| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-03-21 | |||
| 22:19:57 | mriedem | they are having timeouts in their jobs | |
| 22:20:13 | EmilienM | yeah it seems like everything broke for us today | |
| 22:20:23 | EmilienM | centos repos are down 30% of time | |
| 22:20:25 | EmilienM | etc etc | |
| 22:20:27 | mriedem | bauzas: we also have tripleo in the nova experimental queue if you just wanted to run it on the nova change itself | |
| 22:20:34 | EmilienM | no | |
| 22:20:42 | dansmith | we can go ahead and send that min_unit one though right? seems like a reasonable thing we can do for the moment, and it will self-heal if we change it later | |
| 22:20:42 | mriedem | not the same jobs? | |
| 22:20:43 | EmilienM | the experimental is not ovb IIRC | |
| 22:20:44 | bauzas | mriedem: I know that | |
| 22:20:56 | EmilienM | it's multinode, hence it doesn't test baremetal deployment | |
| 22:20:56 | mriedem | i'm fine with sending it in | |
| 22:20:58 | dansmith | actually, by the time we would want to change it we'll be scheduling on the custom class anyway | |
| 22:21:01 | bauzas | dansmith: yeah, I'm about to +W it anyway | |
| 22:21:13 | EmilienM | thanks for that ^ | |
| 22:21:19 | dansmith | I got it | |
| 22:21:29 | bauzas | heh | |
| 22:21:29 | EmilienM | I'll give you feedback as soon I I have one | |
| 22:21:32 | EmilienM | (as usual :)) | |
| 22:21:37 | bauzas | okay | |
| 22:21:46 | bauzas | then me disappears for the day | |
| 22:22:43 | EmilienM | bauzas: bonne nuit! | |
| 22:24:07 | mriedem | we should probably add a functional test in tree for the exact filters at some point | |
| 22:24:27 | mriedem | since the dsvm ironic job doesn't use those | |
| 22:24:51 | bauzas | mriedem: wait, what ? | |
| 22:25:13 | mriedem | bauzas: tripleo is using the exact filters right? | |
| 22:25:17 | bauzas | mriedem: nope | |
| 22:25:23 | mriedem | oh it's the flavor size? | |
| 22:25:27 | mriedem | i guess i'm mixing those up | |
| 22:25:37 | bauzas | mriedem: sec, verifying an ironic job log | |
| 22:26:07 | mriedem | http://logs.openstack.org/98/448098/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-bios-agent_ipmitool-tinyipa-ubuntu-xenial-nv/c0cc973/logs/etc/nova/nova.conf.txt.gz | |
| 22:26:28 | mriedem | scheduler_default_filters = RetryFilter,AvailabilityZoneFilter,RamFilter,DiskFilter,ComputeFilter,ComputeCapabilitiesFilter,ImagePropertiesFilter,ServerGroupAntiAffinityFilter,ServerGroupAffinityFilter,SameHostFilter,DifferentHostFilter | |
| 22:27:09 | bauzas | http://logs.openstack.org/98/448098/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-bios-agent_ipmitool-tinyipa-ubuntu-xenial-nv/c0cc973/logs/screen-n-sch.txt.gz#_2017-03-21_14_40_25_936 | |
| 22:27:18 | bauzas | mriedem: ^ | |
| 22:27:25 | bauzas | mriedem: you're getting confused by the opt | |
| 22:27:40 | bauzas | mriedem: if ironic, then it's not the right opt | |
| 22:27:41 | jroll | bauzas: but, filter_scheduler.use_baremetal_filters = False | |
| 22:27:43 | jroll | http://logs.openstack.org/98/448098/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-bios-agent_ipmitool-tinyipa-ubuntu-xenial-nv/c0cc973/logs/screen-n-sch.txt.gz#_2017-03-21_14_40_25_941 | |
| 22:27:52 | mriedem | http://logs.openstack.org/22/447522/3/check/gate-tripleo-ci-centos-7-nonha-multinode-updates-nv/d7bf42b/logs/etc/nova/nova.conf.txt.gz | |
| 22:27:54 | jroll | mriedem is correct | |
| 22:28:01 | jroll | and this deserves a hearty "lolwtf" | |
| 22:28:08 | mriedem | enabled_filters=RetryFilter,TripleOCapabilitiesFilter,ComputeCapabilitiesFilter,AvailabilityZoneFilter,RamFilter,DiskFilter,ComputeFilter,ImagePropertiesFilter,ServerGroupAntiAffinityFilter,ServerGroupAffinityFilter | |
| 22:28:26 | bauzas | ooooh fuss http://logs.openstack.org/98/448098/2/check/gate-tempest-dsvm-ironic-ipa-wholedisk-bios-agent_ipmitool-tinyipa-ubuntu-xenial-nv/c0cc973/logs/screen-n-sch.txt.gz#_2017-03-21_14_40_25_941 | |
| 22:28:51 | jroll | not that it would test anything, other than they run | |
| 22:28:57 | jroll | because all of our gate hosts are same size | |
| 22:29:00 | jroll | s/hosts/nodes/ | |
| 22:29:14 | mriedem | use_baremetal_filters=False is in the nova.conf for the tripleo job too | |
