Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-05-21
11:39:53 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.binary_type/integer_types/string_types https://review.opendev.org/728094
11:47:48 openstackgerrit Lee Yarwood proposed openstack/nova master: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/578846
11:50:41 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.text_type (1/2) https://review.opendev.org/728109
11:53:17 openstackgerrit Takashi Natsume proposed openstack/nova master: Remove six.text_type (2/2) https://review.opendev.org/728117
12:00:47 openstackgerrit Balazs Gibizer proposed openstack/nova master: Remove inject_file from compute manager and virt driver https://review.opendev.org/729934
12:36:09 openstackgerrit Brin Zhang proposed openstack/nova master: WIP: Cyborg suspend/resume support https://review.opendev.org/729945
12:41:57 openstackgerrit Balazs Gibizer proposed openstack/nova master: Remove compatibility check from migrate_data https://review.opendev.org/729949
13:06:02 openstackgerrit Wenping Song proposed openstack/nova master: cyborg evacuate support https://review.opendev.org/715326
13:09:24 lvdombrkr Hello guys. using Kolla-ansible Train + Centos8, trying to enable hugepages, HOSt info : http://paste.openstack.org/raw/793838/
13:09:57 lvdombrkr creating flavor openstack flavor create hugepage --ram 1024 --disk 10 --vcpus 1 openstack flavor set hugepage --property hw:mem_page_size=1GB
13:10:10 lvdombrkr but when booting VM get this error in logs
13:10:17 lvdombrkr Requested instance NUMA topology cannot fit the given host NUMA topology.\n']
13:10:24 lvdombrkr any clues?
13:11:05 stephenfin lvdombrkr: compare these
13:11:06 stephenfin openstack flavor set hugepage --property hw:mem_page_size=1GB
13:11:15 stephenfin Hugepagesize: 2048 kB
13:11:25 stephenfin (the latter is from http://paste.openstack.org/raw/793838/)
13:11:57 stephenfin do this instead
13:12:01 stephenfin openstack flavor set hugepage --property hw:mem_page_size=2M
13:16:19 lvdombrkr stephenfin: now i had another error : Invalid memory page size '0' (HTTP 400) (Request-ID: req-338bf619-3a54-45c5-9c59-ad8c1d425e91)
13:16:20 lvdombrkr http://paste.openstack.org/show/793839/
13:18:49 stephenfin Ah, apologies. That should have read 'hw:mem_page_size=2MB'
13:18:54 stephenfin That's a rather confusing error message, btw. Might be worth filing a bug for it.
13:22:39 lvdombrkr stephenfin: its works with hw:mem_page_size=2MB
13:22:57 lvdombrkr stephenfin: Thanks in assist!
13:23:18 stephenfin lvdombrkr: Good to hear. Could you open a bug about that confusing error message? (https://bugs.launchpad.net/nova/) Should be easy to fix
13:28:26 lvdombrkr stephenfin: sure, will do now
13:28:31 stephenfin thanks
14:03:54 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Raise useful error for invalid mempage size https://review.opendev.org/729967
14:15:18 openstackgerrit Lee Yarwood proposed openstack/nova master: libvirt: Don't delete disks on shared storage during evacuate https://review.opendev.org/578846
14:50:12 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Raise useful error for invalid mempage size https://review.opendev.org/729967
14:50:26 stephenfin real easy bugfix there if anyone fancies a gawk ^
14:59:13 lyarwood stephenfin: I'll take a look now, would you mind looking at https://review.opendev.org/#/c/729261/ in return?
14:59:35 stephenfin fair
15:00:58 artom Hehe, raise from?
15:01:27 stephenfin artom: lyarwood: http://paste.openstack.org/show/793850/
15:01:38 stephenfin try that with and without the from and it'll be obvious
15:01:53 stephenfin tl;dr: we drop the context from the captured exception
15:01:58 lyarwood TIL
15:02:09 stephenfin python3++
15:13:59 stephenfin lyarwood: awkward question left on https://review.opendev.org/#/c/729261/
15:16:43 lyarwood stephenfin: ack, quick nit on yours otherwise it LGTM
15:23:13 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Raise useful error for invalid mempage size https://review.opendev.org/729967
15:43:21 lyarwood stephenfin: are we good to use f-string in nova now?
