Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-02-26
22:16:59 openstackgerrit Matt Riedemann proposed openstack/nova master: Change HostManager to allow scheduling to other cells https://review.openstack.org/614037
22:16:59 openstackgerrit Matt Riedemann proposed openstack/nova master: Filter duplicates from compute API get_migrations_sorted() https://review.openstack.org/636224
22:17:51 mriedem i feel robbed when i push up 33 changes but gerrritbot only shows 20
22:17:56 aspiers haha
22:18:08 aspiers don't worry, stackalytics will get them all
22:24:43 efried aspiers: wow. Any idea why that other test was interfering?
22:25:50 aspiers efried: COMPUTE_VOLUME_MULTI_ATTACH is present when the SRIOV test is run first, I haven't figured out why yet
22:26:00 efried yeah, that's weird.
22:26:04 aspiers I was chasing completely the wrong trait thanks to testtools
22:28:21 aspiers efried: https://bugs.launchpad.net/testtools/+bug/1817800
22:28:22 openstack Launchpad bug 1817800 in testtools "assertItemsEqual output is misleading when comparing unicode with str" [Undecided,New]
22:29:37 mriedem aspiers: is the driver.capabilities at the class or member level?
22:29:42 mriedem for the LibvirtDriver?
22:29:54 mriedem or FakeDriver - whatever you're using
22:30:15 aspiers in this case it's at the class level, so yeah that would explain the lack of test isolation
22:30:27 aspiers however https://review.openstack.org/#/c/638677/1
22:30:44 mriedem are your tests using the FakeDriver though?
22:30:48 aspiers no
22:30:57 aspiers functional tests use libvirt driver
22:31:26 aspiers (probably foolishly) I stacked that fix on top of your rebased capabilities-to-traits patch which is currently failing
22:31:37 aspiers if I'd put it underneath, it would probably have fixed this
22:31:40 aspiers let me try that now
22:42:30 aspiers BOOM. Works!
22:45:38 openstackgerrit Adam Spiers proposed openstack/nova master: Change LibvirtDriver.capabilities to an instance variable https://review.openstack.org/638677
22:52:14 openstackgerrit Adam Spiers proposed openstack/nova master: Convert driver supported capabilities to compute node provider traits https://review.openstack.org/538498
23:02:34 openstackgerrit Theodoros Tsioutsias proposed openstack/nova master: Add requested_networks to RequestSpec https://review.openstack.org/570201
23:02:35 openstackgerrit Theodoros Tsioutsias proposed openstack/nova master: Enable rebuild for instances in cell0 https://review.openstack.org/570203
23:03:24 openstackgerrit Theodoros Tsioutsias proposed openstack/nova master: Allow rebuild for instances in PENDING state https://review.openstack.org/637585
23:03:24 openstackgerrit Theodoros Tsioutsias proposed openstack/nova master: Introduce the PENDING instance state https://review.openstack.org/566473
23:05:33 melwitt jaypipes: thanks for the reviews :D nice point in patch 2, I replied. maybe I need to redo the whole thing with user_id as non-nullable, and try to populate it in _from_db_object (the way that queued_for_delete does). I began that way originally, but didn't think of using build request for the case where cell_id=NULL :(
23:46:18 melwitt s/queued_for_delete does/BlockDeviceMapping.uuid does/
#openstack-nova - 2019-02-27
00:12:51 melwitt sigh... BuildRequest doesn't have a user_id field. only a project_id field. so, neeeeevermind
00:12:53 openstackgerrit Brin Zhang proposed openstack/nova master: Remove the string check of the flavor attribute 'swap' https://review.openstack.org/639012
00:17:40 sean-k-mooney melwitt: if you wanted to optimise the data migration slightly you coudl do it in 2 phases
00:18:06 sean-k-mooney melwitt: first add the user ids form all instace that have a user id in there request spec
