| *** mriedem has quit IRC | 00:07 | |
| *** ttsiouts has quit IRC | 00:15 | |
| openstackgerrit | Merged openstack/nova master: fix tox python3 overrides https://review.openstack.org/572974 | 00:39 |
|---|---|---|
| openstackgerrit | Merged openstack/nova master: doc follow https://review.openstack.org/#/c/572195 https://review.openstack.org/573091 | 00:39 |
| *** takashin has joined #openstack-placement | 00:50 | |
| *** tetsuro has joined #openstack-placement | 00:57 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all nested providers in tree https://review.openstack.org/559480 | 01:10 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Add microversion for nested allocation candidate https://review.openstack.org/565487 | 01:10 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Fix nits in placement-return-all-resources series https://review.openstack.org/573604 | 01:10 |
| openstackgerrit | gaofei proposed openstack/nova master: Fix typos https://review.openstack.org/574931 | 01:22 |
| openstackgerrit | jiang wei proposed openstack/nova master: Add action initiator attribute to the payload https://review.openstack.org/536243 | 01:35 |
| openstackgerrit | Naichuan Sun proposed openstack/nova master: XenAPI: define a new image handler to use vdi streaming https://review.openstack.org/486475 | 01:59 |
| openstackgerrit | Naichuan Sun proposed openstack/nova master: XenAPI: update the document related to vdi streaming https://review.openstack.org/568444 | 02:00 |
| *** ttsiouts has joined #openstack-placement | 02:37 | |
| *** ttsiouts has quit IRC | 02:39 | |
| *** ttsiouts has joined #openstack-placement | 02:50 | |
| *** ttsiouts has quit IRC | 02:55 | |
| openstackgerrit | Zack Cornelius proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876 | 03:35 |
| *** ttsiouts has joined #openstack-placement | 03:55 | |
| openstackgerrit | Merged openstack/nova master: Implement certificate_utils https://review.openstack.org/479949 | 03:59 |
| openstackgerrit | Merged openstack/nova master: Do not use nova.test in placement.test_microversion https://review.openstack.org/574408 | 04:00 |
| openstackgerrit | Merged openstack/nova master: add consumers generation field https://review.openstack.org/557958 | 04:00 |
| *** ttsiouts has quit IRC | 04:00 | |
| openstackgerrit | Merged openstack/nova master: Plumb trusted_certs through the compute service https://review.openstack.org/561262 | 04:00 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (7) https://review.openstack.org/574974 | 04:39 |
| openstackgerrit | Merged openstack/nova master: Skip ServerShowV254Test.test_rebuild_server in cells v1 job https://review.openstack.org/569454 | 04:46 |
| openstackgerrit | Tony Breeds proposed openstack/nova stable/queens: [Stable Only] Initialise filed_builds in IronicNodeState https://review.openstack.org/574997 | 06:11 |
| *** Kevin_Zheng has joined #openstack-placement | 06:18 | |
| openstackgerrit | jichenjc proposed openstack/nova master: Not reraise DiskNotFound if instance is resized https://review.openstack.org/571410 | 06:45 |
| openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: Initial change set of z/VM driver https://review.openstack.org/523387 | 06:46 |
| openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: Spawn and destroy function of z/VM driver https://review.openstack.org/527658 | 06:46 |
| openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add snapshot function https://review.openstack.org/534240 | 06:46 |
| openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add power actions https://review.openstack.org/543340 | 06:46 |
| openstackgerrit | jichenjc proposed openstack/nova master: z/VM Driver: add get console output https://review.openstack.org/543344 | 06:46 |
| openstackgerrit | jichenjc proposed openstack/nova master: Remove support for /os-fixed-ips REST API https://review.openstack.org/568516 | 06:53 |
| *** ttsiouts has joined #openstack-placement | 07:17 | |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add rules column to instance_group_policy table. https://review.openstack.org/560832 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add InstanceGroupPolicy object https://review.openstack.org/573628 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add policy to InstanceGroup object and api models. https://review.openstack.org/563375 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add policy field to ServerGroup notification object https://review.openstack.org/563401 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Change the anti-affinity Filter to adapt to new policy https://review.openstack.org/571166 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Adapt _validate_instance_group_policy to new policy model https://review.openstack.org/571465 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Microversion 2.63 - Use new format policy in server group https://review.openstack.org/567534 | 07:19 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Fix all invalid obj_make_compatible test case https://review.openstack.org/574240 | 07:19 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Use ThreadPoolExecutor for max_concurrent_live_migrations https://review.openstack.org/563505 | 07:20 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Add full traceback to ExceptionPayload in versioned notifications https://review.openstack.org/564092 | 07:23 |
| *** cdent has joined #openstack-placement | 07:27 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Use nova.test.TestingException https://review.openstack.org/575012 | 07:39 |
| *** takashin has left #openstack-placement | 08:01 | |
| *** e0ne has joined #openstack-placement | 08:06 | |
| openstackgerrit | Merged openstack/nova master: Do not use nova.test in placement.test_deploy https://review.openstack.org/574409 | 08:09 |
| *** cdent has quit IRC | 08:09 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Compute: add support to abort queued live migration https://review.openstack.org/568542 | 08:12 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: API: add support to abort queued live migration in microversion 2.63 https://review.openstack.org/573136 | 08:14 |
| openstackgerrit | jiang wei proposed openstack/nova master: Add action initiator attribute to the payload https://review.openstack.org/536243 | 08:15 |
| *** tssurya has joined #openstack-placement | 08:20 | |
| *** nicolasbock has joined #openstack-placement | 08:57 | |
| openstackgerrit | garyk proposed openstack/nova master: Ensure that the manager does not freeze when greenpool is full https://review.openstack.org/575034 | 08:59 |
| *** nicolasbock has quit IRC | 09:12 | |
| *** ttsiouts has quit IRC | 09:16 | |
| *** ttsiouts has joined #openstack-placement | 09:21 | |
| openstackgerrit | jiang wei proposed openstack/nova master: Add action initiator attribute to the payload https://review.openstack.org/536243 | 10:08 |
| *** ttsiouts has quit IRC | 10:33 | |
| openstackgerrit | Andrey Volkov proposed openstack/nova master: Update nova network info when doing rebuild for evacuate operation https://review.openstack.org/382853 | 10:43 |
| *** cdent has joined #openstack-placement | 10:44 | |
| *** nicolasbock has joined #openstack-placement | 10:46 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Transfer port.resource_request to the scheduler https://review.openstack.org/567268 | 10:52 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Send resource allocations in the port binding https://review.openstack.org/569459 | 10:52 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Test boot with more ports with bandwidth request https://review.openstack.org/573317 | 10:52 |
| *** ttsiouts has joined #openstack-placement | 11:17 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Remove network info stubbing in functional test https://review.openstack.org/570018 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add request_spec.RequestGroup versioned object https://review.openstack.org/568840 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add requested_resources field to RequestSpec https://review.openstack.org/567267 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add bandwidth related standard resource classes https://review.openstack.org/570847 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Transfer port.resource_request to the scheduler https://review.openstack.org/567268 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Send resource allocations in the port binding https://review.openstack.org/569459 | 12:05 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Test boot with more ports with bandwidth request https://review.openstack.org/573317 | 12:05 |
| *** edmondsw has joined #openstack-placement | 12:07 | |
| *** bhagyashri_s is now known as bhagyashris | 12:56 | |
| openstackgerrit | jiang wei proposed openstack/nova master: Add action initiator attribute to the payload https://review.openstack.org/536243 | 12:59 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Validate transport_url in nova-manage cell_v2 commands https://review.openstack.org/569858 | 13:04 |
| openstackgerrit | Eric Fried proposed openstack/os-traits master: normalize_name helper https://review.openstack.org/560107 | 13:21 |
| openstackgerrit | Dan Smith proposed openstack/nova stable/queens: [Stable Only] Initialise failed_builds in IronicNodeState https://review.openstack.org/574997 | 13:28 |