15:43:27 stephenfin yup
15:43:54 lyarwood oh wait backports
15:43:57 lyarwood gah
15:47:29 sean-k-mooney lyarwood: stephen already has used them i think
15:47:46 sean-k-mooney lyarwood: im pretty sure we have merged a couple of patches with fstrings
15:47:55 stephenfin yeah, they're used in the extra spec validator code anyway, though that's not being backported
15:48:11 sean-k-mooney ya i think that is what i was thinking of
15:48:20 sean-k-mooney i think i used them for something too
15:48:36 sean-k-mooney but i think that someting that can be fixed in the backport
15:50:18 gibi nova meeting will start in 10 minutes on #openstack-meeting-3
15:51:02 lyarwood yeah my git grep failed to find them but I was sure we had some in the codebase already
15:51:10 lyarwood for anything that I'm not going to backport back to Queens
15:51:14 lyarwood I'll use them
15:51:15 sean-k-mooney gibi: im proably not going to attend as i have a rather annoying headache right now so going to lie down for a bit
15:51:33 gibi sean-k-mooney: ack, I hope it will be gone soon
15:53:57 sean-k-mooney i think its partly dehydration and partly the wether so ya it should go away on its own
16:09:46 openstackgerrit Dan Smith proposed openstack/nova stable/stein: Make quotas respect instance_list_per_project_cells https://review.opendev.org/729088
16:10:20 dansmith mnaser: lyarwood ^ added missing mock to the test file, which will need to go back to the stein one if that passes tests
16:13:06 lyarwood dansmith: that is the stein one, looks like it was already there in train
16:13:40 dansmith ah dang, I guess I got confused
16:13:44 dansmith which makes the commit message wrong
16:14:28 dansmith I need to coin some sort of term that conveys the common practice of getting lost in following gerrit topic links between branches
16:14:54 dansmith maybe I'm the only one, but I forget which alternate 1985 I'm in pretty easily when following that circular loop
16:15:43 openstackgerrit Dan Smith proposed openstack/nova stable/stein: Make quotas respect instance_list_per_project_cells https://review.opendev.org/729088
16:19:00 artom minmax(manhours, beerhours) ftw
16:37:11 openstackgerrit Lee Yarwood proposed openstack/nova master: compute: Do not allow rescue attempts using volume snapshot images https://review.opendev.org/729261
17:05:38 openstackgerrit Merged openstack/nova master: [Trivial]Add missing white spaces between words in log messages https://review.opendev.org/729880
18:41:41 dansmith can someone help sync me up with where we are on like servicey tokeny things?
18:42:04 dansmith when we're talking to glance, are we using the user's token exclusively still?
18:42:27 dansmith and if we have the servicey tokeny thing, I assume that just helps us for the case where the user
18:42:56 dansmith token has expired, but doesn't give us any other permissions?
18:43:03 dansmith do/can we talk to glance with a more authoritative token specifically, like to look at tasks, which are admin-only?
20:19:51 melwitt dansmith: I don't know the answer to your question but here's a relatively recent change we made to our neutron interaction in a case where we changed to use admin to get specific info that wasn't available to non-admin, in case that is helpful https://github.com/openstack/nova/commit/aab4b7a0e2504c04e08389145bcb1414dea63631
20:22:05 dansmith melwitt: if you had used fewer words, you'd have sounded like an irc bot :)
20:22:25 dansmith melwitt: that's relevant-adjacent I think.. I need glance specifically and I'm sure we could be doing each service N+1 different ways
20:23:03 dansmith BUT, it looks like glance exposes a minor amount of information about the task I'm concerned with to the regular users, so at least at the moment I think I can get away without hitting the task api and thus I can put off having to learn the answer to my question for a while longer :D
20:24:01 melwitt I do sound like a bot. words are hard
20:24:26 melwitt ah, ok. yeah I was thinking if you're looking to elevate the token for certain APIs you need, that's an example of where we've done that before
20:24:55 dansmith okay, yeah, then probably relevant
20:25:13 dansmith but if I can avoid elevation, then that's probably better and it looks like I can, at least at the moment
20:25:21 melwitt yeah, for sure
20:32:40 openstackgerrit Dan Smith proposed openstack/nova master: WIP: Make libvirt able to trigger a backend image copy when needed https://review.opendev.org/656998
21:11:29 openstackgerrit Dan Smith proposed openstack/nova-specs master: Add rbd-glance-multistore spec https://review.opendev.org/730126
21:57:02 openstackgerrit melanie witt proposed openstack/nova stable/stein: Make quotas respect instance_list_per_project_cells https://review.opendev.org/729088
#openstack-nova - 2020-05-22
02:40:00 openstackgerrit zhangboye proposed openstack/nova-specs master: Fix hacking min version to 3.0.1 https://review.opendev.org/730162
03:12:50 openstackgerrit zhaoleilc proposed openstack/nova master: Fix an erroneous grammar in explanatory notes https://review.opendev.org/730176
05:43:39 openstackgerrit Wenping Song proposed openstack/nova master: cyborg evacuate support https://review.opendev.org/715326
06:07:49 openstackgerrit junbo proposed openstack/nova master: use dpdk interface type for ovs dpdk representor port https://review.opendev.org/729874
08:48:03 openstackgerrit Balazs Gibizer proposed openstack/nova stable/queens: Clean up allocation if unshelve fails due to neutron https://review.opendev.org/729540
08:56:29 openstackgerrit Balazs Gibizer proposed openstack/nova stable/queens: Reproduce bug 1862633 https://review.opendev.org/729539
08:56:29 openstack bug 1862633 in OpenStack Compute (nova) "unshelve leak allocation if update port fails" [Medium,Fix released] https://launchpad.net/bugs/1862633 - Assigned to Balazs Gibizer (balazs-gibizer)
08:56:29 openstackgerrit Balazs Gibizer proposed openstack/nova stable/queens: Clean up allocation if unshelve fails due to neutron https://review.opendev.org/729540
09:00:08 openstackgerrit Wenping Song proposed openstack/nova master: WIP: Cyborg shelve/unshelve support https://review.opendev.org/729563
09:24:45 openstackgerrit Merged openstack/nova-specs master: re-propose numa with placement https://review.opendev.org/728009
09:32:10 gibi stephenfin: hi! I'm not sure about your suggestion in https://review.opendev.org/#/c/729924/1/nova/compute/manager.py@a280
09:34:15 stephenfin gibi: Ah, I'm wrong then. I was searching for Migration._from_db_object since it was a static method

Earlier   Later