Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-04-02
02:19:01 openstackgerrit Takashi NATSUME proposed openstack/nova master: [placement] api-ref: Fix parameters https://review.openstack.org/558142
02:42:48 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi(N-R-P):Get vgpu info from `allocations` https://review.openstack.org/521717
02:47:38 tetsuro @fired_bunny: Oh, I realized your message just now. Thanks in advance! I’m going to look into your patch on the top of the branch after lunch. Anyway, Good night!
03:47:07 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi: Use XAPI pool instead of aggregate pool for shared SR migration https://review.openstack.org/554154
05:47:03 openstackgerrit Maciej Kucia proposed openstack/nova-specs master: Fix Non-unique network names in Servers IPs API response https://review.openstack.org/558125
05:50:15 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix the request context in ServiceFixture https://review.openstack.org/558160
06:15:00 openstackgerrit Nguyen Hai proposed openstack/nova-specs master: Enhance nova-specs webpage and clean up repo https://review.openstack.org/551802
06:18:26 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi: Support live migration in pooled multi-nodes environment https://review.openstack.org/489451
06:20:05 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi: Use XAPI pool instead of aggregate pool for shared SR migration https://review.openstack.org/554154
06:35:55 yuqian #openstack-ceilometer
07:19:32 openstackgerrit Chen Hanxiao proposed openstack/nova master: libvirt: guest: introduce blockStats instead of domain.blockStats https://review.openstack.org/526833
07:29:33 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove mox in tests/unit/test_utils.py https://review.openstack.org/557883
07:59:54 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: [RFC] Pick next minimum libvirt / QEMU versions for "Stein" https://review.openstack.org/558171
08:38:30 openstackgerrit Naichuan Sun proposed openstack/nova master: Use update_provider_tree from resource tracker https://review.openstack.org/520246
08:38:30 openstackgerrit Naichuan Sun proposed openstack/nova master: SchedulerReportClient.update_from_provider_tree https://review.openstack.org/533821
08:38:31 openstackgerrit Naichuan Sun proposed openstack/nova master: Make generation optional in ProviderTree https://review.openstack.org/539324
08:38:31 openstackgerrit Naichuan Sun proposed openstack/nova master: Fix nits in update_provider_tree series https://review.openstack.org/531260
08:38:32 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi(N-R-P):Get vgpu info from `allocations` https://review.openstack.org/521717
08:38:32 openstackgerrit Naichuan Sun proposed openstack/nova master: xenapi(N-R-P): Add API to support vgpu resource provider create https://review.openstack.org/520313
08:38:33 openstackgerrit Naichuan Sun proposed openstack/nova master: (WIP)xenapi(N-R-P): Add API to support compute node resource provider update and create https://review.openstack.org/521041
09:46:00 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix a missing white space in exception message https://review.openstack.org/558183
10:35:33 openstackgerrit Wonil Choi proposed openstack/nova master: SR-IOV passthrough: Check PF only if VF is enabled https://review.openstack.org/476642
12:04:22 jaypipes morning supernovas
13:03:07 Spaz-Home Morning folks
13:35:03 Spaz-Home Whew nerve wracking day
13:55:52 edleafe Scheduler subteam meeting in 5 minutes in #openstack-meeting-alt
14:20:04 Tahvok Hey guys! I've changes a flavor to more cpus, which resulted many of instances to have not available flavor. I've changed back the cpu to what it was on the flavor, and now I want to attach back the flavor to the instances.
14:20:23 Tahvok Is there way to just re-attache the flavor to the instances, without actually doing a full resize?
14:39:10 bhagyashris melwitt: Hi, Could you please review https://review.openstack.org/#/c/511825/ ? addressed review comments thank you :)
14:45:17 bhagyashris jaypipes: Hi, As per our dicussion on friday i have followed the steps and test it but still it's taking the compute node DISK_GB inventory and not shared RP disk_gb inventory to boot the instance
14:47:33 jaypipes bhagyashris: so your compute node is not reporting any DISK_GB inventory?
14:48:41 bhagyashris jaypipes: yesh i have removed the DISK_GB inventory for CN RP
14:48:46 bhagyashris yeah*
14:49:36 jaypipes bhagyashris: if that is the case, an instance cannot possibly be consuming inventory of DISK_GB against the compute node...
14:49:42 bhagyashris jaypipes: and one more point that I have observed is after booting the instance I show the DISK_GB inventory record for CN RP
14:51:21 jaypipes bhagyashris: so the compute node resource tracker must be automatically re-adding the DISK_GB inventory to the compute node provider...
14:51:25 jaypipes hmmm
14:51:51 bhagyashris jaypipes: yeah so it seems like patch: https://review.openstack.org/#/c/533396/8 is not reporting issue
14:55:10 jaypipes bhagyashris: ok, thanks for that. looks like we are going to need some method of indicating to a compute node not to automatically inventory local disk space
14:55:14 jaypipes fried_bunny: ^^
14:55:24 fried_bunny noted
14:55:49 jaypipes but first... I must eat something.
14:55:57 jaypipes bhagyashris: I'll have a think on this today.
14:56:26 fried_bunny Guess we'll have to figure out where that inventory is coming from. bhagyashris Which driver? libvirt?
14:56:49 bhagyashris fried_bunny: yes
14:57:19 fried_bunny bhagyashris: code level?
14:57:24 bhagyashris i am using libvirt driver
14:57:50 bhagyashris yes
14:58:01 fried_bunny What level of code? Queens release? Master?
