*** ducttape_ has joined #openstack-nova | 00:15 | |
*** dave-mccowan has quit IRC | 00:17 | |
*** Fdaisuke has quit IRC | 00:31 | |
*** ducttape_ has quit IRC | 00:35 | |
*** yingjun has joined #openstack-nova | 00:37 | |
*** catintheroof has quit IRC | 00:42 | |
*** catintheroof has joined #openstack-nova | 00:43 | |
*** ducttape_ has joined #openstack-nova | 00:47 | |
*** stvnoyes has joined #openstack-nova | 00:50 | |
*** tuanluong has joined #openstack-nova | 00:53 | |
*** hfu has joined #openstack-nova | 00:55 | |
*** hoangcx has joined #openstack-nova | 00:57 | |
*** zz_dimtruck is now known as dimtruck | 00:58 | |
*** ducttape_ has quit IRC | 00:59 | |
*** ducttape_ has joined #openstack-nova | 01:00 | |
*** phuongnh has joined #openstack-nova | 01:01 | |
*** ducttape_ has joined #openstack-nova | 01:01 | |
*** mlavalle has quit IRC | 01:09 | |
*** masayukig_ is now known as masayukig | 01:16 | |
*** tyrola has quit IRC | 01:27 | |
*** tyrola has joined #openstack-nova | 01:28 | |
oomichi | mriedem: is it possible to get your review on https://review.openstack.org/#/c/413453/ ? | 01:33 |
---|---|---|
oomichi | that seems necessary to be backported to stable branches also for avoiding an internal error | 01:33 |
*** yingjun has quit IRC | 01:39 | |
*** Jeffrey4l has quit IRC | 01:39 | |
*** Jeffrey4l has joined #openstack-nova | 01:39 | |
*** ducttape_ has quit IRC | 01:42 | |
*** ducttape_ has joined #openstack-nova | 01:43 | |
*** ducttape_ has quit IRC | 01:47 | |
*** ducttape_ has joined #openstack-nova | 01:48 | |
*** yufei has joined #openstack-nova | 01:50 | |
oomichi | mikal: are you around here today? | 01:54 |
*** ducttape_ has quit IRC | 01:59 | |
*** hongbin has quit IRC | 02:03 | |
*** gongysh has joined #openstack-nova | 02:05 | |
*** ducttape_ has joined #openstack-nova | 02:16 | |
*** gongysh has quit IRC | 02:20 | |
*** hfu_ has joined #openstack-nova | 02:22 | |
*** hfu has quit IRC | 02:23 | |
*** baoli has joined #openstack-nova | 02:24 | |
*** baoli has quit IRC | 02:28 | |
*** ducttape_ has quit IRC | 02:32 | |
takashin | oomichi: Could you review https://review.openstack.org/#/c/323287/, https://review.openstack.org/#/c/413075/, https://review.openstack.org/#/c/308746/ ? | 02:35 |
takashin | oomichi: They got one +2. They need more +2 and W+1. | 02:35 |
*** hfu_ has quit IRC | 02:40 | |
*** gcb has quit IRC | 02:52 | |
*** gcb has joined #openstack-nova | 02:55 | |
*** dave-mccowan has joined #openstack-nova | 02:55 | |
*** dimtruck is now known as zz_dimtruck | 02:56 | |
*** yingjun has joined #openstack-nova | 02:59 | |
*** sacharya has joined #openstack-nova | 03:08 | |
*** hfu has joined #openstack-nova | 03:09 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571 | 03:12 |
*** hongbin has joined #openstack-nova | 03:13 | |
*** mdnadeem has joined #openstack-nova | 03:13 | |
*** yamahata has quit IRC | 03:14 | |
*** sacharya_ has joined #openstack-nova | 03:24 | |
*** sacharya has quit IRC | 03:24 | |
*** baoli has joined #openstack-nova | 03:26 | |
*** baoli has quit IRC | 03:26 | |
*** mdnadeem has quit IRC | 03:30 | |
*** baoli has joined #openstack-nova | 03:31 | |
*** ducttape_ has joined #openstack-nova | 03:32 | |
openstackgerrit | Spencer Yu proposed openstack/nova: [3/3]Replace six.iteritems() with .items() https://review.openstack.org/415080 | 03:34 |
*** ducttape_ has quit IRC | 03:37 | |
*** links has joined #openstack-nova | 03:39 | |
*** gongysh has joined #openstack-nova | 03:41 | |
*** ducttape_ has joined #openstack-nova | 03:45 | |
*** dave-mcc_ has joined #openstack-nova | 03:52 | |
*** xinliang has quit IRC | 03:53 | |
*** xinliang has joined #openstack-nova | 03:54 | |
*** dave-mccowan has quit IRC | 03:55 | |
*** ducttape_ has quit IRC | 03:55 | |
*** tiantian has quit IRC | 03:55 | |
*** zz_dimtruck is now known as dimtruck | 03:56 | |
*** hieulq_ has quit IRC | 03:56 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: RT now adds proper Ironic inventory https://review.openstack.org/404472 | 03:59 |
openstackgerrit | Jay Pipes proposed openstack/nova: rt: explicitly pass compute node to _update() https://review.openstack.org/398472 | 03:59 |
openstackgerrit | Jay Pipes proposed openstack/nova: rt: use a single ResourceTracker object instance https://review.openstack.org/398473 | 03:59 |
*** hieulq has joined #openstack-nova | 04:00 | |
*** takedakn has joined #openstack-nova | 04:02 | |
*** psachin has joined #openstack-nova | 04:10 | |
*** dave-mcc_ has quit IRC | 04:10 | |
*** yingjun has quit IRC | 04:13 | |
*** zhs_ has joined #openstack-nova | 04:15 | |
*** zhs__ has quit IRC | 04:18 | |
*** hongbin_ has joined #openstack-nova | 04:25 | |
*** baoli has quit IRC | 04:25 | |
*** hongbin has quit IRC | 04:26 | |
*** baoli has joined #openstack-nova | 04:26 | |
*** hongbin_ has quit IRC | 04:29 | |
*** ayogi has joined #openstack-nova | 04:34 | |
*** nicolasbock has joined #openstack-nova | 04:35 | |
*** yingjun has joined #openstack-nova | 04:36 | |
*** baoli has quit IRC | 04:39 | |
*** baoli has joined #openstack-nova | 04:40 | |
*** baoli has quit IRC | 04:45 | |
alex_xu | vdrok: just done the review | 04:46 |
*** Jeffrey4l has quit IRC | 04:57 | |
*** Jeffrey4l has joined #openstack-nova | 04:58 | |
*** takedakn has quit IRC | 04:58 | |
*** baoli has joined #openstack-nova | 05:00 | |
*** pbandark has joined #openstack-nova | 05:01 | |
*** ducttape_ has joined #openstack-nova | 05:03 | |
openstackgerrit | Alex Xu proposed openstack/nova: Use JSON-Schema to validate query parameters for keypairs API https://review.openstack.org/398755 | 05:04 |
*** baoli has quit IRC | 05:04 | |
*** prateek has joined #openstack-nova | 05:05 | |
*** haplo37_ has quit IRC | 05:06 | |
*** ducttape_ has quit IRC | 05:07 | |
*** baoli has joined #openstack-nova | 05:09 | |
*** baoli has quit IRC | 05:14 | |
*** haplo37_ has joined #openstack-nova | 05:15 | |
*** tbachman has joined #openstack-nova | 05:18 | |
*** yingjun has quit IRC | 05:21 | |
*** xinliang has quit IRC | 05:28 | |
*** vtech has joined #openstack-nova | 05:37 | |
*** tbachman has quit IRC | 05:42 | |
*** vtech has quit IRC | 05:46 | |
*** yingjun has joined #openstack-nova | 05:49 | |
*** baoli has joined #openstack-nova | 05:50 | |
*** mal66 has joined #openstack-nova | 05:50 | |
*** ratailor has joined #openstack-nova | 05:52 | |
*** baoli has quit IRC | 05:54 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 3 test cases https://review.openstack.org/394738 | 05:55 |
vdrok | alex_xu: thanks! | 06:01 |
alex_xu | vdrok: np | 06:01 |
openstackgerrit | Alex Xu proposed openstack/nova: Limit the min length of string for integer JSON-Schema https://review.openstack.org/415103 | 06:13 |
alex_xu | oomichi: ^ a fix for our json-schema | 06:14 |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571 | 06:14 |
