Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-08-22
19:54:39 mriedem so uh, i could use some reviews on this rc3 potential fix https://review.openstack.org/#/c/594508/
19:54:41 mriedem "fix"
19:55:42 melwitt I'll review the rc3 things today. it's been a distracting day
19:56:59 mriedem i'll review yours if you review mine
19:57:24 mriedem the other one, approved yesterday morning, is still humping its way through the gate
20:01:04 melwitt ok, cool. I saw on the todos etherpad
20:28:36 mriedem dansmith: i'm slowly climbing multi-cell list batch mountain
20:28:41 mriedem nearly at the summit
20:28:49 mriedem oxygen....depleting...
20:41:21 melwitt I found that the console code in compute is storing token auths in both the database and nova-consoleauth for rocky, and I'm struggling to understand why. it was like that before I took on the patch series. I assume it was just so operators could keep using nova-consoleauth as the storage backend for a cycle, if they for some reason wanted to
20:41:48 melwitt initially, I thought maybe it was to support rolling upgrades, but I don't think storing new tokens in nova-consoleauth helps with that
20:46:11 eandersson Is compute_driver used by any other service then nova-compute?
20:46:51 eandersson Can you have nova-compute-ironic on the same box as nova-conductor, api etc?
20:56:38 tssurya mriedem: ack, I'll repropose..
20:57:25 openstackgerrit Dan Smith proposed openstack/nova master: Make CELL_TIMEOUT a constant https://review.openstack.org/594570
20:57:26 openstackgerrit Dan Smith proposed openstack/nova master: Stash the cell uuid on the context when targeting https://review.openstack.org/594571
20:57:27 openstackgerrit Dan Smith proposed openstack/nova master: Make RecordWrapper record RequestContext and expose cell_uuid https://review.openstack.org/594572
20:57:28 openstackgerrit Dan Smith proposed openstack/nova master: Clarify which context is used by do_query() https://review.openstack.org/595250
20:57:29 openstackgerrit Dan Smith proposed openstack/nova master: Batch results per cell when doing cross-cell listing https://review.openstack.org/592698
20:57:30 openstackgerrit Dan Smith proposed openstack/nova master: List instances from all cells explicitly https://review.openstack.org/593717
20:57:31 openstackgerrit Dan Smith proposed openstack/nova master: Make instance_list perform per-cell batching https://review.openstack.org/593131
20:57:32 openstackgerrit Dan Smith proposed openstack/nova master: Record cell success/failure/timeout in CrossCellLister https://review.openstack.org/594265
20:57:33 openstackgerrit Dan Smith proposed openstack/nova master: Optimize global marker re-lookup in multi_cell_list https://review.openstack.org/594577
21:00:03 melwitt eandersson: no, compute_driver shouldn't be used for any other service (though I see one check for it in the novnc proxy since the xen driver doesn't support a new enough novnc version, IIRC). and AFAIK, you could run nova-compute-ironic on the same box as nova-conductor, etc, but I'm not sure it's recommended. maybe TheJulia can advise
21:00:40 eandersson I see - we figured as much, and didn't see any issues
21:00:55 TheJulia best to keep it to where you run ironic-conductor or ironic-api anyhow
21:01:20 TheJulia minimize latency between connections and also keep access to logs relatively close if they are needed
21:01:36 mriedem the api looks at compute_driver in one case for bdm device name
21:01:37 mriedem i think
21:01:38 TheJulia But that is just my take for sanity :)
21:01:52 mriedem if you're using xen
21:02:05 melwitt ok, I didn't see that in my grepping
21:04:24 mriedem eandersson: http://codesearch.openstack.org/?q=is_xenapi&i=nope&files=&repos=
21:04:37 mriedem http://git.openstack.org/cgit/openstack/nova/tree/nova/cmd/novncproxy.py#n43
21:10:20 eandersson Thanks!
21:41:06 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Add /reshaper handler for POST https://review.openstack.org/576927
22:53:55 openstackgerrit Merged openstack/nova master: Filter out instances without a host when populating AZ https://review.openstack.org/594050
22:54:02 openstackgerrit Merged openstack/nova master: Remove ChanceScheduler https://review.openstack.org/576384
#openstack-nova - 2018-08-23
00:04:12 openstackgerrit Sam Morrison proposed openstack/nova master: Allow ability for non admin users to use all filters on server list. https://review.openstack.org/526558
00:36:25 openstackgerrit Matt Riedemann proposed openstack/nova master: [placement] Add /reshaper handler for POST https://review.openstack.org/576927
00:47:20 mriedem anyone, off the top of your head, does shelve of a volume-backed server create a snapshot of the root volume like createImage will?
