bat | 20:42 |
EmilienM | ok, let's make non voting first | 20:43 |
EmilienM | and see how it goes | 20:43 |
mriedem | are you talking 1 job or 6? | 20:43 |
EmilienM | but we need to stop ignoring the world outside devstack | 20:43 |
EmilienM | 1 job | 20:43 |
EmilienM | which takes 50 min max | 20:43 |
mriedem | wfm | 20:43 |
EmilienM | cool | 20:43 |
*** annegentle has joined #openstack-nova | 20:43 | |
openstackgerrit | Dan Smith proposed openstack/nova master: Don't ever cache service versions at zero https://review.openstack.org/428404 | 20:43 |
openstackgerrit | Dan Smith proposed openstack/nova master: Work around deployments with unsupported API services https://review.openstack.org/428415 | 20:43 |
dansmith | mriedem: ^ untested, but might workaround the issue for experimental and unsupported setups | 20:44 |
mriedem | yup | 20:44 |
*** hieulq has quit IRC | 20:48 | |
openstackgerrit | Dan Smith proposed openstack/nova master: Don't ever cache service versions at zero https://review.openstack.org/428404 | 20:48 |
openstackgerrit | Dan Smith proposed openstack/nova master: Clarify the language in the apache wsgi sample https://review.openstack.org/428417 | 20:48 |
dansmith | mriedem: also recommend this, since there's clearly confusion around the word "experimental" | 20:48 |
dansmith | and with that, I'm going to walk away for a bit. | 20:49 |
mriedem | jaypipes: https://review.openstack.org/#/c/428375/ | 20:49 |
EmilienM | mriedem: the tripleo job has finished to run, http://logs.openstack.org/60/428360/2/experimental/gate-tripleo-ci-centos-7-nonha-multinode/6ca334c/logs/subnode-2/var/log/nova/nova-compute.txt.gz | 20:49 |
EmilienM | mriedem: let me know if logs are not enough I can increase it | 20:50 |
EmilienM | mriedem: the error is here: http://logs.openstack.org/60/428360/2/experimental/gate-tripleo-ci-centos-7-nonha-multinode/6ca334c/logs/subnode-2/var/log/nova/nova-api.txt.gz#_2017-02-02_20_42_17_097 | 20:50 |
mriedem | wish i could filter these logs based on level | 20:51 |
mriedem | i expected to see different errors there | 20:51 |
mriedem | like real ERROR level issues | 20:51 |
edleafe | cdent: what about just changing objects.Allocation.used to a NonNegativeIntegerield? | 20:51 |
cdent | edleafe: I thought we established those didn't exist? | 20:52 |
cdent | edleafe: and also, in the past, even if we did have that we still wanted the trap at the schema level | 20:52 |
edleafe | cdent: doh! | 20:53 |
edleafe | NonNegative allows zero | 20:53 |
* edleafe feels especially dumb | 20:53 | |
*** takashin has joined #openstack-nova | 20:53 | |
mriedem | bauzas: if you're still around https://review.openstack.org/#/c/428352/ | 21:00 |
bauzas | mriedem: I'm just writing a comment :) | 21:00 |
*** harlowja has quit IRC | 21:00 | |
mriedem | great | 21:00 |
bauzas | mriedem: there is a py3 issue | 21:00 |
bauzas | but I'm verifying it | 21:01 |
mriedem | wouldn't the py3 job fail on that? | 21:01 |
bauzas | not sure | 21:01 |
mriedem | it's voting | 21:01 |
bauzas | mriedem: do we have meeting ? | 21:01 |
edleafe | meeting today? | 21:01 |
mriedem | crap | 21:01 |
mriedem | yes | 21:01 |
edleafe | bauzas: jinx | 21:01 |
mriedem | short one | 21:01 |
bauzas | mriedem: nevermind my comment on py3 | 21:04 |
*** salv-orlando has quit IRC | 21:05 | |
*** liverpooler has quit IRC | 21:06 | |
*** _ducttape_ has joined #openstack-nova | 21:06 | |
*** raunak has joined #openstack-nova | 21:09 | |
*** karimb has quit IRC | 21:09 | |
tonyb | mriedem: I get the quote I'm just not sure how to interpret it in the context of this discussion | 21:10 |