*** hfu has quit IRC | 06:22 | |
*** gongysh has quit IRC | 06:23 | |
*** adisky_ has joined #openstack-nova | 06:24 | |
oomichi | takashin: one question about https://review.openstack.org/#/c/323287/34/nova/tests/unit/cells/test_cells_scheduler.py | 06:25 |
oomichi | self is added to _fake_instance_update_at_top(), but it is not to fake_build_request_spec() What is difference between them? | 06:26 |
oomichi | takashin: both seem to be used for stub_out() | 06:27 |
oomichi | takashin: ah, maybe it is because the difference of target methods. | 06:28 |
*** ircuser-1 has joined #openstack-nova | 06:30 | |
*** moshele has joined #openstack-nova | 06:31 | |
oomichi | takashin: +W on the unit test one | 06:32 |
*** Satya__ has joined #openstack-nova | 06:32 | |
oomichi | alex_xu: are you around? | 06:32 |
alex_xu | oomichi: yea, i'm here | 06:33 |
takashin | oomichi: fake_build_request_spec() is not an instance mehod. | 06:33 |
oomichi | alex_xu: is it possible to remove your score on https://review.openstack.org/#/c/413075 ? | 06:33 |
*** nandal has joined #openstack-nova | 06:33 | |
oomichi | alex_xu: the description doesn't seem good for users | 06:33 |
takashin | The fake mehods that is added 'self' argument are instance methods. | 06:33 |
*** ducttape_ has joined #openstack-nova | 06:34 | |
takashin | oomichi: The fake mehods that is added 'self' argument are instance methods. | 06:34 |
*** sacharya_ has quit IRC | 06:34 | |
takashin | oomichi: That is the difference. | 06:34 |
oomichi | takashin: yeah I know. then I already +W on thhat | 06:34 |
alex_xu | oomichi: basically we use 'the action' everywhere, maybe it is ok with later patch if someone have better words? | 06:35 |
takashin | oomichi: Thank you so much. | 06:35 |
oomichi | alex_xu: Yeah I know that. OK, let discuss it later with another patch | 06:37 |
alex_xu | oomichi: cool! | 06:37 |
oomichi | alex_xu: no problem, thanks for your quick response :) | 06:38 |
alex_xu | oomichi: np | 06:38 |
*** ducttape_ has quit IRC | 06:38 | |
*** sridharg has joined #openstack-nova | 06:43 | |
oomichi | takashin: btw all patches of bp/remove-mox-ocata are already up now? | 06:46 |
oomichi | https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/remove-mox-ocata | 06:46 |
oomichi | I just want to know when we can finish it | 06:47 |
takashin | oomichi: Sorry. I don't know it. | 06:48 |
oomichi | takashin: no problem, that is just a question | 06:49 |
oomichi | I didn't want to review them *forever ;) | 06:49 |
oomichi | like remove-mox-pike, remove-mox-q**, ... | 06:50 |
*** baoli has joined #openstack-nova | 06:50 | |
oomichi | takashin: one more question about https://review.openstack.org/#/c/308746/40/nova/tests/unit/api/openstack/compute/test_block_device_mapping_v1.py | 06:53 |
oomichi | takashin: _validate_bdm() doesn't return any value like https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1332 | 06:54 |
oomichi | takashin: in test_create_instance_with_volumes_enabled_and_bdms_no_image(), the method is mocked with return_value, why? | 06:55 |
openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Trivial typo fix in spec https://review.openstack.org/415115 | 06:55 |
*** baoli has quit IRC | 06:55 | |
oomichi | in addition, in the test, the patch adds assert_called_once_with() for the method, but the other tests doesn't do that. | 06:55 |
*** tesseract has joined #openstack-nova | 06:56 | |
*** dimtruck is now known as zz_dimtruck | 06:59 | |
*** zz_dimtruck is now known as dimtruck | 06:59 | |
takashin | oomichi: I just replace mox with mock. So if '_validate_bdm' does not return any value in previous, it remains as it was. | 07:01 |
takashin | oomichi: if '_validate_bdm' returns a value int test methods in previous, it remains as it was. | 07:03 |
takashin | oomichi: assert statement is same. | 07:03 |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571 | 07:04 |
oomichi | takashin: ok, that should be fixed for actual test in this patch I think | 07:08 |
*** dimtruck is now known as zz_dimtruck | 07:09 | |
takashin | oomichi: Old tests were not good? | 07:09 |
oomichi | takashin: yeah, I think so because the private method doesn't return value at all | 07:09 |
takashin | oomichi: Thank you for your comments. I will fix them. | 07:13 |
oomichi | takashin: cool, I will check it after updating | 07:13 |
*** mjura has joined #openstack-nova | 07:14 | |
*** mjura has quit IRC | 07:20 | |
*** mjura has joined #openstack-nova | 07:20 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 7 test cases https://review.openstack.org/395436 | 07:27 |
*** pcaruana has joined #openstack-nova | 07:31 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 7 test cases https://review.openstack.org/395436 | 07:32 |
*** sam_nowitzki has joined #openstack-nova | 07:33 | |
Kevin_Zheng | oomichi: hi could you have a look on https://review.openstack.org/#/c/414529/ ? it failed heat and ceilometer gate, they have fixed in their own project already but we may still have fix it since we dont know who else is using "interface" to init novaclient | 07:36 |
*** sacharya has joined #openstack-nova | 07:36 | |
*** gongysh has joined #openstack-nova | 07:37 | |
oomichi | Kevin_Zheng: I cannot understand the patch purpose TBH | 07:39 |
oomichi | Kevin_Zheng: The patch way doesn't work for _LW() I think | 07:40 |
Kevin_Zheng | Hm, I will have a check again | 07:40 |
Kevin_Zheng | Thanks | 07:40 |
*** sacharya has quit IRC | 07:41 | |
*** ekuris has joined #openstack-nova | 07:42 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Trivial typo fix in spec https://review.openstack.org/415115 | 07:43 |
*** gcb has quit IRC | 07:46 | |
*** pcaruana has quit IRC | 07:50 | |
*** pcaruana has joined #openstack-nova | 07:51 | |
*** hfu has joined #openstack-nova | 07:55 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Trivial typo fix in spec https://review.openstack.org/415115 | 07:58 |
openstackgerrit | Spencer Yu proposed openstack/nova: [3/3]Replace six.iteritems() with .items() https://review.openstack.org/415080 | 08:02 |
*** ducttape_ has joined #openstack-nova | 08:04 | |
*** sahid has joined #openstack-nova | 08:06 | |
*** gcb has joined #openstack-nova | 08:08 | |
*** ducttape_ has quit IRC | 08:09 | |
vsaienk0 | Hello oomichi, could you please help to merge https://review.openstack.org/#/c/414214/ this patch fix issue that is a blocker for Ironic CI. | 08:10 |
oomichi | vsaienk0: ok, I am checking the code and bug report | 08:18 |
*** mal66 has quit IRC | 08:21 | |
oomichi | vsaienk0: I cannot catch the point of the patch yet. | 08:22 |
oomichi | vsaienk0: in https://review.openstack.org/#/c/414214/4/nova/scheduler/client/report.py , the change seems to skip storing the data if vcpus(and other) is 0 instead of storing 0 | 08:22 |