00:48:50 mriedem oh we don't even snapshot for volume-backed
00:48:53 mriedem we just offload
00:49:46 mriedem well hells to the yeah
00:49:52 mriedem something went right with shelve for a change
01:12:41 openstackgerrit Takashi NATSUME proposed openstack/nova master: Adds view builders for keypairs controller https://review.openstack.org/347289
01:16:58 Kevin_Zheng :)
02:37:37 openstackgerrit zhufl proposed openstack/nova master: Blacklist test_create_server_with_tags for nova-cells-v1 https://review.openstack.org/595399
03:24:15 openstackgerrit Chen proposed openstack/nova master: Normalize dashless 'resource provider create' uuid https://review.openstack.org/567191
04:12:18 openstackgerrit Chen proposed openstack/nova master: Fix create_resource_provider docstring https://review.openstack.org/595453
04:57:33 openstackgerrit melanie witt proposed openstack/nova master: Correct the release notes related to nova-consoleauth https://review.openstack.org/595455
06:50:49 openstackgerrit zhufl proposed openstack/nova master: Blacklist test_create_server_with_tags for nova-cells-v1 https://review.openstack.org/595399
07:41:18 zigo http://paste.openstack.org/show/728659/ <--- Excuse my french but ... WTF ?!?
08:25:29 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Only lock when race condition happens in context.set_target_cell https://review.openstack.org/595530
09:21:02 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Only lock when race condition happens in context.set_target_cell https://review.openstack.org/595530
09:39:44 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Add functional test to verify presence of policy https://review.openstack.org/595559
11:33:14 openstackgerrit Moshe Levi proposed openstack/nova master: libvirt: set vfio driver in interface hostdev https://review.openstack.org/595592
11:34:23 mriedem ahoy mateys!
11:42:05 sean-k-mooney mriedem: so i have not got a chance to test your patches for the livemigration issue yet(ill do that today) but i have noticed one other issue.
11:42:27 sean-k-mooney its not so much a bug just something unfortunate
11:44:01 sean-k-mooney mriedem: when hybrid_plug=false and we are not useing dpdk we hit https://github.com/openstack/os-vif/blob/master/vif_plug_ovs/ovs.py#L216-L223
11:44:37 sean-k-mooney in this config os-vif only creates the bridge and the actual pluggin is done by the hypervior e.g. libvirt
11:44:55 sean-k-mooney so we dont get the speed up on live migrate
11:47:16 mriedem what speed up on live migrate?
11:47:17 sean-k-mooney i know how to fix this in two way and ill start working on it after i test your changes but ya. i noticed up to a 20 second network outage when using kernel ovs with ovs contrac firewall driver but none when using ip tables as a result.
11:47:58 sean-k-mooney ya mriedem effectivly the plug does not happen untill libvirt creates the vm on the dest.
11:48:04 mriedem ok
11:48:08 mriedem i'll take your word for it :)
11:48:14 jaypipes morning supernovas
11:48:32 mriedem hello jay
11:48:43 sean-k-mooney mriedem: its the same behavior as before multiple port bindings so not a regression or bug just not faster.
11:48:50 sean-k-mooney jaypipes: enjoy your PTO?
11:48:54 jaypipes mriedem: luckily not much happened or was discussed on the ML while I was away
11:49:05 jaypipes sean-k-mooney: very much, thank you Sean. :)
11:49:31 mriedem gibi: can you take a look at this rc3-related bug workaround? https://review.openstack.org/#/c/594508/ - rc3 is supposed to be today
11:49:53 mriedem gibi: but withhold a +W until sean-k-mooney is able to test it in his env
11:51:07 gibi mriedem: ack, witholding my love but reading the patch ;)
11:53:20 sean-k-mooney mriedem: oh i forgot RC3 was today. i need to restack but i should have the env running in 20 mins or so
11:54:36 mriedem sean-k-mooney: *no pressure!*
12:09:38 mriedem moshele: can you report a bug for https://review.openstack.org/#/c/595592/ ?
12:09:41 mriedem it's a regression in rocky
12:11:07 moshele mriedem: sure I will
12:11:09 mriedem melwitt: ^
12:11:19 mriedem throw it on the big board donny!
12:12:05 mriedem it's fun
12:13:36 mriedem sahid: you should take a look at moshele's patch
12:13:46 mriedem moshele: what sets the driver_name to vhost?
12:14:54 moshele mriedem: the macvtap port which is virtio tap connected to VF so for that case it should be fine
12:14:54 sahid mriedem ah yes this bug has been reported downstream yesterday i think
12:15:00 sahid moshele: thanks for your patch
12:15:18 moshele sahid: np :)
12:15:44 mriedem moshele: sorry i don't understand that statement,
12:15:46 mriedem looking at your paste
12:15:46 mriedem : libvirtError: unsupported configuration: Unknown PCI device <driver name='vhost'/> has been specified
12:15:57 mriedem and looking at https://review.openstack.org/#/c/595592/1/nova/virt/libvirt/designer.py
12:16:06 mriedem i guess in this case the net_type was 'direct'?
12:16:11 mriedem so L120
12:17:17 sahid mriedem: are you sure vfio is a valid driver?
12:17:22 sahid moshele:
12:17:55 ykarel jroll, hi

Earlier   Later