*** ducttape_ has quit IRC | 21:10 | |
*** karimb has joined #openstack-nova | 21:10 | |
*** raunak has quit IRC | 21:10 | |
*** dave-mccowan has quit IRC | 21:10 | |
mriedem | tonyb: that the reqs team can wait :) | 21:11 |
mriedem | and enjoy some cake while they are waiting | 21:11 |
tonyb | mriedem: :) | 21:11 |
tonyb | the cake is a lie! | 21:11 |
mriedem | unless it's mikal's birthday | 21:12 |
*** slaweq has quit IRC | 21:12 | |
*** raunak has joined #openstack-nova | 21:12 | |
*** catintheroof has quit IRC | 21:14 | |
*** catintheroof has joined #openstack-nova | 21:15 | |
*** pshedimb has joined #openstack-nova | 21:16 | |
pshedimb | If anyone has idea about multi node devstack (latest version) setup | 21:17 |
pshedimb | My comupte node is generating this error when creating an instance | 21:17 |
pshedimb | ERROR nova.compute.manager [req-392e02de-349a-4791-b390-4449ff466427 None None] No compute node record for host fmwpaxeon39 | 21:17 |
pshedimb | Intially I had placement config error | 21:17 |
pshedimb | I copied placement config from controller to compute node | 21:18 |
pshedimb | And now I'm having this error | 21:18 |
*** catintheroof has quit IRC | 21:19 | |
*** Apoorva has joined #openstack-nova | 21:19 | |
diana_clarke | pshedimb: is placement-api in your ENABLED_SERVICES? | 21:22 |
*** raunak has quit IRC | 21:23 | |
dansmith | that error isn't related to placement | 21:23 |
*** _ducttape_ has quit IRC | 21:23 | |
dansmith | and it's not an error if it's followed by "created a service record..." | 21:23 |
*** ducttape_ has joined #openstack-nova | 21:24 | |
pshedimb | dansmith, On controller placement-api is enabled | 21:25 |
pshedimb | On compute node, I copied placement-api related conf to nova.conf manually | 21:25 |
*** andrewbogott has quit IRC | 21:26 | |
*** andrewbogott has joined #openstack-nova | 21:26 | |
*** raunak has joined #openstack-nova | 21:26 | |
mriedem | bauzas: thanks for https://review.openstack.org/#/c/428352/ | 21:26 |
diana_clarke | dansmith: oh, I just recall that being one of the answers to the same question here: http://eavesdrop.openstack.org/irclogs/%23openstack-nova/%23openstack-nova.2017-02-01.log.html#t2017-02-01T13:08:33 | 21:27 |
pshedimb | dansmith, It is not followed by " "created a service record..." | 21:27 |
mriedem | bauzas: if you still have some stomach left, https://review.openstack.org/#/c/427499/ | 21:27 |
*** srwilkers has joined #openstack-nova | 21:27 | |
diana_clarke | nevermind | 21:27 |
dansmith | diana_clarke: the compute node record is service and old-school RT related | 21:27 |
bauzas | mriedem: I had an open question on https://review.openstack.org/#/c/428352/ about older allocation records | 21:28 |
bauzas | mriedem: I'm not sure we auto-heal previous records | 21:28 |
mriedem | bauzas: probably a question for jaypipes | 21:28 |
mriedem | or cdent | 21:28 |
bauzas | yup | 21:28 |
bauzas | folks ? | 21:28 |
* cdent looks | 21:28 | |
bauzas | mriedem: dansmith: oh man, totally forgot to amend the nova-status check for filters | 21:29 |
mriedem | i've got to run for awhile | 21:29 |
mriedem | bauzas: that's not stopship imo | 21:29 |
mriedem | and we could backport if needed | 21:29 |
* mriedem is out | 21:29 | |
*** mriedem has quit IRC | 21:29 | |
bauzas | mriedem: okay, go to your meeting | 21:29 |
cdent | bauzas: we check allocations every periodic job in the resource tracker. if what's on the server is different from what we have now, we rewrite it | 21:29 |
bauzas | cdent: that's what I thought but I wanted to be sure | 21:30 |