oomichi | vsaienk0: according to the bug report, "request stuck at nova" is a problem. I cannot follow why this patch can solve the problem | 08:24 |
vsaienk0 | oomichi: when placement client reports 0 to API request takes a lot of time (more than 6 sec) | 08:26 |
vsaienk0 | Lock "compute_resources" released by "nova.compute.resource_tracker._update_available_resource" :: held 6.935s inner | 08:26 |
vsaienk0 | oomichi: please check http://logs.openstack.org/39/404239/14/check/gate-tempest-dsvm-ironic-ipa-wholedisk-pxe_snmp-tinyipa-ubuntu-xenial-nv/8f98498/logs/screen-n-cpu.txt.gz#_2016-12-21_13_24_24_307 | 08:26 |
*** mdnadeem has joined #openstack-nova | 08:26 | |
*** nicolasbock has quit IRC | 08:27 | |
*** ekuris has quit IRC | 08:27 | |
vsaienk0 | oomichi: client shouldn't report 0 to placement API as it is not valid value and will not accepted anyway, API returns 400 | 08:27 |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142 | 08:27 |
vsaienk0 | oomichi: whit patch proposed the task takes lock for less than a second (0.148s) http://logs.openstack.org/50/414650/2/check/gate-tempest-dsvm-ironic-ipa-partition-pxe_ipmitool-tinyipa-ubuntu-xenial/9e223e7/logs/screen-n-cpu.txt.gz#_2016-12-24_08_22_59_774 | 08:30 |
vsaienk0 | s/whit/with | 08:30 |
*** zhs_ has quit IRC | 08:31 | |
*** zhs_ has joined #openstack-nova | 08:31 | |
*** ekuris has joined #openstack-nova | 08:42 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142 | 08:44 |
*** palexster has quit IRC | 08:48 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Update py35 functional tests blacklist https://review.openstack.org/415149 | 08:51 |
*** palexster has joined #openstack-nova | 08:54 | |
*** gcb has quit IRC | 08:59 | |
*** zzzeek has quit IRC | 09:00 | |
*** prateek has quit IRC | 09:00 | |
*** zzzeek has joined #openstack-nova | 09:00 | |
*** hfu has quit IRC | 09:01 | |
*** ralonsoh has joined #openstack-nova | 09:02 | |
*** baoli has joined #openstack-nova | 09:05 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove mox in test_block_device_mapping_v1.py https://review.openstack.org/308746 | 09:07 |
*** baoli has quit IRC | 09:10 | |
vsaienk0 | oomichi: I've replied in the bug https://bugs.launchpad.net/ironic/+bug/1651678/comments/14 | 09:12 |
openstack | Launchpad bug 1651678 in OpenStack Compute (nova) "boot server request randomly hanging at n-cpu side, and didn't get to Ironic" [High,In progress] - Assigned to John L. Villalovos (happycamp) | 09:12 |
*** hfu has joined #openstack-nova | 09:12 | |
*** prateek has joined #openstack-nova | 09:15 | |
*** fragatina has joined #openstack-nova | 09:16 | |
*** gcb has joined #openstack-nova | 09:17 | |
*** bhagyashris has joined #openstack-nova | 09:19 | |
oomichi | vsaienk0: I put a question about that | 09:21 |
vsaienk0 | oomichi: unfortunately I'm not familiar with placement API enough, I hope that jaypipes will check your comment. Thank you for your time! | 09:30 |
*** gongysh has quit IRC | 09:30 | |
oomichi | vsaienk0: I also am not good at placement API so well ;) Yeah, it is nice to wait for jay's response | 09:31 |
*** ducttape_ has joined #openstack-nova | 09:35 | |
*** zenoway has joined #openstack-nova | 09:38 | |
*** irenab_ has joined #openstack-nova | 09:38 | |
*** irenab_ has quit IRC | 09:38 | |
*** sacharya has joined #openstack-nova | 09:39 | |
*** ekuris has quit IRC | 09:39 | |
*** prateek has quit IRC | 09:39 | |
*** ducttape_ has quit IRC | 09:39 | |
*** zenoway has quit IRC | 09:40 | |
*** zenoway has joined #openstack-nova | 09:40 | |
*** karthiks has joined #openstack-nova | 09:42 | |
*** sacharya has quit IRC | 09:43 | |
*** ekuris has joined #openstack-nova | 09:45 | |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 09:47 |
*** nicolasbock has joined #openstack-nova | 09:47 | |
*** karthiks has quit IRC | 09:49 | |
*** karthiks has joined #openstack-nova | 09:50 | |
*** yingjun has quit IRC | 09:51 | |
*** ekuris has quit IRC | 09:51 | |
*** baoli has joined #openstack-nova | 09:52 | |
*** baoli has quit IRC | 09:56 | |
*** liusheng has quit IRC | 09:59 | |
*** liusheng has joined #openstack-nova | 10:00 | |
*** prateek has joined #openstack-nova | 10:00 | |
*** artom has joined #openstack-nova | 10:00 | |
*** artom_ has quit IRC | 10:01 | |
*** ekuris has joined #openstack-nova | 10:06 | |
*** tuanluong has quit IRC | 10:08 | |
*** sam_nowitzki has quit IRC | 10:16 | |
*** sam_nowitzki has joined #openstack-nova | 10:16 | |
*** Satya__ has quit IRC | 10:28 | |
openstackgerrit | Alex Xu proposed openstack/nova: [POC][placement] Add Traits API to placement service https://review.openstack.org/376200 | 10:29 |
*** sam_nowitzki has quit IRC | 10:30 | |
*** kristian__ has joined #openstack-nova | 10:46 | |
*** ratailor has quit IRC | 10:55 | |
*** yufei has quit IRC | 10:56 | |
*** hoangcx has quit IRC | 10:57 | |
*** ducttape_ has joined #openstack-nova | 11:05 | |
*** takashin_ has joined #openstack-nova | 11:05 | |
*** takashin has quit IRC | 11:07 | |
*** ducttape_ has quit IRC | 11:10 | |
*** slaweq has joined #openstack-nova | 11:21 | |
pkoniszewski | hey, can someone unabandon this change https://review.openstack.org/#/c/328997/ ? I'd like to push it forward | 11:22 |
*** ralonsoh has quit IRC | 11:24 | |
*** ociuhandu has quit IRC | 11:25 | |
*** slaweq has quit IRC | 11:30 | |
tdurakov | pkoniszewski: hi, won't it be enough to just resubmit a new version? | 11:34 |
*** phuongnh has quit IRC | 11:40 | |
*** zenoway has quit IRC | 11:51 | |
*** zenoway has joined #openstack-nova | 11:51 | |
*** ociuhandu has joined #openstack-nova | 11:53 | |
*** zenoway has quit IRC | 11:56 | |
*** nandal has quit IRC | 12:00 | |
*** kristian__ has quit IRC | 12:02 | |
*** kristian__ has joined #openstack-nova | 12:02 | |
pkoniszewski | tdurakov: i'm getting an error from remote that the change is closed | 12:12 |
*** prateek has quit IRC | 12:13 | |
*** rfolco has joined #openstack-nova | 12:17 | |
*** rfolco has quit IRC | 12:17 | |
*** rfolco has joined #openstack-nova | 12:18 | |
*** hfu has quit IRC | 12:20 | |
*** mdnadeem has quit IRC | 12:21 | |
*** hfu has joined #openstack-nova | 12:23 | |
*** baoli has joined #openstack-nova | 12:23 | |
*** baoli has quit IRC | 12:27 | |
*** sridharg has quit IRC | 12:33 | |
openstackgerrit | sahid proposed openstack/nova: libvirt: update the active migrations ds to support migration result https://review.openstack.org/409260 | 12:33 |
openstackgerrit | sahid proposed openstack/nova: libvirt: pass to live migration monitor the API result https://review.openstack.org/409261 | 12:33 |