14:58:28 bhagyashris fried_bunny: on current master
14:59:00 fried_bunny jaypipes, bhagyashris: It should be noted that we don't officially support shared providers yet. So I'm not sure how broken this really is.
14:59:06 bhagyashris fried_bunny: commit: 02e357e7c2d2ccbab0a1f6b5a807d11f1ef72d46
15:00:08 fried_bunny The DISK_GB inventory that's showing up in your CN RP - is it the inventory you expected from your sharing RP, or is it actually the system's local disk?
15:00:31 bhagyashris fried_bunny: not getting clearly you mean to say the shared RP is not supported on current master?
15:00:53 fried_bunny That's correct.
15:01:01 fried_bunny We were just talking about this in the sched meeting.
15:01:43 fried_bunny We could "fix" this "problem", but it's part of a larger thing that we've been carrying from release to release for quite a while.
15:02:07 fried_bunny jaypipes will have to answer whether there's some aspect of this that counts as a regression.
15:02:38 fried_bunny otherwise... I suspect we may have to wait to fix this properly.
15:04:51 bhagyashris fried_bunny: currently in inventory table the CN RP and shared RP DISK_GB value as same but what i expect is when i boot instance at that time it should take DISK_GB from shared Rp and VPCU and MEMoR_MB from CN RP
15:05:59 fried_bunny bhagyashris: What I'm trying to find out is whether we're reporting the inventory from the sharing provider as if it was local; or if we're reporting the local inventory itself and ignoring the shared. Can you perhaps resize one of those disks so that we can tell which is happening?
15:10:11 edleafe ugh, yeah I see 'em
15:10:17 edleafe fried_bunny: ^^
15:11:26 bhagyashris fried_bunny: it still taking the CN RP inventory and ignores the shareed RP inventory
15:17:12 fried_bunny bhagyashris: Okay. Like I said, unless we can declare this to be a regression, it's probably not going to be something we "fix" right now. Let's ask jaypipes when he gets back from breakfast.
15:18:58 bhagyashris fried_bunny: ok thank you for your inputs as i am working on IST time zone so it's almost EOD so i will discuss tomorrow :)
15:20:05 fried_bunny bhagyashris: Roger that. I'll try to follow up with jaypipes and one or the other of us will let you know tomorrow (either in IRC or we'll send an email)
15:39:59 openstackgerrit Eric Fried proposed openstack/nova master: Don't declare traits synced if they ain't https://review.openstack.org/558068
15:39:59 openstackgerrit Eric Fried proposed openstack/nova master: Test case: traits don't sync if first access fails https://review.openstack.org/558066
15:46:11 openstackgerrit Eric Berglund proposed openstack/nova master: WIP: PowerVM: Cold Migrate & Resize https://review.openstack.org/553583
15:48:30 fried_bunny jaypipes, edleafe: https://review.openstack.org/#/c/558068/1/nova/api/openstack/placement/objects/resource_provider.py@142
15:54:24 edleafe fried_bunny: in the middle of my own rabbit hole :) Will look soon
15:54:29 fried_bunny oy vay
15:55:02 jaypipes zzzeek: hey, what is the magic incantation to allow an enginefacade-managed session to start and close its own transaction?
15:55:34 zzzeek jaypipes: do you mean, you want to explicitly do that part yourself?
15:55:45 zzzeek jaypipes: the enginefacade ohhh
15:55:50 jaypipes zzzeek: allowing the transaction to be committed and exceptions raised in "outer" enginefacade sessions won't roll the inner one back
15:55:53 zzzeek jaypipes: you mean indepdendent of the rest
15:55:57 jaypipes ya
15:56:02 zzzeek jaypipes: yeah writer.independent
15:56:16 jaypipes oh crap, why can't I ever remember it. thanks zzzeek :)
15:56:18 zzzeek jaypipes: that means whole new DB connection for that block
15:56:34 fried_bunny jaypipes: Okay, so I decorate _trait_sync with that, and we're good?
15:57:03 zzzeek jaypipes fried_bunny well use .independent sparingly if it's in a tight loop you are putting a lot more load on things
15:57:16 jaypipes fried_bunny: yes.
15:57:20 fried_bunny zzzeek: Cool man, it should only get run once, ever.
15:57:21 zzzeek then again before enginefacade *all* of nova was like that :)
15:57:28 zzzeek fried_bunny: then you're golden
15:57:53 fried_bunny zzzeek, jaypipes: works. Thanks!
15:58:07 jaypipes np (all hail zzzeek)
15:58:15 zzzeek jaypipes: so true! :)
16:02:44 openstackgerrit Eric Fried proposed openstack/nova master: Use an independent transaction for _trait_sync https://review.openstack.org/558068
16:02:47 fried_bunny jaypipes, zzzeek: ^
16:02:49 fried_bunny Thanks for the help.
16:32:39 openstackgerrit Eric Berglund proposed openstack/nova master: PowerVM Driver: vSCSI volume driver https://review.openstack.org/526094
16:47:47 edleafe Is there some reason you can't add columns to two different tables in a single migration? IOW, do they have to be done separately?
16:49:22 edleafe oh, nvm - stupid typo
17:53:15 openstackgerrit Vladyslav Drok proposed openstack/nova master: Ignore some updates from virt driver https://review.openstack.org/523006
18:00:28 openstackgerrit Ed Leafe proposed openstack/nova master: Add handlers for consumer_generation https://review.openstack.org/557959
18:00:28 openstackgerrit Ed Leafe proposed openstack/nova master: Add columns for generation to consumers and allocations https://review.openstack.org/557958

Earlier   Later