Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-03-11
08:46:10 openstackgerrit Takashi NATSUME proposed openstack/nova master: doc: Fix a typo https://review.openstack.org/642382
09:08:04 openstackgerrit Balazs Gibizer proposed openstack/nova master: fix race in test_interface_detach_with_port_with_bandwidth_request https://review.openstack.org/642387
09:49:54 gibi bauzas: hi! there is a simple test only patch https://review.openstack.org/#/c/641309/ that need a second core
10:10:12 openstackgerrit zhaixiaojun proposed openstack/nova master: Fix bug: non-admin user can not filter instance by instance-uuid https://review.openstack.org/642403
10:16:47 openstackgerrit zhaixiaojun proposed openstack/nova master: Fix bug: non-admin user can not filter instance by instance-uuid https://review.openstack.org/642403
10:23:29 openstackgerrit Surya Seetharaman proposed openstack/nova master: Delete orphaned RS and IM records of archived instances https://review.openstack.org/560042
10:34:37 openstackgerrit ya.wang proposed openstack/nova master: Update docs: User token times out during long-running operations https://review.openstack.org/642410
11:54:51 openstackgerrit Balazs Gibizer proposed openstack/nova master: Use Selection object to fill request group mapping https://review.openstack.org/638711
12:09:05 openstackgerrit Merged openstack/nova master: doc: Fix a typo https://review.openstack.org/642382
12:51:10 awalende So, I have a flavor which takes 720gb memory. An instance running on stress with this flavor is taking 770gb+ memory...the limit of my machine. Linux Kernel kicks in and stops the qemu process resulting in a crashed instance. I have writeback cache enabled. How can I estimate the true memory usage of a guest?
12:56:15 efried awalende: I'm afraid I can't help you with that one. Others in this channel might be able to, if you hang around for a bit. However, it sounds more like a qemu/libvirt question than an OpenStack or nova one.
13:00:46 awalende Ah, okay. I'll ask around. Thanks :)
13:04:04 openstackgerrit Surya Seetharaman proposed openstack/nova master: Delete orphaned RS and IM records of archived instances https://review.openstack.org/560042
13:04:06 efried good luck
13:19:57 Alon_KS Does Openstack support multiple iothreads for 1 RBD (disk)?
13:20:44 Alon_KS or do i must use multiple disks to take advantage of multiple iothreads?
13:23:12 efried Alon_KS: The couple of people whose names I associate with "knows something about RBD" aren't in channel at the moment. Can you check back in a couple of hours and tag cfriesen or sean-k-mooney ?
13:24:02 Alon_KS sure! will do, Thanks efried
13:24:30 efried good luck
13:30:54 efried n-sch meeting in 30 minutes in #openstack-meeting-alt
13:41:07 openstackgerrit Balazs Gibizer proposed openstack/nova master: Reproduce bug #1819460 in functional test https://review.openstack.org/642444
13:41:08 openstack bug 1819460 in OpenStack Compute (nova) "instance stuck in BUILD state due to unhandled exceptions in conductor" [Low,New] https://launchpad.net/bugs/1819460 - Assigned to Balazs Gibizer (balazs-gibizer)
13:43:35 openstackgerrit Eric Fried proposed openstack/nova master: WIP: remove fake_libvirt_utils.create_image https://review.openstack.org/642445
13:46:48 zhubx Can anyone else here take some time to help to review https://review.openstack.org/#/c/640271/ Thanks.
13:47:23 mriedem gibi: a few small things in https://review.openstack.org/#/c/642387/
13:48:48 gibi mriedem: looking
13:54:48 gibi mriedem: gerrit died on me but as far as I see you are correct. I will respin the patch soon
13:55:23 mriedem ack
13:55:26 mriedem yeah dead for me too
13:55:43 edleafe I had just tried to post a long comment when it died on me. <sigh>
14:19:52 openstackgerrit sahid proposed openstack/os-vif master: ci: re-enable openstack-python35-jobs https://review.openstack.org/642465
14:26:39 openstackgerrit Balazs Gibizer proposed openstack/nova master: Reproduce bug #1819460 in functional test https://review.openstack.org/642444
14:26:41 openstack bug 1819460 in OpenStack Compute (nova) "instance stuck in BUILD state due to unhandled exceptions in conductor" [Low,New] https://launchpad.net/bugs/1819460 - Assigned to Balazs Gibizer (balazs-gibizer)
14:26:41 openstackgerrit Balazs Gibizer proposed openstack/nova master: Handle placement error during re-schedule https://review.openstack.org/642470
14:30:32 openstackgerrit Balazs Gibizer proposed openstack/nova master: Handle placement error during re-schedule https://review.openstack.org/639608
14:36:07 openstackgerrit Matt Riedemann proposed openstack/nova master: Use Selection object to fill request group mapping https://review.openstack.org/638711
14:36:28 mriedem easy optimization for the bw series ^
14:38:57 openstackgerrit Balazs Gibizer proposed openstack/nova master: fix race in test_interface_detach_with_port_with_bandwidth_request https://review.openstack.org/642387
14:39:05 gibi mriedem, efried: fixed the nits in ^^
14:39:42 efried gibi: +2 thanks
14:39:45 gibi efried: thanks
14:40:08 gibi efried: there is a functinal test coverage only patch as well https://review.openstack.org/#/c/641309/ that needs a second core
14:40:37 efried gibi: Yeah, I saw that and tried to grok it last week but gave up :)
14:40:57 gibi efried: no worries I will try to hook bauzas into that
14:41:11 efried or somebody who knows wtf a macvtap is
14:41:36 gibi yeah :)
14:41:39 efried one of these days I'm going to have to learn *anything* about networking.