*** nicolasbock has quit IRC | 12:34 | |
*** ducttape_ has joined #openstack-nova | 12:36 | |
*** takashin_ has left #openstack-nova | 12:38 | |
*** ducttape_ has quit IRC | 12:40 | |
*** sahid has quit IRC | 12:46 | |
*** BlackDex has quit IRC | 13:02 | |
*** sridharg has joined #openstack-nova | 13:09 | |
*** moshele has quit IRC | 13:23 | |
*** sahid has joined #openstack-nova | 13:24 | |
*** moshele has joined #openstack-nova | 13:25 | |
*** blaisebo_ has joined #openstack-nova | 13:27 | |
*** liverpooler has joined #openstack-nova | 13:33 | |
*** ducttape_ has joined #openstack-nova | 13:33 | |
*** yingjun has joined #openstack-nova | 13:36 | |
*** BlackDex has joined #openstack-nova | 13:36 | |
*** sacharya has joined #openstack-nova | 13:40 | |
*** haplo37_ has quit IRC | 13:41 | |
*** sacharya has quit IRC | 13:45 | |
*** haplo37_ has joined #openstack-nova | 13:50 | |
*** ducttape_ has quit IRC | 13:54 | |
*** jaypipes has joined #openstack-nova | 13:55 | |
jaypipes | mriedem: mornin. | 13:56 |
jaypipes | mriedem: you working today? | 13:56 |
*** nklenke has joined #openstack-nova | 14:00 | |
*** dave-mccowan has joined #openstack-nova | 14:01 | |
*** blaisebo_ has quit IRC | 14:14 | |
openstackgerrit | sahid proposed openstack/nova: libvirt: update the active migrations ds to support migration result https://review.openstack.org/409260 | 14:14 |
openstackgerrit | sahid proposed openstack/nova: libvirt: pass to live migration monitor the API result https://review.openstack.org/409261 | 14:14 |
*** moshele has quit IRC | 14:15 | |
jaypipes | sahid: hey, I'm fixing the unit test failures in https://review.openstack.org/#/c/400877/ | 14:16 |
jaypipes | sahid: I rebased that patch series yesterday and accidentally introduced them. | 14:16 |
*** maestropandy has joined #openstack-nova | 14:18 | |
*** maestropandy has left #openstack-nova | 14:18 | |
sahid | jaypipes: yes, no worries I have to address your comments anyways so | 14:19 |
*** dave-mccowan has quit IRC | 14:20 | |
*** ducttape_ has joined #openstack-nova | 14:20 | |
jaypipes | sahid: thx man, sorry about that | 14:20 |
*** yingjun has quit IRC | 14:20 | |
jaypipes | sahid: all those pep8 failures, BTW, are unrelated. I kicked off rechecks on those ones. | 14:21 |
*** lamt has joined #openstack-nova | 14:22 | |
*** mlavalle has joined #openstack-nova | 14:24 | |
sahid | jaypipes: ok i will check that :) | 14:29 |
*** moshele has joined #openstack-nova | 14:33 | |
*** dave-mccowan has joined #openstack-nova | 14:35 | |
pkoniszewski | sahid: hey, can you unabandon this change https://review.openstack.org/#/c/328997/ ? I'd like to push it forward | 14:44 |
*** Jeffrey4l has quit IRC | 15:02 | |
*** ociuhandu has quit IRC | 15:06 | |
*** tjones has joined #openstack-nova | 15:07 | |
*** ekuris has quit IRC | 15:07 | |
*** Jeffrey4l has joined #openstack-nova | 15:09 | |
*** tjones has quit IRC | 15:11 | |
*** Jeffrey4l has quit IRC | 15:12 | |
openstackgerrit | sahid proposed openstack/nova: objects: add new field cpu_emulator_threads_policy https://review.openstack.org/385022 | 15:14 |
openstackgerrit | sahid proposed openstack/nova: numa: add numa constraints for emulator threads policy https://review.openstack.org/385023 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: libvirt: return a CPU overhead if isolate emulator threads requested https://review.openstack.org/385364 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: objects: add new field cpus_overhead in NUMACell https://review.openstack.org/400874 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: numa: fit instance toplogy with cpu overhead onto cores https://review.openstack.org/400875 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: virt: update overhead to take into account vCPUs https://review.openstack.org/385024 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: libvirt: make emulator threads to run on the reserved pCPU https://review.openstack.org/400878 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: numa: take into account cpus overhead https://review.openstack.org/400876 | 15:15 |
openstackgerrit | sahid proposed openstack/nova: numa: update numa usage incording the cpu overhead https://review.openstack.org/400877 | 15:15 |
sahid | pkoniszewski: yes sure I did not notice that it got abandonned | 15:15 |
*** ayogi has quit IRC | 15:16 | |
sahid | jaypipes: did you noce this weird thing with 'api_samples' https://review.openstack.org/#/c/385023 | 15:17 |
sahid | i never pushed that but it looks to be there since th beginning | 15:18 |
jaypipes | sahid: yeah, that's a weird one. hoping it was a blip... | 15:19 |
*** pbandark has quit IRC | 15:21 | |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 15:24 |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: libvirt: pass libvirt reference methods to migration https://review.openstack.org/328997 | 15:24 |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: libvirt: live-migrate updates NUMA and cpus in the XML https://review.openstack.org/286744 | 15:24 |
pkoniszewski | sahid: thanks! | 15:24 |
openstackgerrit | Jay Pipes proposed openstack/nova: rt: explicitly pass compute node to _update() https://review.openstack.org/398472 | 15:27 |
openstackgerrit | Jay Pipes proposed openstack/nova: rt: use a single ResourceTracker object instance https://review.openstack.org/398473 | 15:27 |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 15:27 |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: libvirt: pass libvirt reference methods to migration https://review.openstack.org/328997 | 15:27 |
openstackgerrit | Pawel Koniszewski proposed openstack/nova: libvirt: live-migrate updates NUMA and cpus in the XML https://review.openstack.org/286744 | 15:27 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: report client handle InventoryInUse https://review.openstack.org/404990 | 15:27 |
*** ociuhandu has joined #openstack-nova | 15:27 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: minor refactor _allocate_for_instance() https://review.openstack.org/405705 | 15:27 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: RT now adds proper Ironic inventory https://review.openstack.org/404472 | 15:27 |
*** sridharg has quit IRC | 15:30 | |
*** baoli has joined #openstack-nova | 15:32 | |
*** links has quit IRC | 15:33 | |
*** psachin has quit IRC | 15:34 | |
*** dave-mccowan has quit IRC | 15:34 | |
*** zz_dimtruck is now known as dimtruck | 15:35 | |
mriedem | jaypipes: i am, had to take laura and maya to the airport this morning but now i'm home | 15:36 |
mriedem | working all week | 15:36 |
*** baoli has quit IRC | 15:36 | |
*** baoli has joined #openstack-nova | 15:39 | |
*** sacharya has joined #openstack-nova | 15:41 | |
*** baoli has quit IRC | 15:44 | |
mriedem | jianghuaw: on https://review.openstack.org/#/c/365505/ - can the citrix xenserver ci have a run against that patch with the tempest.api.compute.servers.test_attach_interfaces.AttachInterfacesTestJSON.test_create_list_show_delete_interfaces test enabled? | 15:45 |