| openstackgerrit | Dan Smith proposed openstack/nova stable/queens: [Stable Only] Initialise failed_builds in IronicNodeState https://review.openstack.org/574997 | 13:28 |
| *** mriedem has joined #openstack-placement | 13:30 | |
| mriedem | cdent: efried: jaypipes: please make sure my eyes aren't deceiving me https://review.openstack.org/#/c/565405/23/nova/api/openstack/placement/objects/resource_provider.py@1917 | 13:31 |
| jaypipes | mriedem: reading it now. | 13:32 |
| jaypipes | mriedem: responded. | 13:34 |
| cdent | i concur | 13:34 |
| mriedem | jaypipes: ok you're doing that in this change or a follow up? | 13:34 |
| mriedem | oh i see -W | 13:35 |
| mriedem | ack | 13:35 |
| jaypipes | mriedem: actively fixing it right now. | 13:35 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Add status field to version document at / https://review.openstack.org/575117 | 13:46 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix bug to api-ref https://review.openstack.org/571375 | 13:49 |
| jaypipes | mriedem: guh... | 13:55 |
| jaypipes | mriedem: so, in that case, AllocationList.get_by_consumer_id() will return 0 records even when there are allocations for that consumer when the online data migration has not been run. | 13:59 |
| jaypipes | mriedem: I can fix that if you think it's worthwhile. | 13:59 |
| jaypipes | mriedem: in fact, I *did* just fix it.. | 13:59 |
| jaypipes | mriedem: I can *push* the fix if you think it's worthwhile :) | 13:59 |
| jaypipes | mriedem: we have online data migrations. then we have "really online" data migrations. | 14:01 |
| mriedem | jaypipes: allocations for that consumer - even though we didn't record the consumer info, right? | 14:03 |
| mriedem | i think that's what you mean, we have allocs but not the consumer record b/c the allocs were created w/o a project/user | 14:03 |
| jaypipes | right | 14:03 |
| mriedem | i don't think we can regress that otherwise, unless you want to make it a separate change *before* the one that adds the Allocation.consumer field (if that's possible?) - ensuring the consumer at the time we read the allocation is a kind of online data migration we've done quite a few times in nova, | 14:04 |
| alex_xu | efried: quick question, what case we will generate traits dynamically, in case https://review.openstack.org/#/c/560107 | 14:04 |
| mriedem | like moving things from cell dbs to api dbs | 14:04 |
| jaypipes | this is handling the micro-scenario for when some user calls AllocationList.get_all_by_consumer_id() before the operator has run the online data migrations and we want that user to get accurate info back, not an empty list | 14:04 |
| mriedem | jaypipes: yeah returning an empty list would be a regression so we can't do that imo | 14:05 |
| mriedem | maybe see if dansmith agrees on the 'really online' data migration nature of this | 14:05 |
| mriedem | for example, | 14:05 |
| * dansmith reads | 14:06 | |
| mriedem | that would probably screw up heal_allocations because it would think there are no allocations for a given instance and then overwrite them | 14:06 |
| jaypipes | mriedem: I have it already fixed/coded... | 14:06 |
| mriedem | jaypipes: with a test? | 14:06 |
| jaypipes | mriedem: it ain't gonna hurt anything, and I've added a TODO to remove. | 14:06 |
| jaypipes | mriedem: yep. | 14:06 |
| mriedem | i'm ok with it if dan is | 14:06 |
| efried | alex_xu: Example would be a trait corresponding to a feature name that we pull straight off of the hardware. Or a vendor/product ID. | 14:06 |
| dansmith | I'm not sure I understand.. just doing the migration of that consumer on read? | 14:07 |
| mriedem | dansmith: https://review.openstack.org/#/c/565405/23/nova/api/openstack/placement/objects/resource_provider.py@1917 | 14:07 |
| mriedem | dansmith: but yes | 14:07 |
| dansmith | mriedem: then yes | 14:07 |
| mriedem | dansmith: as coded, ^ will make GET /allocations/{consumer_id} not return anything if there is no consumer record | 14:07 |
| mriedem | which is a regression | 14:07 |
| mriedem | so let's just online migrate on read | 14:07 |
| efried | alex_xu: Here's a concrete example, actually. In PowerVM, we're at least initially going want to support consumers attaching a *specific* device. In that case, we may wish to tag the device with its location code, the POWER equivalent of a PCI address, which can contain characters outside of the accepted set. | 14:08 |
| alex_xu | efried: ah....I guess that is coming from gpu? I thought jaypipes will hate vendor/projuct ID in the trait | 14:08 |
| dansmith | mriedem: yeah that's uncool for sure | 14:08 |