00:18:33 sean-k-mooney then call done to the cell and grab it out of the instance table for the remainder
00:19:04 melwitt yeah. that might be a good idea
00:19:13 sean-k-mooney you may not be able to rely on it being present in the request spec but you can conditionally use it to minimise the down calls
00:19:23 melwitt for this, it was about whether we can make the InstanceMapping.user_id field non-nullable. and I think because of the aforementioned things, we can't
00:19:48 melwitt no user_id on BuildRequest, and user_id not guaranteed to be set on RequestSpec
00:19:49 mriedem yonglihe: i replied in https://review.openstack.org/#/c/621474/ and i'm ok with the server_groups as a list since that seems to be the consensus
00:20:38 sean-k-mooney melwitt: if this change lands in stien you could make it non nullable in train
00:21:22 sean-k-mooney the only donwside to that would be for FFU you would have to run the online migration. although i assume this could be done as an offline migration too
00:21:47 melwitt sean-k-mooney: yeah. I don't understand whether it needs to be a major object version bump or not, but I can ask about that next cycle. because it would be good to make it non-nullable whenever we can
00:22:27 melwitt yes, offline (nova-manage) migration is the one that does the cell database reads
00:23:26 melwitt we're going to have a blocker migration in T also, to ensure all migrated, and then be able to remove transitionary online/active migration code
00:23:43 sean-k-mooney ok if its an offline data migration that is executed centrally on the controller that is not too much of a headache for ffu
00:23:57 melwitt right
00:24:52 sean-k-mooney thanks for +1'ing the os-vif releae patch by the way.
00:25:17 sean-k-mooney hopefully it will get merged soon but ill follow up tomorrow if not
00:25:46 sean-k-mooney night o/
00:29:36 melwitt np, gnight
00:44:24 openstackgerrit Matt Riedemann proposed openstack/python-novaclient master: Fix changes-before values in an instance action test https://review.openstack.org/638849
00:46:54 openstackgerrit Adam Spiers proposed openstack/nova master: Convert driver supported capabilities to compute node provider traits https://review.openstack.org/538498
01:01:58 openstackgerrit Artom Lifshitz proposed openstack/nova master: Make the use of the CastAsCall fixture configurable https://review.openstack.org/639428
01:01:58 openstackgerrit Artom Lifshitz proposed openstack/nova master: RPC changes to prepare for NUMA live migration https://review.openstack.org/634605
01:01:59 openstackgerrit Artom Lifshitz proposed openstack/nova master: [WIP needs more tests] Full NUMA live migration support https://review.openstack.org/634606
01:08:19 jaypipes melwitt: hmm, how user_id isn't guaranteed on request spec or build request is very odd to me...
01:08:40 jaypipes one would think both of those would by their very nature have a user associated with them. :)
01:09:05 melwitt jaypipes: build request doesn't even have a user_id field (wah wah). but request spec waits until conductor to set project_id and user_id
01:10:10 melwitt and yeah, I agree. I think it's because user_id wasn't "required" when they were first added as tables
01:10:31 melwitt there's definitely a user associated with them. just not recorded so well
01:21:59 openstackgerrit Ghanshyam Mann proposed openstack/nova stable/queens: DNM: For testing multinode and tempest-slow job https://review.openstack.org/623992
01:47:46 openstackgerrit Merged openstack/nova master: Mention SR-IOV cold migration limitation in admin docs https://review.openstack.org/603909
01:51:09 yonglihe mriedem, got that, working on it now, thanks.