*** sacharya has quit IRC | 15:46 | |
mriedem | which you could maybe do with a check for the patch 365505 in the CI setup so the tempest blacklist is modified | 15:46 |
*** openstackgerrit_ has joined #openstack-nova | 15:56 | |
*** openstackgerrit_ has quit IRC | 15:57 | |
*** ducttape_ has quit IRC | 15:59 | |
*** mjura has quit IRC | 16:03 | |
*** sacharya has joined #openstack-nova | 16:04 | |
*** dtp has joined #openstack-nova | 16:06 | |
*** erlon has joined #openstack-nova | 16:16 | |
*** tesseract has quit IRC | 16:22 | |
*** brad[] has quit IRC | 16:29 | |
*** brad[] has joined #openstack-nova | 16:31 | |
*** sahid has quit IRC | 16:32 | |
jaypipes | mriedem: trying to figure out wtf this is... http://logs.openstack.org/76/400876/11/check/gate-nova-pep8-ubuntu-xenial/5484680/console.html | 16:34 |
mriedem | hmm, just on this change or seeing on multiple? i'll logstash. | 16:35 |
jaypipes | mriedem: it's also on an earlier one in the same series: http://logs.openstack.org/23/385023/13/check/gate-nova-pep8-ubuntu-xenial/786f6f0/console.html | 16:36 |
jaypipes | mriedem: but those pep8 errors don't have anything to do with code in those patches. | 16:37 |
jaypipes | mriedem: those patches don't touch api_samples. | 16:37 |
mriedem | does it need to be rebased? | 16:37 |
jaypipes | shittttttttt | 16:37 |
jaypipes | this is my fault. | 16:37 |
jaypipes | mfer. | 16:37 |
jaypipes | https://review.openstack.org/#/c/385023/ | 16:37 |
mriedem | heh, that would do it :) | 16:37 |
jaypipes | ^^ accidentally added some leftover api_samples crap :( | 16:38 |
jaypipes | gah. I'm an idiot. | 16:38 |
mriedem | accidents happen | 16:38 |
*** moshele has quit IRC | 16:39 | |
mriedem | JayF: are there patches up anywhere for this yet? https://blueprints.launchpad.net/nova/+spec/ironic-rescue-mode | 16:40 |
*** toure has quit IRC | 16:41 | |
mriedem | looks like there are patches up for Ironic but they are all in merge conflict | 16:41 |
openstackgerrit | Jay Pipes proposed openstack/nova: numa: add numa constraints for emulator threads policy https://review.openstack.org/385023 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: libvirt: return a CPU overhead if isolate emulator threads requested https://review.openstack.org/385364 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: numa: fit instance toplogy with cpu overhead onto cores https://review.openstack.org/400875 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: virt: update overhead to take into account vCPUs https://review.openstack.org/385024 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: libvirt: make emulator threads to run on the reserved pCPU https://review.openstack.org/400878 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: numa: take into account cpus overhead https://review.openstack.org/400876 | 16:42 |
openstackgerrit | Jay Pipes proposed openstack/nova: numa: update numa usage incording the cpu overhead https://review.openstack.org/400877 | 16:42 |
*** kfarr has joined #openstack-nova | 16:46 | |
mriedem | gibi: are the o.vo changes needed for https://blueprints.launchpad.net/nova/+spec/json-schema-for-versioned-notifications basically all in and released? | 16:47 |
mriedem | i see 2 nova changes for that bp but they look stalled | 16:47 |
mriedem | and a single o.vo change outstanding | 16:47 |
mriedem | also, do we need to raise the minimum required o.vo in global-requirements for the changes that have been released? | 16:50 |
artom | mriedem, thanks for the -W, as you guessed, it's still WIP :) | 16:51 |
mriedem | artom: :) | 16:51 |
artom | Though I'll have to do BDM in the same patch, as it's the same microversion, no? | 16:51 |
mriedem | i'm doing my post-xmas bp status check | 16:51 |
artom | At least the API stuff | 16:51 |
mriedem | artom: for the rest api yes, | 16:52 |
mriedem | but the internal plumbing can be separate | 16:52 |
artom | mriedem, right | 16:52 |
vdrok | jaypipes sorry to bother, but could you respond in https://review.openstack.org/414214, or update if needed based on oomichi cooments? mriedem could you look at that too? ironic gate is half dead because of that :( | 16:53 |
vdrok | and good morning :) | 16:54 |
mriedem | vdrok: jaypipes: do we know why that's happening? | 16:55 |
vdrok | mriedem: yeah, ironic reports 0s, lemme find that place | 16:55 |
mriedem | yeah but why | 16:55 |
vdrok | ironic driver rather | 16:55 |
vdrok | mriedem: https://review.openstack.org/414603 | 16:56 |
vdrok | mriedem: I think that in case of "fake" nodes, we don't have any properties like cpus/memory populated. also when the node is not in a usable state, like in maintenance, the same thing happens | 16:57 |
vdrok | so that in case of maintenance, node's resources are not reported as used, but as absent, for example | 16:58 |
mriedem | or if the power state on the node is ERROR i guess | 16:58 |
vdrok | yup | 16:59 |
mriedem | hmm, so, what if we previously reported inventory for a given ironic node in the nova resource tracker, but then the node goes into error state, or maintenance mode, should nova delete the inventory for that resource provider so the scheduler doesn't think it has inventory available and try to schedule to it? | 17:00 |
mriedem | i can ask that in the review | 17:00 |
vdrok | yup, thanks. just wanted to ask to keep an eye on that if possible :) as we have like 5-6 big patches blocked on the gate because of that | 17:02 |
vdrok | and I'm not sure I know enough things about resource proivders to respond | 17:03 |
mriedem | jaypipes: ok question in https://review.openstack.org/#/c/414214/ for you | 17:03 |
jaypipes | mriedem: k | 17:03 |
*** slaweq has joined #openstack-nova | 17:04 | |
jaypipes | mriedem: soon as I'm done rebasing one of my patch series for the tenth time this week, I'll hop on that ;) | 17:05 |
*** ducttape_ has joined #openstack-nova | 17:06 | |
*** ducttape_ has quit IRC | 17:11 | |
artom | Has anyone ever noticed now block device code is a plate of spaghetti? With squid-based sauce :D | 17:13 |
mriedem | artom: mdbooth had a ML thread on that awhile back | 17:14 |
jaypipes | mriedem: good points. I'll update and throw a fresh one up. | 17:14 |
artom | mriedem, heh, I should have added an /s or something :) | 17:15 |
mriedem | artom: http://lists.openstack.org/pipermail/openstack-dev/2016-June/097529.html | 17:15 |
artom | It's clearly a known problem, but the solution would be to... what, scrap everything and do block_device_v3? | 17:15 |
mriedem | to be fair, there are only like 4 or 5 different "BDMs" in nova | 17:15 |
mriedem | dropping bdm v1 might be a decent start to eliminate some legacy cruft | 17:16 |
jaypipes | mriedem: yeah, we should totally add another type of BDM. | 17:16 |
jaypipes | mriedem: call it BDSM. | 17:16 |
mriedem | but bdm v1 is in the REST API | 17:16 |
mriedem | jaypipes: i see what you did there | 17:16 |