| 22:29:19 | bauzas | I have something in my pipe about working on functional tests for filtres | |
| 22:29:20 | jroll | ya | |
| 22:29:36 | mriedem | anyway, a regression functional test is needed here obviously | |
| 22:29:38 | bauzas | we could check thart | |
| 22:29:42 | bauzas | yeah | |
| 22:31:17 | bauzas | jroll: any reason why the ironic job isn't using the right opt values ? | |
| 22:31:33 | jroll | bauzas: idk, but it wouldn't make a difference | |
| 22:31:44 | jroll | bauzas: our "nodes" in the gate are all the same size | |
| 22:31:54 | jroll | so it doesn't test much, but I don't mind adding it | |
| 22:31:59 | bauzas | well | |
| 22:32:14 | bauzas | the main problem is that we tell our ops to opt-in that :) | |
| 22:32:34 | jroll | sure | |
| 22:32:38 | bauzas | if we don't use the same flags that the ones we ask for our operators, then... :p | |
| 22:32:41 | jroll | don't disagree we should add it | |
| 22:32:50 | jroll | but even so, I could make it a noop and it would still pass our jobs | |
| 22:33:05 | bauzas | doesn't devstack turn that flag on automatically ? | |
| 22:33:14 | bauzas | I mean for lib/ironic | |
| 22:33:24 | jroll | the baremetal filters? apparently not | |
| 22:34:50 | bauzas | I thought it was the case | |
| 22:35:00 | jroll | I thought so too | |
| 22:35:06 | jroll | but CI just proved us wrong :) | |
| 22:35:11 | bauzas | I only played a very few with devstack installing ironic | |
| 22:35:31 | bauzas | so I basically know the workflow but I thought it was done this way | |
| 22:37:11 | bauzas | mmmm, not the case in lib/ironic AFAICS | |
| 22:38:04 | jroll | bauzas: https://review.openstack.org/#/c/448303/ | |
| 22:40:37 | bauzas | jroll: tbh, can't find in the dev howto where to set that conf flag... | |
| 22:41:02 | bauzas | speaking of https://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack | |
| 22:41:17 | jroll | bauzas: it's in the install guide https://docs.openstack.org/project-install-guide/baremetal/draft/configure-integration.html#configure-compute-to-use-the-bare-metal-service | |
| 22:42:01 | bauzas | okay | |
| 22:44:12 | bauzas | anyway, time for calling it a day | |
| 22:46:22 | jroll | indeed, night bauzas | |
| 23:44:49 | openstackgerrit | Dan Smith proposed openstack/nova master: Remove the Allocation.create() method. https://review.openstack.org/416751 | |
| 23:44:49 | openstackgerrit | Dan Smith proposed openstack/nova master: Add check for invalid inventory amounts https://review.openstack.org/428800 | |
| 23:44:50 | openstackgerrit | Dan Smith proposed openstack/nova master: Add check for invalid allocation amounts https://review.openstack.org/416752 | |
| #openstack-nova - 2017-03-22 | |||
| 00:18:45 | mriedem | dansmith: you want to get these too? https://review.openstack.org/#/q/I1f5b772bff8fa2dc508c4bdd3dd1f32838594786,n,z | |
| 00:19:23 | dansmith | yar | |
| 00:21:21 | dansmith | and with that, | |
| 00:21:28 | dansmith | eye meow tah | |
| 00:21:36 | mriedem | cute | |
| 01:40:36 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Refactor a test method including 7 test cases https://review.openstack.org/395436 | |
| 01:40:52 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Refactor a test method including 3 test cases https://review.openstack.org/394738 | |
| 01:54:34 | openstackgerrit | Bing Li proposed openstack/python-novaclient master: Remove log translations https://review.openstack.org/448336 | |
| 01:55:09 | mriedem | zul: https://review.openstack.org/#/c/448253/ isn't approved on master or stable/ocata | |
| 01:56:15 | mriedem | oh i guess it's on master | |
| 01:56:18 | mriedem | need stable/ocata though | |
| 01:56:27 | zul | mriedem: whoops sorry | |
| 01:56:37 | mriedem | that's fine, the change is merged on master | |
| 01:56:41 | mriedem | but also proposed for ocata | |
| 01:57:12 | zul | mriedem: a little trigger happy ;) | |
| 01:57:30 | mriedem | so excited to see that fixed i guess | |
| 01:57:39 | mriedem | or just tired of hearing about it internally | |
| 02:00:45 | zul | mriedem: a little from column A a little from column B | |
| 02:01:57 | openstackgerrit | Shunli Zhou proposed openstack/nova master: Remove mox from cells/test_cells_rpc_driver.py https://review.openstack.org/447412 | |