02:03:58 openstackgerrit melanie witt proposed openstack/nova master: Add get_counts() to InstanceMappingList https://review.openstack.org/638072
02:03:58 openstackgerrit melanie witt proposed openstack/nova master: Add online data migration for populating user_id https://review.openstack.org/633351
02:03:59 openstackgerrit melanie witt proposed openstack/nova master: Use instance mappings to count server group members https://review.openstack.org/638324
02:03:59 openstackgerrit melanie witt proposed openstack/nova master: WIP Count instances from mappings and cores/ram from placement https://review.openstack.org/638073
03:53:29 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Add compute service support for attach/detach root volume https://review.openstack.org/614750
04:21:06 openstackgerrit Merged openstack/python-novaclient master: Fix changes-before values in an instance action test https://review.openstack.org/638849
05:13:22 openstackgerrit Merged openstack/nova master: Avoid BadRequest error log on volume attachment https://review.openstack.org/581453
05:37:46 openstackgerrit Takashi NATSUME proposed openstack/nova master: Remove duplicate cleanup in functional tests https://review.openstack.org/636996
06:11:08 openstackgerrit Yongli He proposed openstack/nova master: Adds the server group info into show server detail API. https://review.openstack.org/621474
06:16:23 openstackgerrit Yongli He proposed openstack/nova master: Adds the server group info into show server detail API. https://review.openstack.org/621474
06:45:33 openstackgerrit Boxiang Zhu proposed openstack/nova master: Convert to raw format into rbd volume https://review.openstack.org/638080
06:59:30 openstackgerrit Michael Still proposed openstack/nova master: We no longer need rootwrap. https://review.openstack.org/554438
06:59:31 openstackgerrit Michael Still proposed openstack/nova master: Cleanup the _execute shim in nova/network. https://review.openstack.org/639581
06:59:31 openstackgerrit Michael Still proposed openstack/nova master: Move final bridge commands to privsep. https://review.openstack.org/639580
06:59:32 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Detach/Attach root volume API changes https://review.openstack.org/623981
07:19:48 openstackgerrit Yongli He proposed openstack/nova master: Adds the server group info into show server detail API. https://review.openstack.org/621474
07:30:02 openstackgerrit Yongli He proposed openstack/nova master: Add server sub-resource topology API https://review.openstack.org/621476
08:25:51 openstackgerrit Sagar Waghmare proposed openstack/nova master: db sync prints stack-trace on invalid version https://review.openstack.org/639591
08:35:29 openstackgerrit Boxiang Zhu proposed openstack/nova master: Convert to raw format into rbd volume https://review.openstack.org/638080
08:46:53 openstackgerrit garyk proposed openstack/nova master: Better handle live migration abort https://review.openstack.org/635440
09:02:34 openstackgerrit Boxiang Zhu proposed openstack/nova master: Convert to raw format into rbd volume https://review.openstack.org/638080
09:20:32 openstackgerrit Lee Yarwood proposed openstack/nova master: WIP status: Ensure templated connection URLs are rendered before use https://review.openstack.org/639607
09:24:40 openstackgerrit Balazs Gibizer proposed openstack/nova master: Handle placement error during re-schedule https://review.openstack.org/639608
09:24:40 openstackgerrit Balazs Gibizer proposed openstack/nova master: Fup for the bandwidth series https://review.openstack.org/639159
09:32:18 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Detach/Attach root volume API changes https://review.openstack.org/623981
09:40:41 aspiers can anyone help me understand why this failed? http://logs.openstack.org/77/638677/2/gate/grenade-py3/e4fb83b/
11:25:04 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Detach/Attach root volume API changes https://review.openstack.org/623981
11:47:59 bauzas gibi: hola
11:48:32 bauzas gibi: do you know where we set VCPU inventories for https://review.openstack.org/#/c/631559/ ?
11:49:03 bauzas FWIW, we have a VCPU inventory that only accepts two VCPUs :)
11:49:12 bauzas {u'VCPU': {u'allocation_ratio': 16.0, u'total': 2, u'reserved': 0, u'step_size': 1, u'min_unit': 1, u'max_unit': 2}, u'MEMORY_MB': {u'allocation_ratio': 1.5, u'total': 4096, u'reserved': 512, u'step_size': 1, u'min_unit': 1, u'max_unit': 4096}, u'DISK_GB': {u'allocation_ratio': 1.0, u'total': 128, u'reserved': 0, u'step_size': 1, u'min_unit': 1, u'max_unit': 128}}
11:49:12 bauzas (Pdb) compute_inventory
11:50:13 bauzas oh my bad
11:52:07 sean-k-mooney bauzas: technically the vcpu inventory will accept 32 cpu allocation but each allcoation can only contin 2
11:52:28 bauzas yup, so that's not the problem I have then

Earlier   Later