*** Raildo has joined #openstack-nova | 17:16 | |
jaypipes | mriedem: you like that, don't you. naughty boy. | 17:17 |
mriedem | please. more. | 17:17 |
artom | What I'm trying to figure out right now is why the device name (which is used to match the db objects to the instance xml stuff) in libvirt is different when attaching a volume, which means the tag doesn't show up in the metadata | 17:17 |
artom | It goes up and down through so many layers just to come up with a device name... | 17:17 |
artom | I'm sure there were valid historical reasons for this to be like that, but... damn | 17:18 |
jaypipes | where is mdbooth when you need him, eh? | 17:18 |
jaypipes | artom: don't be so sure :) | 17:18 |
mriedem | artom: you know the libvirt driver ignores the user-supplied device name from the REST API right? | 17:18 |
mriedem | not sure if that's part of your issue or not | 17:18 |
mriedem | artom: would a uuid on the bdm help at all? | 17:19 |
*** Raildo is now known as raildo | 17:19 | |
artom | mriedem, maybe... but it happens even if the API doesn't supply a name | 17:19 |
artom | mriedem, not unless it's also in the instance xml | 17:19 |
* artom keeps adding logs everywhere and continues digging | 17:20 | |
mriedem | artom: are you talking about this part? https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/driver.py#L7660 | 17:20 |
artom | mriedem, nah, I'm talking about what happens at attach time | 17:21 |
mriedem | sure, but are you saying that tagged_bdms here https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/driver.py#L7671 | 17:22 |
mriedem | is missing the bdm created at attach because the name doesn't match? | 17:22 |
mriedem | or is bdm.device_name just not even set on attach? | 17:22 |
artom | mriedem, https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/driver.py#L7697 fails to get a bdm | 17:22 |
mriedem | have you added logging of tagged_bdms to make sure you are seeing the one from the REST API call? | 17:23 |
artom | Last time I logged it, it was 'hda' in the instance xml (the dev.target_dev variable) and 'vdb' in the DB object | 17:23 |
artom | Hrmm, I'd have to double-check because you're sowing doubt in my mind, but I'm pretty sure there was only one volume to speak of | 17:24 |
mriedem | i assume you've been through reserve_block_device_name in the compute manager already? | 17:24 |
artom | mriedem, down that rabbit hole now | 17:25 |
mriedem | yeah, which calls the libvirt driver's get_device_name_for_instance method | 17:25 |
artom | Yep | 17:25 |
*** baoli has joined #openstack-nova | 17:25 | |
mriedem | which sets block_device_obj.device_name = None because it has to else blockinfo.get_info_from_bdm doesn't work correctly, because tight coupling is all the rage | 17:26 |
openstackgerrit | Jay Pipes proposed openstack/nova: Return uuid attribute for aggregates https://review.openstack.org/415031 | 17:26 |
artom | I've ran out of wtfs to give :) | 17:26 |
artom | I just want it to work | 17:26 |
mriedem | i would add some logging to see what get_info_from_bdm is returning | 17:27 |
artom | mriedem, yeah, doing that now, I'll figure it out eventually | 17:27 |
mriedem | having said that, | 17:27 |
mriedem | i think mdbooth has pointed out that if you reboot or recreate the domain xml in any way, the devices might change within the guest and it would probably screw this all up anyway | 17:28 |
mriedem | *changed | 17:28 |
artom | When you say within the guest, you mean in the instance xml? | 17:28 |
mriedem | i.e. if i attach 3 volumes, they are vdb, vdc and vdd, then reboot the guest, those might have changed and my tags are now out of whack | 17:28 |
mriedem | yeah | 17:29 |
artom | Because device name is all we have to match the DB objects (which contain tags) to the instance XML info (which contain everything else) | 17:29 |
artom | If that match isn't stable, we can scrap block device tagging altogether | 17:29 |
mriedem | ok, well, danpb or mdbooth or dansmith could probably elaborate on this more, but i thought that basically bdm device name can't be trusted in the xml vs the bdm db | 17:30 |
*** baoli has quit IRC | 17:30 | |
*** gouthamr has joined #openstack-nova | 17:32 | |
artom | mriedem, there should be a way to retroactively adjust the DB to match the XML, I'd think... I'll push a WIP that passes the new Tempest test and ask for feedback | 17:33 |
artom | This device name thing is all that's left before I can do that | 17:33 |
mriedem | artom: looking at http://logs.openstack.org/03/414603/1/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/1d0aeb9/logs/screen-n-cpu.txt.gz#_2016-12-23_15_49_19_009 | 17:38 |
mriedem | there is a vda bdm in that xml | 17:38 |
mriedem | and the serial is the volume uuid | 17:38 |
mriedem | seems you could map things that way, as the bdm.volume_id field in the db is that serial id | 17:38 |
mriedem | in the xml | 17:38 |
*** baoli has joined #openstack-nova | 17:39 | |
artom | mriedem, that seems to be much smarter than the current way | 17:39 |
mriedem | well assuming serial is always set, but i'm not sure when it wouldn't be | 17:40 |
mriedem | but then again, there could be wrinkles for that in the bowels of the block_device_info code so who knows | 17:40 |
mriedem | i see _volume_snapshot_delete in the libvirt driver relying on it though | 17:41 |
mriedem | artom: this is what sets that in the disk config xml https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/volume/volume.py#L57 | 17:42 |
mriedem | and this is what sets it in bdm.connection_info on attach https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/block_device.py#L260 | 17:43 |
mriedem | so seems like a safe fallback | 17:43 |
*** ijw has joined #openstack-nova | 17:43 | |
*** slaweq has quit IRC | 17:44 | |
*** raildo has quit IRC | 17:47 | |
artom | mriedem, thanks, I'll try that :) | 17:47 |
artom | You've saved me from my rabbit hole | 17:47 |
artom | mriedem, wait, where is the volume uuid in the block device DB object? In connection_info somewhere? | 17:50 |
artom | Grah, you've said it yourself, volume_id | 17:51 |
artom | Which is a string and not a uuid, which makes me weary, but it's the best I've got | 17:51 |
*** hfu has quit IRC | 17:52 | |
*** baoli has quit IRC | 17:55 | |
*** baoli has joined #openstack-nova | 17:57 | |
*** mdnadeem has joined #openstack-nova | 17:57 | |
mriedem | yeah bdm.volume_id is the volume uuid from cinder | 17:58 |
*** links has joined #openstack-nova | 17:59 | |
mriedem | artom: the tricky part might be when the bdm.volume_id is set in the database during attach | 18:00 |
mriedem | although for attach volume you should be ok, i was thinking of something like this https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/block_device.py#L381 | 18:00 |
mriedem | which is boot from volume, where nova creates the volume from blank/image/snapshot | 18:00 |