cdent | bauzas: you want me to leave that response on the review too, or okay? | 21:30 |
bauzas | cdent: so the patch above will autofix allocations after the 60-sec default period, cool | 21:30 |
cdent | yes | 21:30 |
bauzas | cdent: nope, not needed, I +Wd it | 21:31 |
cdent | great | 21:31 |
*** sapcc-bot has quit IRC | 21:33 | |
*** carthaca_ has quit IRC | 21:33 | |
*** Swami has quit IRC | 21:34 | |
*** cleong has quit IRC | 21:34 | |
*** Jeffrey4l has quit IRC | 21:35 | |
*** sapcc-bot has joined #openstack-nova | 21:35 | |
*** tpatzig_ has joined #openstack-nova | 21:35 | |
*** Swami has joined #openstack-nova | 21:36 | |
*** tpatzig_ has quit IRC | 21:37 | |
*** ijw has joined #openstack-nova | 21:38 | |
*** Jeffrey4l has joined #openstack-nova | 21:38 | |
*** cdent has quit IRC | 21:38 | |
*** dimtruck is now known as zz_dimtruck | 21:46 | |
*** karimb has quit IRC | 21:46 | |
*** dave-mccowan has joined #openstack-nova | 21:47 | |
*** sambetts is now known as sambetts|afk | 21:49 | |
*** avolkov has quit IRC | 21:53 | |
*** srwilkers has left #openstack-nova | 21:53 | |
*** salv-orlando has joined #openstack-nova | 21:57 | |
*** karimb has joined #openstack-nova | 22:03 | |
*** flwang has left #openstack-nova | 22:03 | |
openstackgerrit | Jiri Suchomel proposed openstack/nova master: Retry few times when creating the initial connection to libvirtd. https://review.openstack.org/416928 | 22:04 |
*** vsaienko has quit IRC | 22:04 | |
*** vsaienko has joined #openstack-nova | 22:05 | |
*** satyar_ has quit IRC | 22:06 | |
*** thorst_ has quit IRC | 22:06 | |
*** breitz has quit IRC | 22:14 | |
*** salv-orl_ has joined #openstack-nova | 22:15 | |
*** slaweq has joined #openstack-nova | 22:16 | |
*** salv-orlando has quit IRC | 22:18 | |
*** crushil has quit IRC | 22:19 | |
*** esberglu has quit IRC | 22:24 | |
*** esberglu has joined #openstack-nova | 22:24 | |
*** mriedem has joined #openstack-nova | 22:24 | |
mriedem | i'm back | 22:24 |
*** jheroux has quit IRC | 22:25 | |
*** xinli has quit IRC | 22:25 | |
*** harlowja has joined #openstack-nova | 22:25 | |
*** breitz has joined #openstack-nova | 22:26 | |
*** ducttape_ has quit IRC | 22:26 | |
EmilienM | mriedem: https://review.openstack.org/#/c/428424/ is available for review when you have time (as PTL you need to sign-off) | 22:28 |
*** esberglu has quit IRC | 22:28 | |
*** AAR has joined #openstack-nova | 22:29 | |
*** AAR has left #openstack-nova | 22:29 | |
*** tlian has joined #openstack-nova | 22:30 | |
*** ducttape_ has joined #openstack-nova | 22:31 | |
*** zz_dimtruck is now known as dimtruck | 22:31 | |
*** thorst_ has joined #openstack-nova | 22:31 | |
*** harlowja has quit IRC | 22:32 | |
*** harlowja has joined #openstack-nova | 22:33 | |
*** takashin has left #openstack-nova | 22:33 | |
*** takashin has joined #openstack-nova | 22:34 | |
*** Sukhdev has joined #openstack-nova | 22:34 | |
mriedem | EmilienM: i'm not going to dig into that one today probably | 22:35 |
*** thorst_ has quit IRC | 22:35 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova master: Add functional tests for cold migration to same host https://review.openstack.org/414926 | 22:36 |
melwitt | dansmith: do you know which part of the scheduler/placement code treats RequestSpec root_gb as 0 if BFV during scheduling? I haven't found it yet | 22:38 |
EmilienM | mriedem: ok, just don't forget it please | 22:38 |
*** slaweq has quit IRC | 22:38 | |
*** annegentle has quit IRC | 22:38 | |
*** marst has quit IRC | 22:39 | |
*** erlon has quit IRC | 22:40 | |
dansmith | melwitt: here's where we request it: https://review.openstack.org/#/c/417961/41/nova/scheduler/filter_scheduler.py | 22:42 |