| mriedem | jaypipes: bombs away | 14:08 |
| dansmith | mriedem: migrate on read is a common pattern, so not sure why we wouldn't do that if we have to migrate in order to return the things we need | 14:08 |
| mriedem | jaypipes: probably good to call it out in the commit message also | 14:08 |
| mriedem | since it's more than just a refactor then | 14:08 |
| efried | alex_xu: Yes. | 14:08 |
| efried | to both | 14:08 |
| mriedem | dansmith: agree | 14:08 |
| mriedem | just making sure | 14:08 |
| alex_xu | efried: oops, I shouldn't ask that | 14:09 |
| * mriedem notes alex_xu must now go back to work in the salt mines | 14:10 | |
| efried | alex_xu: Example of a Power loc code: U78CB.001.WZS0JZB-P1-C10 | 14:10 |
| jaypipes | mriedem: ack | 14:10 |
| efried | alex_xu: I think they can sometimes have * in them too | 14:11 |
| * alex_xu should never ask that kind of awkward question next time | 14:11 | |
| efried | alex_xu: Where were you going with that anyway, suggesting that we shouldn't have normalize_name at all? | 14:12 |
| openstackgerrit | Dan Smith proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876 | 14:13 |
| efried | jaypipes: if you get a chance, this should be a quick and easy one: https://review.openstack.org/#/c/560107/ | 14:15 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Enforce placement minimum in nova.cmd.status https://review.openstack.org/568965 | 14:21 |
| efried | mriedem: This ^ is going to fail because it wasn't in place when I submitted https://review.openstack.org/#/c/515811/ | 14:22 |
| efried | I need to bump the cmd.status minimum and rebase ^ on top of that. | 14:22 |
| mriedem | efried: ack ping me when it's up | 14:23 |
| efried | rgr | 14:23 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] replace deprecated accept.best_match https://review.openstack.org/575127 | 14:26 |
| cdent | efried: ^ another quicky bug fix | 14:27 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Add status field to version document at / https://review.openstack.org/575117 | 14:39 |
| cdent | alex_xu: updated docs on ^ | 14:39 |
| efried | cdent: I think that bug report has a bad copy/paste | 14:44 |
| efried | cdent: never mind | 14:44 |
| cdent | efried: no worries. pause a minute if you're thinking about +w the status field one. may be one more channge | 14:46 |
| efried | cdent: ack | 14:46 |
| efried | cdent: I wasn't going to fast-approve that. | 14:46 |
| *** ttsiouts has quit IRC | 14:51 | |
| *** ttsiouts has joined #openstack-placement | 14:53 | |
| *** ttsiouts has quit IRC | 14:56 | |
| *** ttsiouts has joined #openstack-placement | 14:57 | |
| openstackgerrit | Eric Fried proposed openstack/nova master: Update nova-status & docs: require placement 1.25 https://review.openstack.org/575140 | 15:00 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Enforce placement minimum in nova.cmd.status https://review.openstack.org/568965 | 15:00 |
| efried | mriedem: ^ | 15:00 |
| mriedem | efried: i think i've mentioned this before, but to bludgeon you again https://review.openstack.org/#/c/515811/20/releasenotes/notes/granular-extra-specs-50b26b8f63717942.yaml@2 | 15:05 |
| mriedem | efried: just one thing on https://review.openstack.org/#/c/575140/ | 15:07 |
| efried | mriedem: ack on the docs for granular - I'm guessing if you mentioned it before, my response was: waiting until nrp is in place. | 15:11 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Update nova-status & docs: require placement 1.25 https://review.openstack.org/575140 | 15:15 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Enforce placement minimum in nova.cmd.status https://review.openstack.org/568965 | 15:15 |
| efried | mriedem: Done | 15:15 |
| openstackgerrit | Brianna Poulos proposed openstack/nova master: Add trusted_image_certificates to REST API https://review.openstack.org/486204 | 15:17 |
| openstackgerrit | Brianna Poulos proposed openstack/nova master: Add notification support for trusted_certs https://review.openstack.org/563269 | 15:17 |
| openstackgerrit | Brianna Poulos proposed openstack/nova master: Add certificate validation docs https://review.openstack.org/560158 | 15:17 |
| openstackgerrit | Brianna Poulos proposed openstack/nova master: WIP: Add trusted certs to feature support matrix docs https://review.openstack.org/574890 | 15:17 |
| openstackgerrit | Brianna Poulos proposed openstack/nova master: Remove max_size parameter from fake_libvirt_utils.fetch_*image methods https://review.openstack.org/574911 | 15:17 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Add status and links fields to version document at / https://review.openstack.org/575117 | 15:19 |