mriedem | the connection_info passed into driver.attach_volume also has that 'serial' entry though | 18:01 |
*** ijw has quit IRC | 18:06 | |
*** ociuhandu has quit IRC | 18:08 | |
jaypipes | mriedem: https://review.openstack.org/#/q/topic:bp/flavor-notifications+status:open looks pretty good to go to me. | 18:14 |
mriedem | yowser https://review.openstack.org/#/c/398171/ | 18:16 |
mriedem | ok i'll add to my list | 18:16 |
*** ociuhandu has joined #openstack-nova | 18:20 | |
*** adisky_ has quit IRC | 18:29 | |
*** ociuhandu has quit IRC | 18:31 | |
*** ociuhandu has joined #openstack-nova | 18:33 | |
*** hfu has joined #openstack-nova | 18:33 | |
*** david-lyle has joined #openstack-nova | 18:34 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: RT now adds proper Ironic inventory https://review.openstack.org/404472 | 18:35 |
artom | mriedem, well crap - I just tried it, and the serial wasn't set in the xml | 18:36 |
artom | So it looks like serial isn't 100% either | 18:36 |
artom | But maybe it's easier to fix than device_name | 18:37 |
*** ducttape_ has joined #openstack-nova | 18:37 | |
*** links has quit IRC | 18:38 | |
mriedem | artom: at what point is _build_device_metadata called during the attach_volume flow? | 18:41 |
*** ducttape_ has quit IRC | 18:42 | |
*** Zer0Byte__ has joined #openstack-nova | 18:44 | |
*** Zer0Byte__ has quit IRC | 18:44 | |
*** dtp has quit IRC | 18:44 | |
mriedem | artom: because at this point https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/driver.py#L1197 that conf object should have serial set due to https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/virt/libvirt/volume/volume.py#L57 | 18:44 |
*** mdnadeem has quit IRC | 18:49 | |
*** pcaruana has quit IRC | 18:52 | |
*** Zer0Byte__ has joined #openstack-nova | 18:54 | |
*** kristian__ has quit IRC | 19:04 | |
*** kristian__ has joined #openstack-nova | 19:04 | |
artom | mriedem, at the end of attach_volume in the compute api | 19:09 |
artom | And at this point, it's 20:00 here and I have a beer in me, so see you tomorrow! | 19:09 |
mriedem | artom: ttyl | 19:10 |
*** david-lyle has quit IRC | 19:12 | |
*** david-lyle has joined #openstack-nova | 19:17 | |
mriedem | jaypipes: -1 on that base flavor notifications patch https://review.openstack.org/#/c/398171/ - comments inline | 19:21 |
*** kristian__ has quit IRC | 19:22 | |
*** kristian__ has joined #openstack-nova | 19:22 | |
openstackgerrit | Akshil Verma proposed openstack/python-novaclient: Remove HTTP error code when creating flavor https://review.openstack.org/415289 | 19:23 |
jaypipes | mriedem: having a hell of a time understanding how to add api_sample_tests for a specific microversion. Would you mind checking out http://logs.openstack.org/31/415031/2/check/gate-nova-tox-db-functional-ubuntu-xenial/6655fc4/testr_results.html.gz and giving some advice if you have some spare time today? | 19:24 |
*** david-lyle has quit IRC | 19:26 | |
jaypipes | have I mentioned today how much I dislike the api_sample_tests? | 19:26 |
*** kristian__ has quit IRC | 19:27 | |
mriedem | jaypipes: for one thing, your api response sample for v2.41 doesn't have the uuid field in it https://review.openstack.org/#/c/415031/2/doc/api_samples/os-aggregates/v2.41/aggregate-post-resp.json | 19:28 |
jaypipes | ffs | 19:29 |
mriedem | i'm leaving some other comments | 19:30 |
*** rfolco has quit IRC | 19:31 | |
mriedem | jaypipes: we're going to want a bp and spec for that api change, which i think we can make an exception for and fast track it, | 19:33 |
mriedem | but we wouldn't normally allow a bug fix in to add a microversion to the rest api w/o a spec | 19:33 |
jaypipes | mriedem: that was the only tpl file that was missing the uuid field. | 19:34 |
openstackgerrit | Jay Pipes proposed openstack/nova: Return uuid attribute for aggregates https://review.openstack.org/415031 | 19:34 |
jaypipes | mriedem: so that's definitely not the issue, since all 5 tests are failing. | 19:34 |
mriedem | but aren't all 5 creating an aggregate first? | 19:34 |
mriedem | so it fails on that | 19:34 |
jaypipes | hmm. | 19:34 |
jaypipes | ok, well we'll see if that works :) | 19:35 |
mriedem | yeah they all have to create an aggregate first, then do things to it like add a host, or list aggregates from that created one | 19:36 |
jaypipes | mriedem: after a while, your mind just kind of numbs to the copy/pasted JSON files :( | 19:36 |
mriedem | yeah i had issues with that when i was doing get me a network, oomichi had to point out my issues | 19:36 |
jaypipes | this is partly why I greatly prefer the Gabbi way of testing specific microversion changes. it's a whole lot more explicit and obvious what you're verifying. | 19:36 |
*** ijw has joined #openstack-nova | 19:41 | |
mriedem | jaypipes: i imagine this wasn't a problem because we don't have a functional test for update? https://review.openstack.org/#/c/415031/2/doc/api_samples/os-aggregates/v2.41/aggregate-update-post-resp.json | 19:44 |
mriedem | weird you do have test_update_aggregate | 19:46 |
*** kristian__ has joined #openstack-nova | 19:51 | |
mriedem | jianghuaw: the xenserver ci fails hard on this https://review.openstack.org/#/c/406059/ - the virt driver can't even load the xenapi session object | 19:55 |
*** kristian__ has quit IRC | 19:56 | |
*** dtp has joined #openstack-nova | 20:00 | |
*** ducttape_ has joined #openstack-nova | 20:07 | |
*** yamahata has joined #openstack-nova | 20:08 | |
*** ducttape_ has quit IRC | 20:12 | |
*** moshele has joined #openstack-nova | 20:13 | |
*** ijw has quit IRC | 20:14 | |
mriedem | cfriesen: fixed http://docs.openstack.org/releasenotes/nova/newton.html#id3 | 20:17 |
*** kristian__ has joined #openstack-nova | 20:29 | |
*** yamahata has quit IRC | 20:31 | |
*** dave-mccowan has joined #openstack-nova | 20:37 | |
*** kristia__ has joined #openstack-nova | 20:43 | |
*** kristian__ has quit IRC | 20:43 | |
mriedem | jroll: how are microversions handled in ironicclient? at least when nova is using it. i.e. https://review.openstack.org/#/c/247904/15/ironicclient/v1/node.py isn't passing 1.27 but that's the microversion in the ironic rest api for that change: | 20:51 |
mriedem | https://review.openstack.org/#/c/216730/33/doc/source/dev/webapi-version-history.rst | 20:51 |
*** slaweq has joined #openstack-nova | 21:00 | |
*** hogepodge has quit IRC | 21:02 | |
*** mrhillsman has quit IRC | 21:02 | |
*** hogepodge has joined #openstack-nova | 21:03 | |
*** codebauss has joined #openstack-nova | 21:04 | |
*** codebauss is now known as mrhillsman | 21:04 | |
*** hogepodge has quit IRC | 21:08 | |
*** ijw has joined #openstack-nova | 21:10 | |
*** code4coffee has joined #openstack-nova | 21:12 | |
openstackgerrit | Sujitha proposed openstack/nova: Change tags to default field in Instance object. https://review.openstack.org/415298 | 21:14 |