mriedem | edleafe: are you working on the unit tests in https://review.openstack.org/#/c/428375/ ? | 22:42 |
dansmith | melwitt: but looking at that, we probably have to adjust that logic? | 22:42 |
*** ducttape_ has quit IRC | 22:43 | |
dansmith | melwitt: like if is_bfv(), don't include root_gb in there | 22:43 |
melwitt | dansmith: yeah, I think so. right | 22:43 |
dansmith | for some reason I had in my head that the math there equaling zero means we don't ask for DISK_GB, | 22:44 |
dansmith | but if we want root_gb!=0 then we need one more if in there | 22:44 |
dansmith | melwitt: you want me to work that up? | 22:44 |
mriedem | i swear i mentioned something about that to bauzas during that review | 22:45 |
mriedem | because the report client was checking is_bfv | 22:45 |
dansmith | I did too | 22:45 |
dansmith | but in irc | 22:45 |
melwitt | yeah, the reporting is working I think. that means compute hosts won't advertise less disk available if they have bfv instances on them | 22:45 |
*** esberglu has joined #openstack-nova | 22:45 | |
openstackgerrit | Ed Leafe proposed openstack/nova master: Add check for invalid allocation amounts https://review.openstack.org/416752 | 22:45 |
dansmith | yeah | 22:45 |
dansmith | that's been there for a while | 22:45 |
melwitt | but that doesn't address the issue of asking for disk when during a request if bfv | 22:46 |
dansmith | yep, it's a two-liner I think | 22:46 |
melwitt | -when | 22:46 |
edleafe | mriedem: yeah - just updated https://review.openstack.org/416752first | 22:46 |
melwitt | dansmith: yeah, if you want to add that, sure. what's there currently is just making sure disk isn't requested if the flavor has 0 disk. but the bfv thing is flavor can have disk but it ends up on a volume and not on the compute host | 22:47 |
dansmith | yep yep | 22:47 |
*** kaisers_ has quit IRC | 22:47 | |
mriedem | https://review.openstack.org/#/c/417961/22/nova/scheduler/filter_scheduler.py@151 | 22:48 |
mriedem | was my comment / question i think | 22:48 |
dansmith | gdi gerrit | 22:49 |
mriedem | it seems like the ultimate agreement was we needed to check for bfv in there, and then didn't | 22:50 |
melwitt | okay, so bauzas was saying wait until cinder is telling the capacity and then check against that. in the future | 22:50 |
mriedem | which would still be a different resource provider | 22:50 |
mriedem | is what dan clarified | 22:51 |
*** mtanino has joined #openstack-nova | 22:51 | |
melwitt | okay, so that means we have to pass by this RP then? sorry I don't know how this works | 22:51 |
mriedem | you'd have an aggregate for the shared storage pool | 22:51 |
mriedem | with compute RPs in that aggregate | 22:51 |
dansmith | that's just to get the math right, | 22:51 |
dansmith | we still don't want to charge it against the compute node if it's a volume | 22:51 |
-openstackstatus- NOTICE: Restarting gerrit due to performance problems | 22:51 | |
melwitt | yeah | 22:51 |
dansmith | we have BDMs by this point, but I'm not sure they make it to the scheduler to make this easy ... | 22:52 |
*** salv-orl_ has quit IRC | 22:53 | |
*** edmondsw_ has quit IRC | 22:53 | |
mriedem | edleafe: so https://review.openstack.org/#/c/428375/ depends on https://review.openstack.org/416752 ? | 22:54 |
mriedem | edleafe: if so, can you rebase them to stack up that way? | 22:54 |
*** edmondsw has joined #openstack-nova | 22:54 | |
edleafe | mriedem: I separated them at cdent's request | 22:54 |