| *** efried has quit IRC | 15:20 | |
| *** efried has joined #openstack-placement | 15:22 | |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] replace deprecated accept.best_match https://review.openstack.org/575127 | 15:27 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Add status and links fields to version document at / https://review.openstack.org/575117 | 15:35 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: placement: Allocation.consumer field https://review.openstack.org/565405 | 15:37 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: rework allocation handler _allocations_dict() https://review.openstack.org/565407 | 15:37 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: Add a microversion for consumer generation support https://review.openstack.org/565604 | 15:37 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: Fix nits from change Id609789ef6b4a4c745550cde80dd49cabe03869a https://review.openstack.org/574324 | 15:37 |
| jaypipes | mriedem, efried, cdent: ^^ ok, I think I addressed all your comments. | 15:37 |
| efried | ack | 15:37 |
| cdent | I think I will pause to cut the grass | 15:41 |
| *** ttsiouts has quit IRC | 15:43 | |
| * jroll hands cdent some scissors | 15:44 | |
| mriedem | jaypipes: that's not really what i was expecting - i was expecting a targeted data migration for the one missing consumer per set of allocations | 16:10 |
| cdent | thanks jroll | 16:12 |
| *** tssurya has quit IRC | 16:18 | |
| openstackgerrit | Eric Berglund proposed openstack/nova master: Powervm configuration cleanup https://review.openstack.org/575171 | 16:47 |
| openstackgerrit | Eric Berglund proposed openstack/nova master: Powervm configuration cleanup https://review.openstack.org/575171 | 16:48 |
| *** e0ne has quit IRC | 17:03 | |
| jaypipes | mriedem: I decided to go this route because I wanted to re-use as much code as possible | 17:06 |
| jaypipes | mriedem: and not add one-off functions that do pretty much the same thing just for one consumer. | 17:07 |
| mriedem | jaypipes: i get that but i don't think it's something we can do | 17:07 |
| jaypipes | mriedem: how so? | 17:07 |
| mriedem | as noted inline, if the 1001st consumer is the one i needed, we'll get an error | 17:07 |
| jaypipes | mriedem: there is a while True: there. | 17:07 |
| mriedem | plus it's a perf hit for every time through here | 17:07 |
| mriedem | rather than 1. consumer is missing, 2. migrate for that consumer, 3. done - and don't need to do it again | 17:08 |
| mriedem | oh right missed the while True | 17:08 |
| jaypipes | mriedem: we would need to do the same for each consumer returned in AllocationList.get_by_resource_provider() as well. | 17:08 |
| jaypipes | mriedem: so that would be two one-off places that did pretty much the same thing. | 17:08 |
| mriedem | these one off online data migrations are pretty common | 17:09 |
| mriedem | and get dropped over time | 17:09 |
| mriedem | think of all the online data migratoins to add uuids to records in the nova cell db | 17:09 |
| dansmith | jaypipes: with that, we'd end up with one api thread suddenly going off the rails to migrate the whole db, and if another came along, it would start doing the same because transactions, right? | 17:09 |
| jaypipes | mriedem: yes, I'm aware of them. I wrote a good number of them. | 17:09 |
| mriedem | this is why i figured it would be a separate change | 17:10 |
| jaypipes | dansmith: it's a single trx per 1000 records. | 17:10 |
| jaypipes | dansmith: it's not like it would be interwoven. | 17:10 |
| openstackgerrit | Jan Gutter proposed openstack/nova-specs master: Spec to implement vRouter HW offloads https://review.openstack.org/567148 | 17:11 |
| dansmith | jaypipes: well, two concurrent requests will both try to migrate the first 1000 records yeah? | 17:11 |
| jaypipes | mriedem: so do you want me to undo this? | 17:11 |
| jaypipes | dansmith: in single queries, but yes. | 17:11 |
| mriedem | jaypipes: i know i don't want to migrate everything in a GET request yes | 17:12 |
| dansmith | yeah, so I think what'll happen is all concurrent queries will be a random mixed bag of failbait until one of them succeeds in finishing the whole thing | 17:12 |
| mriedem | i expected a one-off per consumer | 17:12 |
| jaypipes | is this a case of us trying too hard to fix every little edge case all in one? | 17:12 |
| dansmith | jaypipes: ask me what I want you to do | 17:12 |
| dansmith | jaypipes: no it's not | 17:12 |
| jaypipes | dansmith: absolutely nothing would have been "broken" without these changes. | 17:12 |
| dansmith | jaypipes: except nothing will work until the migration happened right? | 17:12 |
