Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-02-06
16:09:09 mriedem what am i being passive aggressive about?
16:09:30 mriedem i hit this same thing from time to time and have to remember to delete pycs manually which is annoying
16:10:20 sean-k-mooney ill update the docs tox env but i think we need to add graphviz as a docs dep too
16:10:29 mriedem graphviz is a binary
16:10:32 mriedem it's in bindep
16:10:48 sean-k-mooney you are right but why do we now need it for docs
16:10:53 sean-k-mooney we didnt in the past
16:10:59 mriedem we've needed it for a long time
16:11:02 mriedem are you on a new system?
16:11:18 sean-k-mooney im on a system that has never had devstack run
16:11:20 sean-k-mooney so yes
16:11:28 sean-k-mooney ok ill just install it
16:12:09 jackding mriedem: runway for https://review.openstack.org/#/c/620706/ ends tomorrow. I believe I have addressed your comments. Do you have sometime to revisit?
16:12:12 mriedem stephenfin: this is what i was referring to http://lists.openstack.org/pipermail/openstack-dev/2014-September/045873.html
16:12:38 stephenfin mriedem: "we used to have something in tox.ini to delete those on each run to avoid that but someone had an opinion and removed it"
16:12:44 stephenfin mriedem: I figured that was aimed at me, in which case it seemed like I was being singled out for what was almost certainly a change made for good reasons
16:12:50 stephenfin as opposed to "opinions"
16:13:01 mriedem i was referring to that thread
16:13:13 mriedem which was related to https://review.openstack.org/#/c/338231/
16:13:25 mriedem which was yours, but that was just "don't create pycs rather than delete them" which as you can see i was +2 on
16:13:50 stephenfin Yeah, we reverted that though 233160644f4f0a17177fffe8a1e6db1341e61ad4
16:13:56 mriedem i know
16:14:38 sean-k-mooney anyway ill push a patch to add the delete to the docs job if ye feel like reviewing
16:14:40 stephenfin Cool. So this has just been missed is all. Surprised it didn't come up before now, tbh
16:14:56 mriedem sean-k-mooney: just do what stephenfin asked for
16:15:05 mriedem '{[testenv]commands}'
16:15:17 sean-k-mooney ya that was what i was going to do
16:15:42 sean-k-mooney i was looking up the syntax :)
16:16:37 stephenfin mriedem: But, crucially, I didn't remove it (or anything else, I'd like to think) the first day because of "opinions". I gave sound reasons in that commit message. I just missed the tox issue, which was the reason for the later revert
16:16:39 stephenfin That is all :)
16:16:59 stephenfin Speaking of, as soon as we support a new enough version of tox, we can revert that revert again
16:17:27 stephenfin https://github.com/tox-dev/tox/commit/336f4f6bd8b53223f940fc5cfc43b1bbd78d4699
16:20:05 melwitt gibi_off: thank you for the note! I hope you are having a nice recovery
16:27:07 mriedem stephenfin: question inline https://review.openstack.org/#/c/635170/
16:27:37 stephenfin mriedem: Yeah, I am
16:28:10 stephenfin sean-k-mooney, jangutter: You should probably look at the above too
16:28:44 sean-k-mooney stephenfin: for vhostuser i dont think they work
16:28:52 sean-k-mooney stephenfin: that is why we dont set them today
16:29:14 sean-k-mooney stephenfin: libvirt sets tehm on the tap device but vhost user does not have a tap
16:29:53 jangutter sean-k-mooney: yah vhost-user has a socket.
16:30:56 sean-k-mooney stephenfin: jangutter well the main point is that only way to do ratelimiting with vhost-user that im aware of is via dpdk/ovs
16:31:24 jangutter sean-k-mooney: ya, there's no netdev to set nifty options like rxqueuelen, etc.
16:31:29 sean-k-mooney tc cant do ratelimiting on a unix socket escpially since that socket is part of the contol plane not the data plane
16:31:55 sean-k-mooney jangutter: well you can set the queue lenght in the qemu commandline
16:32:02 sean-k-mooney and we do i think
16:32:26 sean-k-mooney but qemu does not do the rate limiting tc did i think
16:33:51 openstackgerrit sean mooney proposed openstack/nova master: cleanup *.pyc files in docs tox envs https://review.openstack.org/635210
16:33:59 jangutter sean-k-mooney: drat, already running out of caffeine, I'm thinking of txquelen for tc qdiscs (like ip link set txqueuelen 500 dev eth0)
16:34:47 jangutter sean-k-mooney: yah, setting things like number of descriptors in vhost-user is something entirely different and doesn't have to do with rate limiting.