14:42:34 gibi efried: sean-k-mooney
14:42:37 gibi efried: sean-k-mooney can help
15:01:06 mriedem efried: i can't say i fully grok this ironicclient interface change https://review.openstack.org/#/c/640879/ - questions inline
15:05:35 mriedem is the issue that on stable (queens or pike), valid_interfaces as an option didn't exist, only interface, and they have interface=internal but we don't pass that along or something, so when we get the ksa adapter it returns the public endpoint which we use to construct the ironicclient?
15:07:46 efried mriedem: ack, looking.
15:34:58 mriedem efried: also comments on your 'stop using PlacementDirect' change https://review.openstack.org/#/c/640887/
15:40:04 efried ack
15:53:52 openstackgerrit Balazs Gibizer proposed openstack/nova master: Documentation for bandwidth support https://review.openstack.org/642064
16:20:26 zbr we need some input regarding changing default NovaHWMachineType value to make it cross distro compatible: https://review.openstack.org/#/c/642443/
16:20:56 weshay dansmith bauzas can you folks recommend a libvirt vm machine types? https://review.openstack.org/#/c/642443/1/deployment/nova/nova-compute-container-puppet.yaml is pc > pc-i440fx-rhel7.6.0
16:21:36 dansmith weshay: kashyap is probably your best bet for that question, but he and bauzas are at a church retreat this week
16:21:51 weshay ah k
16:22:06 weshay thanks dansmith
16:22:13 dansmith yup
16:23:18 mriedem zbr: i'd ask kashyap but he's not in channel right now
16:23:27 mriedem oh heh
16:23:29 mriedem jinx
16:23:49 zbr mriedem: please add those that can help to this review.
16:24:04 mriedem done
16:24:21 zbr i am curious why we used a pinned version instead of the "pc" one.
16:58:36 openstackgerrit Merged openstack/nova master: fix race in test_interface_detach_with_port_with_bandwidth_request https://review.openstack.org/642387
17:39:01 openstackgerrit guang-yee proposed openstack/nova master: pass endpoint interface to Ironic client https://review.openstack.org/640879
17:44:39 openstackgerrit Chris Dent proposed openstack/os-traits master: Update bug links for storyboard https://review.openstack.org/642529
17:51:42 openstackgerrit Chris Dent proposed openstack/os-resource-classes master: Update bug links for storyboard https://review.openstack.org/642530
19:16:33 melwitt o/
19:34:13 openstackgerrit Adrian Chiris proposed openstack/nova master: SR-IOV Live migration indirect port support https://review.openstack.org/620115
19:34:28 openstackgerrit Merged openstack/os-traits master: Update bug links for storyboard https://review.openstack.org/642529
19:36:47 openstackgerrit Eric Fried proposed openstack/nova master: Stop using PlacementDirect https://review.openstack.org/640887
19:37:00 efried mriedem, cdent: ^
19:38:10 efried Hi melwitt. In case you haven't seen this: https://review.openstack.org/#/c/641391/
19:38:23 efried I don't think there's really contention anymore, but in case you want to weigh in.
19:39:20 openstackgerrit Adrian Chiris proposed openstack/nova master: libvirt: auto detach/attach sriov ports on migration https://review.openstack.org/629589
19:39:30 melwitt efried: I saw and had thought it looked like folks had agreed to hold until train. thanks for confirming
19:39:39 efried rgr
19:42:56 openstackgerrit Merged openstack/os-resource-classes master: Update bug links for storyboard https://review.openstack.org/642530
19:43:44 mriedem melwitt: https://blueprints.launchpad.net/nova/stein should be fully up to date now as well
19:44:32 melwitt mriedem: fantastic, thank you
20:23:36 openstackgerrit Eric Fried proposed openstack/nova master: DNM: inline fake_libvirt_utils.create_image mocks https://review.openstack.org/642445
20:23:54 openstackgerrit Michael Still proposed openstack/nova master: Make privsep.chown mocking for libvirt snapshot tests less magic. https://review.openstack.org/642134
20:23:55 openstackgerrit Michael Still proposed openstack/nova master: Remove fake_libvirt_utils from the cache concurrency tests. https://review.openstack.org/642556
20:23:55 openstackgerrit Michael Still proposed openstack/nova master: Remove fake_libvirt_utils from connection tests. https://review.openstack.org/642557
20:23:56 openstackgerrit Michael Still proposed openstack/nova master: Remove fake_libvirt_utils from snapshot tests. https://review.openstack.org/642558
20:25:07 efried mikal: o/
20:27:45 mikal efried: Hi
20:28:27 efried mikal: In case you wanted to chat realtime about https://review.openstack.org/#/c/641489/ (fake_libvirt_utils)
20:29:18 mikal I'm happy to chat, but wont really be online for a couple of hours. I need to get the kids off to school first. I would be interested in what you think of the alternative approach used above though.
20:30:10 efried ah, looking.
20:32:08 efried mikal: oic, this way you gotta do it (at least) one whole module at a time. I was going to do it one mocked method at a time. I think the former makes for fewer, bigger patches. But I like it.
20:32:58 efried I thought there was a base class somewhere that was inherited by multiple test modules though. If that's the case, you'll either have to do all of those in one big swatch; or move the monkeypatch from the base into the sub-modules so you can piecemeal em.
20:43:11 mikal efried: there isn't much left after those four patches -- test_virt_drivers (small); and test_imagebackend (bigger, but not huge). There is some cleanup afterwards as the module is removed, which I was going to do method by method, but it keeps it reviewable, incremental, and games the statistics to make us all look good.
20:43:45 mikal efried: Oh, and don't discount the added benefit of long patch series using a lot more gate resources than a single big patch. We wouldn't want the gate to get bored.
20:51:14 efried mikal: my series was probably going to be longer.
20:51:23 efried mikal: I'm on board with your approach

Earlier   Later