| mriedem | well, allocations that don't already have consumers wouldn't be returned | 17:13 |
| jaypipes | dansmith: the only thing that would have occurred is AllocationList.get_all_by_consumer_id() would have return 0 records for a time period before the operator ran the migration. | 17:13 |
| dansmith | jaypipes: right, which could be days | 17:13 |
| jaypipes | dansmith: only for old ass allocations. | 17:13 |
| dansmith | jaypipes: some people don't run those until they hit the blocker for the *next* upgrade | 17:13 |
| jaypipes | mriedem, dansmith: honestly, why do we even bother with having anything in the nova-manage online_data_migrations CLI tool, then? if every time I need to write a "really online" migration that does exactly the same thing. | 17:14 |
| mriedem | i know that heal_allocations would heal these 'missing' allocations for the old ones with no consumers if nothing was returned, but we can't really assume people are going to be running that unless they hit problems | 17:14 |
| dansmith | jaypipes: because not every record will be touched, so you have to make sure you've finished the job at some point | 17:15 |
| mriedem | so you can run it offline? | 17:15 |
| dansmith | and so you can run it offline, but that's actually the second reason | 17:15 |
| jaypipes | mriedem: why bother at all with the CLI tool if every time I'm going to be asked to write a "really online" migration that only operates on a subset of the data? | 17:15 |
| dansmith | jaypipes: most of the ones we've had in the past were written as "really online"and the CLI one just tickled those to happen | 17:15 |
| dansmith | so it wasn't two | 17:15 |
| dansmith | but you're optimizing really deep, which is good, which means you have two ways of doing it | 17:15 |
| mriedem | i guess i don't see this as any different than the several 'add a missing uuid value on read' that we've done in the past | 17:16 |
| mriedem | except it's obviously not as simple to write that | 17:16 |
| dansmith | the only one I know of that wasn't like that was the flavor one and we did that because flavors aren't user-created and we simplified some things by not having it happen "really online" | 17:17 |
| dansmith | mriedem: right, semantically it's the same | 17:17 |
| dansmith | not as straightforward, granted, but.. | 17:17 |
| jaypipes | just tell me what you want me to do, then. you want me to remove the highly optimal "do this all in a single INSERT query" routine that was originally routined and only do the suboptimal "insert one record at a time when you notice it" approach? | 17:18 |
| dansmith | jaypipes: I want you to be on PTO like you're supposed to, let us deal with this problem, or wait until you get back. Because I really don't want to have to call your mother... :) | 17:18 |
| dansmith | and | 17:19 |
| jaypipes | dansmith: thx, but I'd prefer to push past this and get it done. so just let me know your prefernce and I'll make it happen. | 17:19 |
| mriedem | i prefer the migrate single record on read pattern we use everywhere | 17:20 |
| dansmith | jaypipes: the second option is not suboptimal for use within a single query. whether or not you have two for single and bulk, or make the bulk one less efficient by using the same routine, I don't really care | 17:20 |
| mriedem | but also that we can hack on this too | 17:20 |
| jaypipes | mriedem: *only* that? i.e. remove the original consumer_obj.ensure_incomplete_consumers()? | 17:20 |
| mriedem | no that's used via the CLI right? | 17:21 |
| dansmith | yes | 17:21 |
| jaypipes | mriedem: yes | 17:21 |
| mriedem | we need both | 17:21 |
| dansmith | right | 17:21 |
| jaypipes | ok, gimme fifteen minutes. | 17:21 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova master: Add InstanceNUMATopology to LibvirtLiveMigrateData https://review.openstack.org/566398 | 17:24 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova master: libvirt live migration: fit NUMA topology on dest https://review.openstack.org/567242 | 17:24 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova master: Service version check for NUMA live migration https://review.openstack.org/566723 | 17:24 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova master: WIP: Libvirt live migration: update NUMA XML for dest https://review.openstack.org/575179 | 17:24 |
| openstackgerrit | Merged openstack/nova master: api-ref: expand on various bdm parameters https://review.openstack.org/574805 | 17:50 |