16:34:51 mriedem sean-k-mooney: weren't you +1 on https://review.openstack.org/#/c/565471/ previously? this change for bw was split out from that
16:37:09 sean-k-mooney mriedem: no dont think i had ever left a +1 on that one specifrically but i did remove my -1
16:37:46 sean-k-mooney i have not reviewed it recently
16:38:31 sean-k-mooney mriedem: stephenfin anyway i dont think it would break vhost-user but i think it would have no effect
16:38:58 sean-k-mooney in anycase we should not set it as it cannot work with vhost-user so its misleading
16:40:55 mriedem sean-k-mooney: ok i remember you asking me to look at it :)
16:40:59 mriedem so maybe i assumed you had
16:42:23 sean-k-mooney mriedem: yes i did sorry i had planned too re review but i didnt get to it yet sorry
16:44:11 mriedem you're forgiven sean
16:44:23 melwitt bauzas: you're not in #openstack-placement, but there's a extraction update meeting at the top of the hour, hangout url to be linked in #openstack-placement
16:45:45 bauzas melwitt: I'll be there
16:45:51 melwitt cool
16:46:01 bauzas melwitt: for some reason, ZNC dropped me from -placement, but I'll rejoin
16:46:40 bauzas melwitt: thanks for the notice :)
16:47:47 jangutter sean-k-mooney, stephenfin: https://github.com/libvirt/libvirt/blob/568a417224e7b4656b9959b03154d45566480d65/src/conf/netdev_bandwidth_conf.h#L38
16:49:00 sean-k-mooney jangutter right so based on ^ libvirt will ignore it for vhos-user if it is set
16:49:45 jangutter sean-k-mooney, stephenfin: yah, in fact, if that's canonical, it's a nice checklist of "supported" interfaces.
16:49:51 openstackgerrit Michal Arbet proposed openstack/nova master: Fix python3 compatibilit https://review.openstack.org/635220
16:50:25 openstackgerrit Michal Arbet proposed openstack/nova master: Fix python3 compatibility https://review.openstack.org/635220
16:57:10 jackding mriedem: just let you know that https://review.openstack.org/#/c/620706/ and https://review.openstack.org/#/c/603844/ is ready to review, thanks
17:01:23 cdent for those not in #openstack-placement placement checkin hangout at https://hangouts.google.com/call/C5h9TyVk5BMfSO8vGxZ_AEEE
17:10:51 stephenfin jangutter, sean-k-mooney: Right, so we were setting the option for OVS vHost User interfaces but it wasn't doing anything
17:11:24 stephenfin jangutter, sean-k-mooney: In that case, is it fair to fold it back into the following patch and just ignore the fact that it's being set, given that it does nothing?
17:11:56 stephenfin Far as I can see, the only side effect will be this https://review.openstack.org/#/c/565471/7/nova/tests/unit/virt/libvirt/test_vif.py
17:16:14 sean-k-mooney stephenfin: we should not set it if its not supported
17:16:47 stephenfin OK,so I guess I should pivot that bug to say "this should *not* be set"?
17:17:04 sean-k-mooney yes i think so
17:35:01 openstackgerrit Artom Lifshitz proposed openstack/nova master: [WIP] Create claims when live migrating https://review.openstack.org/634606
17:35:02 openstackgerrit Artom Lifshitz proposed openstack/nova master: [WIP] Send dest NUMA info inside LiveMigrateData https://review.openstack.org/634828
17:35:02 openstackgerrit Artom Lifshitz proposed openstack/nova master: [WIP] New objects to transmit NUMA config from dest to source https://review.openstack.org/634827
17:35:03 openstackgerrit Artom Lifshitz proposed openstack/nova master: [Very WIP] Update NUMA XML during live migration https://review.openstack.org/635229
17:36:01 artom cfriesen, ^^ I know you wanted to play with early code. So it's still very early, but once that top [Very WIP] patch is just WIP it'll be ready for testing - by which I mean, I'll have enough code written to start actually running it myself in my dev env
17:36:24 artom So it'll most likely still be broken, actually :)
17:36:55 artom My plan is to write some automated tests with a tempest plugin we have, and iterate running those/fixing bugs until the feature works. Then unit tests will get fixed/written.
17:38:36 cfriesen artom: cool. I'm just finishing up some other downstream stuff, but thanks for the heads-up.
17:47:31 mriedem speaking of gibi's bw provider series, i'm +2 on the bottom 2 patches https://review.openstack.org/#/c/567268/
17:47:36 mriedem to help keep that moving
18:10:07 artom cfriesen, yeah, no huge rush, just an FYI
18:16:05 openstackgerrit Matt Riedemann proposed openstack/nova master: WIP: Fix legacy-grenade-dsvm-neutron-multinode-live-migration https://review.openstack.org/634962
18:31:39 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Update the PCI passthrough guides https://review.openstack.org/635243
18:31:57 stephenfin mriedem: You might have some interest in that little doc patch ^
18:32:24 stephenfin Inspired by stephenfin's struggles getting a PCI environment set up locally again
18:40:15 cdent mriedem, melwitt, jaypipes: Assuming we all still have jobs and openstack is still a thing come the opening of train, I'd like to work on making shared disk a thing. Do we have anything other than effort that is blocking that?
19:08:19 mordred cdent: I think that also assumes that the rule of law still holds somewhat and the seas haven't risen and swallowed us right?
19:08:38 mriedem cdent: unknown unknowns
19:08:52 mriedem i have a devstack-plugin-ceph patch that sets up a shared DISK_GB provider
19:08:59 mriedem but that job needs to be made multi-node to be interesting
19:09:13 mriedem https://review.openstack.org/#/c/586363/
19:09:22 mriedem plus that existing bug which enumerates some known issues and the ptg etherpad
19:09:41 mriedem the todo from the ptg was write up a spec with the known issues and how we deal with those so we have a clear picture
19:09:55 mriedem then the rest i guess we find out from integration testing

Earlier   Later