*** yamahata has joined #openstack-nova | 21:14 | |
code4coffee | hey guys, I am trying to get console working with virt_type=lxc, I am starting to doubt if it's at all possible based on my limited python experience. Could someone elaborate on the matter? | 21:15 |
mriedem | code4coffee: there are no console APIs supported for LXC, see http://docs.openstack.org/developer/nova/support-matrix.html | 21:24 |
*** sacharya has quit IRC | 21:26 | |
code4coffee | mriedem: thanks for the link, serial seems to be unknown which would be most interesting for me. I've been using proxmox previously which supports novnc html5 client with lxc containers, I'm wondering if there is any technical reason why it can not be implemented in nova | 21:33 |
*** ducttape_ has joined #openstack-nova | 21:38 | |
*** ijw has quit IRC | 21:39 | |
*** david-lyle has joined #openstack-nova | 21:42 | |
mriedem | code4coffee: i'm not sure, and that support matrix may very well be out of date | 21:43 |
mriedem | we had an lxc CI job in the nova experimental queue, it might be worth finding out if that job ever ran any of the serial console tests | 21:43 |
*** ducttape_ has quit IRC | 21:44 | |
*** david-lyle has quit IRC | 21:48 | |
*** ijw has joined #openstack-nova | 21:48 | |
*** moshele has quit IRC | 21:54 | |
*** dave-mcc_ has joined #openstack-nova | 21:54 | |
*** Zer0Byte__ is now known as jose-phillips | 21:55 | |
*** dave-mccowan has quit IRC | 21:56 | |
*** sacharya has joined #openstack-nova | 21:57 | |
mriedem | jaypipes: if a tag in the tags table is Unicode(80) https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/models.py#L1464 | 22:00 |
mriedem | jaypipes: does that means it's really == String(4*80)? | 22:00 |
*** ducttape_ has joined #openstack-nova | 22:07 | |
code4coffee | mriedem: as far as I understand it's not supported, because in libvirt driver _create_consoles method, any non kvm or qemu virt_type will result in pty console, and later console retrieval code only supports console with type of tcp. Which means it shouldn't work. | 22:11 |
*** avolkov has quit IRC | 22:18 | |
*** avolkov has joined #openstack-nova | 22:18 | |
*** jose-phi_ has joined #openstack-nova | 22:19 | |
*** jose-phillips has quit IRC | 22:19 | |
mriedem | Kevin_Zheng: https://review.openstack.org/#/c/394321/ needs to be rebased and I think it can be substantially smaller | 22:24 |
mriedem | jaypipes: was asking because of this https://review.openstack.org/#/c/394321/17/nova/db/sqlalchemy/api_migrations/migrate_repo/versions/031_build_requests_add_tags.py@27 | 22:25 |
*** slaweq has quit IRC | 22:25 | |
*** jose-phillips has joined #openstack-nova | 22:28 | |
*** jose-phi_ has quit IRC | 22:29 | |
*** jose-phillips has quit IRC | 22:30 | |
*** jose-phillips has joined #openstack-nova | 22:31 | |
*** dimtruck is now known as zz_dimtruck | 22:35 | |
*** kristia__ has quit IRC | 22:36 | |
*** code4coffee has quit IRC | 22:39 | |
*** sacharya has quit IRC | 22:41 | |
*** slaweq has joined #openstack-nova | 22:42 | |
*** sacharya has joined #openstack-nova | 22:44 | |
*** david-lyle has joined #openstack-nova | 22:44 | |
*** sacharya has quit IRC | 22:48 | |
*** sacharya has joined #openstack-nova | 22:49 | |
*** david-lyle has quit IRC | 22:49 | |
openstackgerrit | Sujitha proposed openstack/nova: Change tags to default field in Instance object. https://review.openstack.org/415298 | 22:53 |
*** sacharya has quit IRC | 22:53 | |
*** vtech has joined #openstack-nova | 22:57 | |
mriedem | jaypipes: so your aggregates uuid functional api samples tests are failing because the microversion in the request is 2.1 rather than 2.41 for some reason | 22:58 |
*** dave-mcc_ has quit IRC | 22:58 | |
openstackgerrit | Sujitha proposed openstack/nova: Add more field's in InstancePayload https://review.openstack.org/407128 | 23:00 |
openstackgerrit | Sujitha proposed openstack/nova: Adding tags field to InstancePayload https://review.openstack.org/407228 | 23:00 |
Kevin_Zheng | mriedem: thanks for the review, will do it when I got to the office | 23:04 |
openstackgerrit | Sujitha proposed openstack/nova: Change tags to default field in Instance object. https://review.openstack.org/415298 | 23:04 |
mriedem | jaypipes: ok two things, first you need to set the microversion on the self.api object in the functional test | 23:05 |
mriedem | https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/tests/functional/api_sample_tests/test_hypervisors.py#L142 | 23:05 |
mriedem | like that ^ | 23:05 |
mriedem | that's a bug imo that the base class doesn't handle that setup for you | 23:05 |
mriedem | jaypipes: this is your other issue: https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/objects/aggregate.py#L260 | 23:05 |
mriedem | when constructing the Aggregate object, we don't load the uuid field | 23:06 |
mriedem | well, later on we set a uuid if there isn't one in the db https://github.com/openstack/nova/blob/d442488ebe19a6b3ddea6ec8668f92c7e2a0a644/nova/objects/aggregate.py#L287 | 23:07 |
*** ducttape_ has quit IRC | 23:12 | |
*** efried1 has joined #openstack-nova | 23:14 | |
*** efried has quit IRC | 23:15 | |
*** vtech has quit IRC | 23:21 | |
*** amotoki has quit IRC | 23:28 | |
*** hongbin has joined #openstack-nova | 23:29 | |
*** amotoki has joined #openstack-nova | 23:35 | |
*** takashin has joined #openstack-nova | 23:35 | |
*** lamt has quit IRC | 23:36 | |
*** kristian__ has joined #openstack-nova | 23:36 | |
*** sacharya has joined #openstack-nova | 23:36 | |
*** zul has quit IRC | 23:40 | |
*** kristian__ has quit IRC | 23:41 | |
*** dtp has quit IRC | 23:43 | |
*** zul has joined #openstack-nova | 23:43 | |
mriedem | jaypipes: ok so i see what's up - each test creates a new aggregate, which automatically sets the uuid to a new random value, so the response has a random uuid each time. the api samples have a static uuid, so the compare fails. other api samples tests that use a uuid stub out the DB API calls to return a fake with hard-coded uuids that match what's in the api samples docs. | 23:44 |
*** hogepodge has joined #openstack-nova | 23:44 | |
*** david-lyle has joined #openstack-nova | 23:45 | |
*** ijw has quit IRC | 23:45 | |
*** zhs_ has quit IRC | 23:45 | |
*** zhs_ has joined #openstack-nova | 23:46 | |
*** zz_dimtruck is now known as dimtruck | 23:46 | |
*** slaweq has quit IRC | 23:47 | |
*** ducttape_ has joined #openstack-nova | 23:49 | |
*** david-lyle has quit IRC | 23:49 | |
*** sacharya has quit IRC | 23:50 | |
*** sacharya has joined #openstack-nova | 23:50 | |
*** avolkov has quit IRC | 23:51 | |
*** amotoki has quit IRC | 23:54 | |
*** sacharya has quit IRC | 23:55 | |
*** hogepodge has quit IRC | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!