| jaypipes | dansmith, mriedem: ok, was 45 mins, not 15.... | 18:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: placement: Allocation.consumer field https://review.openstack.org/565405 | 18:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: rework allocation handler _allocations_dict() https://review.openstack.org/565407 | 18:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: Add a microversion for consumer generation support https://review.openstack.org/565604 | 18:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: Fix nits from change Id609789ef6b4a4c745550cde80dd49cabe03869a https://review.openstack.org/574324 | 18:04 |
| *** openstackgerrit has quit IRC | 18:04 | |
| mriedem | that gave me plenty of time for hot dogs and tater salad | 18:14 |
| mriedem | the lunch of champions | 18:14 |
| jaypipes | mriedem: nice. | 18:14 |
| jaypipes | mriedem, dansmith: lemme know if the above was more what you were expecting. thx. | 18:14 |
| *** openstackgerrit has joined #openstack-placement | 18:16 | |
| openstackgerrit | Dan Smith proposed openstack/nova master: Always read-deleted=yes on lazy-load https://review.openstack.org/575190 | 18:16 |
| *** e0ne has joined #openstack-placement | 18:20 | |
| openstackgerrit | Zack Cornelius proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876 | 18:23 |
| *** mriedem has quit IRC | 18:46 | |
| *** mriedem has joined #openstack-placement | 18:47 | |
| *** tssurya has joined #openstack-placement | 18:56 | |
| *** cdent has quit IRC | 19:26 | |
| *** e0ne has quit IRC | 19:34 | |
| *** e0ne has joined #openstack-placement | 19:40 | |
| *** e0ne has quit IRC | 19:43 | |
| mriedem | oooooo yeah https://www.youtube.com/watch?v=6U8JlcB_BzA | 19:49 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Add trusted_image_certificates to REST API https://review.openstack.org/486204 | 19:53 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Add notification support for trusted_certs https://review.openstack.org/563269 | 19:53 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Add certificate validation docs https://review.openstack.org/560158 | 19:53 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: WIP: Add trusted certs to feature support matrix docs https://review.openstack.org/574890 | 19:53 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove max_size parameter from fake_libvirt_utils.fetch_*image methods https://review.openstack.org/574911 | 19:53 |
| *** e0ne has joined #openstack-placement | 19:56 | |
| *** e0ne has quit IRC | 19:58 | |
| mriedem | we should probably ignore these warnings in test runs | 20:20 |
| mriedem | UserWarning: Policy placement:resource_providers:aggregates:list failed scope check. The token used to make the request was project scoped but the policy requires ['system'] scope. This behavior may change in the future where using the intended scope is required | 20:20 |
| mriedem | since they are super spammy | 20:20 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Don't heal allocations for deleted servers https://review.openstack.org/575222 | 20:24 |
| *** ttsiouts has joined #openstack-placement | 20:34 | |
| *** openstackgerrit has quit IRC | 21:04 | |
| *** openstackgerrit has joined #openstack-placement | 21:20 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Ignore UserWarning for scope checks during test runs https://review.openstack.org/575237 | 21:20 |
| *** tssurya has quit IRC | 21:22 | |
| *** nicolasbock has quit IRC | 21:27 | |
| *** edmondsw has quit IRC | 21:33 | |
| *** edmondsw has joined #openstack-placement | 21:35 | |
| *** edmondsw has quit IRC | 21:40 | |
| *** takashin has joined #openstack-placement | 21:52 | |
| mriedem | jaypipes: thanks, i think there is just a couple of typos to fix in https://review.openstack.org/#/c/565405/ and then it looks great | 21:57 |
| *** ttsiouts has quit IRC | 22:15 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: placement: Allocation.consumer field https://review.openstack.org/565405 | 22:19 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: rework allocation handler _allocations_dict() https://review.openstack.org/565407 | 22:19 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Add a microversion for consumer generation support https://review.openstack.org/565604 | 22:19 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix nits from change Id609789ef6b4a4c745550cde80dd49cabe03869a https://review.openstack.org/574324 | 22:19 |
| mriedem | jaypipes: fixed the nits, +2 | 22:37 |
| *** mriedem has quit IRC | 22:40 | |
| openstackgerrit | Zack Cornelius proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876 | 22:42 |
| *** ttsiouts has joined #openstack-placement | 22:56 | |
| *** ttsiouts has quit IRC | 23:01 | |
| *** tssurya has joined #openstack-placement | 23:22 | |
| *** tssurya has quit IRC | 23:27 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!