*** marst has joined #openstack-nova | 22:56 | |
mriedem | sdague: jaypipes: if you're around, would be good to get this in https://review.openstack.org/#/c/427499/ | 22:57 |
mriedem | b/c anyone using nova-status when upgrading to ocata will be possibly busted w/o it | 22:57 |
mriedem | well, the status check will fail | 22:57 |
mriedem | which we probably don't want to happen on it's maiden voyage | 22:57 |
*** burt has quit IRC | 22:58 | |
*** edmondsw has quit IRC | 22:58 | |
*** burgerk_ has joined #openstack-nova | 23:00 | |
*** panda is now known as panda|zZ | 23:00 | |
edleafe | So what's the preferred way of dealing with keystone auth exceptions when placement isn't configured? Seems that 726 tests hit the compute node destroy path, which will call placement in https://review.openstack.org/#/c/428375/ | 23:02 |
edleafe | I would prefer to configure it, if possible :) | 23:02 |
*** tbachman has quit IRC | 23:02 | |
*** burgerk has quit IRC | 23:03 | |
*** burgerk_ has quit IRC | 23:04 | |
*** mdrabe has quit IRC | 23:06 | |
*** jamesden_ has quit IRC | 23:06 | |
openstackgerrit | Dan Smith proposed openstack/nova master: Request zero root disk for boot-from-volume instances https://review.openstack.org/428481 | 23:08 |
dansmith | melwitt: working on tests now, but want to see a run of that ^ to make sure I've correctly guessed which of the ten BDM formats we have right there | 23:08 |
mriedem | edleafe: the @safe_connect decorator | 23:10 |
dansmith | if we had more than a couple hours I'd just rev the rpc api, but.. | 23:10 |
edleafe | mriedem: heh, I was just trying that out... | 23:10 |
*** dimtruck is now known as zz_dimtruck | 23:11 | |
mriedem | edleafe: although you might need to use the scheduler report client / placement fixture that bauzas used in his scheduler tests | 23:12 |
edleafe | mriedem: we'll see | 23:12 |
mriedem | dansmith: root_bdm() can return None | 23:15 |
mriedem | so that will likely explode for a bunch of tests | 23:15 |
dansmith | mriedem: I know | 23:15 |
dansmith | duh | 23:15 |
dansmith | sheesh | 23:15 |
dansmith | just wanted to make sure YOU knew | 23:15 |
dansmith | and 6 is the number of tests | 23:15 |
*** zz_dimtruck is now known as dimtruck | 23:15 | |
mriedem | plus | 23:17 |
mriedem | PLUS | 23:17 |
mriedem | according to https://github.com/openstack/nova/blob/master/nova/objects/block_device.py#L364 it has to be per-instance | 23:17 |
dansmith | mriedem: I don't think that applies here, | 23:18 |
mriedem | i meant tempest tests btw | 23:18 |
dansmith | because this is a BDM that we replicate for each, | 23:18 |
dansmith | not a set of bdms from multuple instances | 23:18 |
dansmith | multiple | 23:18 |
mriedem | oh right | 23:18 |
mriedem | copies the list, and then sets the instance uuid on each bdm in thelist | 23:19 |
dansmith | yar | 23:19 |
mriedem | multuple is py4 btw | 23:19 |
mriedem | which is an openstack goal for T | 23:19 |
dansmith | badumpum | 23:19 |
*** thorst_ has joined #openstack-nova | 23:20 | |
dansmith | ugh, requestspec is the new bdm | 23:20 |
*** trevormc has quit IRC | 23:21 | |
*** owalsh-afk has joined #openstack-nova | 23:24 | |
dansmith | also I'm pretty sure we're serializing to legacy, then passing it to a method that unserializes it | 23:24 |
dansmith | but I don't even want to think about it right now | 23:24 |
melwitt | yeah, there's a small piece in between that can't yet handle objects, so it does that | 23:24 |
dansmith | no, | 23:24 |
dansmith | I think it's direct | 23:24 |
dansmith | lemme get links | 23:24 |
melwitt | oh. I know there's a thing where it serializes to pass to a specific method, then after it comes back it turns it back into an object. or something | 23:25 |
dansmith | https://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L865 | 23:25 |
dansmith | https://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L595 | 23:25 |
dansmith | yeah | 23:25 |
*** takashin has left #openstack-nova | 23:25 | |
openstackgerrit | Dan Smith proposed openstack/nova master: Request zero root disk for boot-from-volume instances https://review.openstack.org/428481 | 23:25 |
dansmith | mriedem: tests never failed if you push before jenkins finished | 23:26 |
melwitt | I wonder if I could bolt my sweet functional test onto that patch to further test it. or I could stack it over yours https://review.openstack.org/#/c/355091/8/nova/tests/functional/test_boot_from_volume.py | 23:27 |
dansmith | that is the worst test I've ever seen | 23:27 |
* dansmith just wants to balance jaypipes' gushing | 23:28 | |
*** dave-mccowan has quit IRC | 23:28 | |
melwitt | I was about to be like *gasp* how could you say that! | 23:28 |
dansmith | heh | 23:28 |
*** baoli has quit IRC | 23:28 | |
dansmith | yeah, I say stack it on for now at least | 23:29 |
melwitt | cool. this is exciting | 23:29 |
* dansmith thinks melwitt needs her meter recalibrated | 23:29 | |
melwitt | I feel like I'm the champion of boot from volume and root_gb = 0 at this point | 23:30 |
gmann | mriedem: alex_xu this one from kevin on release note (we should have this on etherpad) - https://review.openstack.org/#/c/427978/ | 23:31 |
mriedem | gmann: alex_xu split the release note out https://review.openstack.org/#/c/428285/ | 23:32 |
mriedem | i'm reviewing and cleaning it up now | 23:32 |
gmann | mriedem: cool, that also fine | 23:32 |
gmann | mriedem: i did not see doc patch update as per comments - https://review.openstack.org/#/c/421760/ | 23:32 |
mriedem | dansmith: is that like a 5 second rule? | 23:33 |
*** gouthamr has quit IRC | 23:33 | |
dansmith | mriedem: similar, but with less salmonella | 23:33 |
gmann | mriedem: if we are taking that too, i can work on that | 23:33 |
*** esberglu has quit IRC | 23:33 | |
*** Sukhdev has quit IRC | 23:33 | |
mriedem | gmann: api-ref can be in pike | 23:33 |
mriedem | imo | 23:33 |
*** tblakes has quit IRC | 23:33 | |
mriedem | api-ref is built from master anyway | 23:33 |
gmann | yea, that makes no hurry | 23:34 |
melwitt | mriedem: that reminds me, the cells v2 step-by-step stuff is also not related to deadline right? since it's devref | 23:34 |
*** owalsh-afk is now known as owalsh | 23:34 | |
*** tlian has quit IRC | 23:35 | |
mriedem | melwitt: let's say yes | 23:37 |
melwitt | cools | 23:38 |
*** owalsh is now known as owalsh-afk | 23:38 | |
melwitt | my test didn't work -- NoValidHost. I wonder if I need to do something to enable placement in functional test | 23:39 |
openstackgerrit | Matt Riedemann proposed openstack/nova master: Add release note for filter/sort whitelist https://review.openstack.org/428285 | 23:39 |
mriedem | gmann: see how this looks ^ i touched up the wording and fancied it up a bit | 23:39 |
mriedem | melwitt: placement fixture? | 23:39 |
* melwitt looks | 23:39 | |
mriedem | https://github.com/openstack/nova/blob/master/nova/tests/fixtures.py#L1175 | 23:39 |
melwitt | thanks | 23:40 |
melwitt | I <3 fixtures | 23:40 |
jaypipes | mriedem: done. | 23:40 |
mriedem | thanks | 23:41 |
mriedem | oh also, i might just +W this myself if no one else does https://review.openstack.org/#/c/428344/ | 23:41 |
dansmith | mriedem: I'll get it | 23:42 |
mriedem | thanks | 23:42 |
dansmith | mriedem: we should also put the wsgi clarification in, I need to de-parent it | 23:42 |
mriedem | ok | 23:42 |
jaypipes | mriedem: lemme review real quick. | 23:42 |
mriedem | this is going to be the most gd docstastic release ever | 23:42 |
melwitt | mriedem: I looked at that earlier and wondered if you should note that [cells] options aren't required for cells v2 in the first blurb too | 23:42 |
mriedem | shrug | 23:43 |
mriedem | it's approved now so i have no incentive | 23:43 |
mriedem | :) | 23:43 |
openstackgerrit | Dan Smith proposed openstack/nova master: Clarify the language in the apache wsgi sample https://review.openstack.org/428417 | 23:43 |
dansmith | mriedem: ^ | 23:43 |
mriedem | wipped as bauzas would say | 23:43 |
dansmith | heh | 23:44 |
openstackgerrit | Ed Leafe proposed openstack/nova master: Delete a compute node's resource provider when node is deleted https://review.openstack.org/428375 | 23:44 |
edleafe | mriedem: jaypipes: ^^ Had to rebase this on https://review.openstack.org/#/c/404472 to get safe_connect | 23:47 |
*** salv-orlando has joined #openstack-nova | 23:48 | |
*** amotoki has quit IRC | 23:48 | |
melwitt | looks like it's still trying to claim the flavor root_gb | 23:48 |
*** vsaienko has quit IRC | 23:49 | |
dansmith | melwitt: in your test or somewhere else? | 23:51 |
dansmith | melwitt: or do you mean on the compute node? | 23:51 |
melwitt | dansmith: running my test I see it making the claim with the instance flavor root_gb on the compute host | 23:52 |
dansmith | melwitt: that's expected right? | 23:52 |
dansmith | the claim stuff knows nothing about this, | 23:52 |
melwitt | dansmith: yeah but it will fail because the compute host free disk figure will kick it out | 23:52 |
dansmith | but we should be reporting the right thing to placement, and not considering it on scheduling | 23:52 |
*** salv-orlando has quit IRC | 23:52 | |
dansmith | until we remove that stuff, right? | 23:53 |
melwitt | so it can't schedule to a compute host that isn't reporting enough local disk to fit the flavor | 23:53 |
dansmith | so maybe we need the claims bit of your patch? | 23:53 |
*** claudiub has quit IRC | 23:53 | |
*** karimb has quit IRC | 23:53 | |
dansmith | I wish we didn't though, since that's a bunch of extra pieces :/ | 23:54 |
melwitt | it appears so. it's pretty ugly though https://review.openstack.org/#/c/355091/8/nova/compute/resource_tracker.py@155 | 23:54 |
dansmith | yeah | 23:54 |
melwitt | I dunno what resource providers plan is for fixing this | 23:55 |
dansmith | I don't know that we keep the claims stuff past ocata | 23:55 |
dansmith | I'm not sure though | 23:55 |
dansmith | melwitt: your stuff in claims all says "remove this after resource providers" | 23:55 |
melwitt | yeah, that wasn't based on knowing how it would be solved. my understanding was that RP fixes all the things | 23:56 |
mriedem | i'm going to go on a limb and say, | 23:56 |
mriedem | i really like joe walsh | 23:56 |
dansmith | well, when we get allocations in the scheduler it definitely goes away, AFAIK | 23:56 |
melwitt | okay. then I should change my TODOs to say that. I didn't know what specifically would make those changes obsolete in the future | 23:57 |
dansmith | man, that's a lot of copying the instance object every time we report resources | 23:58 |
mriedem | gmann: found a problem in the reno | 23:58 |
mriedem | fixing quick | 23:58 |
*** dtp has quit IRC | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!