| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-01-26 | |||
| 20:13:05 | bauzas | dansmith: oh good point, I missed the grenade Depends-On | |
| 20:13:34 | openstackgerrit | Chris Dent proposed openstack/nova: DNM: Don't clear the tracked instances in resource tracker https://review.openstack.org/425885 | |
| 20:15:11 | bauzas | dansmith: FWIW, I think matt asked for updating as well the devref | |
| 20:15:34 | bauzas | dansmith: I just provided a reno note, but maybe we should do that in a follow-up ? | |
| 20:15:43 | dansmith | yeah | |
| 20:15:51 | bauzas | not needing to be merged for o-3 but still be good for RC1 | |
| 20:16:07 | cdent | dansmith: that review ^^ is checking one of investigations into the redundancies in the resourece_tracker | |
| 20:16:40 | bauzas | dansmith: FWIW, release mgmt people accept for us a few FFE | |
| 20:16:56 | bauzas | the only problem is that we have RC1 next week | |
| 20:17:27 | bauzas | so in case the gate is bad, we could possibly cut o-3 and then merge the needed changes right after | |
| 20:19:51 | bauzas | oooooh | |
| 20:20:06 | bauzas | the requirements are frozen now :( | |
| 20:20:12 | bauzas | andreykurilin: ^ | |
| 20:22:00 | bauzas | WTF... http://logs.openstack.org/61/417961/31/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/cd57fce/console.html#_2017-01-26_20_10_22_249335 | |
| 20:22:03 | bauzas | dansmith: ^ | |
| 20:22:51 | dansmith | bauzas: that would be an expected error if you're not getting resources from the other node right? | |
| 20:23:38 | bauzas | hhhhmpppfff http://logs.openstack.org/61/417961/31/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/cd57fce/logs/new/screen-n-sch.txt.gz?level=INFO#_2017-01-26_20_06_58_563 | |
| 20:23:43 | bauzas | lemme see | |
| 20:24:00 | bauzas | how it can be possible now we accept older computes | |
| 20:28:35 | mriedem | how's it going? | |
| 20:30:08 | dansmith | mriedem: fungi bumped the devstack fix in the gate, when my change was #1 and almost done, then mine failed on a known bug at #2, currently re-running | |
| 20:30:20 | dansmith | mriedem: while I wait, I'm making a fungi-shaped voodoo doll | |
| 20:30:33 | dansmith | mriedem: sounds like something is still up with bauzas' junk | |
| 20:30:52 | dansmith | mriedem: and I'm eating a chicken pot pie as comfort food | |
| 20:30:53 | dansmith | end of status | |
| 20:31:00 | dansmith | oh great, that devstack fix is failing and going to reset the gate | |
| 20:31:11 | mriedem | i thought bauzas saw the doctor about his junk issue | |
| 20:31:22 | dansmith | the cream isn't working apparently | |
| 20:31:25 | fungi | dansmith: https://www.seedsnow.com/blogs/news/7186766-how-to-make-a-mushroom-out-of-radish | |
| 20:31:30 | mriedem | le creme | |
| 20:31:38 | dansmith | fungi: hah | |
| 20:32:05 | bauzas | mriedem: unrelated, requirements is now frozen so https://review.openstack.org/#/c/425827/ is in the weeds for capping u-c | |
| 20:32:35 | bauzas | mriedem: *la* crème | |
| 20:33:04 | fungi | dansmith: i think clarkb actually promoted it, but i'll see if i can dig it back up | |
| 20:33:17 | dansmith | fungi: oh, well it's not too late for me to change this into a clarkb doll | |
| 20:33:33 | dansmith | fungi: it needed doing, I'm just annoyed | |
| 20:35:05 | mriedem | bauzas: but we can still release novaclient 7.1.0 can't we? | |
| 20:35:11 | bauzas | of course | |
| 20:35:21 | bauzas | just saying that for u-c | |
| 20:35:54 | bauzas | mriedem: there is a patch in the gate for adding a reno note | |
| 20:35:59 | mriedem | the wifi in our room sucks so maybe sometime this afternoon i'll be able to load some of these links | |
| 20:36:06 | bauzas | mriedem: that's why I'm putting WIP | |
| 20:36:19 | mriedem | reno for what? | |
| 20:36:40 | bauzas | mriedem: for the token fix | |
| 20:36:44 | bauzas | in the client | |
| 20:36:55 | bauzas | we agreed to not wait for the CLI | |
| 20:37:40 | bauzas | but given the CLI included the reno note mentioning the regression bugfix of 7.0.0, I preferred asking andreykurilin to provide a separate change for the reno file and I fast-approved it | |
| 20:38:15 | bauzas | mriedem: https://review.openstack.org/#/c/425822/1 is the patch in question | |
| 20:38:38 | openstackgerrit | Vladik Romanovsky proposed openstack/nova: Adding vlans field to Device tagging metadata https://review.openstack.org/405951 | |
| 20:49:09 | mriedem | bauzas: so the fix mentioned in https://review.openstack.org/#/c/425822/ is already fixed and merged? | |
| 20:49:32 | bauzas | mriedem: yup, the python binding | |
| 20:49:49 | bauzas | mriedem: not the CLI | |
| 20:50:51 | mriedem | https://github.com/openstack/python-novaclient/commit/9940e3fe0e47ff5f2f6d05c9564d17fc19ca0f5c | |
| 20:50:52 | mriedem | ok | |
| 20:51:07 | bauzas | mriedem: yup https://review.openstack.org/#/c/419441/ | |
| 20:53:31 | fungi | dansmith: mriedem: i've reenqueued and promoted 425620,7 now | |
| 20:53:46 | dansmith | fungi: and my change too, right? :D | |
| 20:53:49 | dansmith | no, I know. | |
| 20:54:29 | dansmith | I know | |
| 20:55:06 | dansmith | bauzas: looks like your bottom two might be merge-able soon | |
| 20:56:12 | bauzas | dansmith: I'm still puzzled by why grenade-multi is unhappy after our changes | |
| 20:57:47 | bauzas | http://logs.openstack.org/61/417961/31/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/cd57fce/logs/new/screen-n-sch.txt.gz?#_2017-01-26_20_06_58_560 tells me that node-[...]154 is not present | |
| 21:00:23 | cdent | bauzas, dansmith: I need to checkout, anything I should chase tomorrow morning? | |
| 21:00:54 | dansmith | cdent: as of now, I think we're just hoping to land the time-sensitive stuff before tomorrow | |
| 21:01:04 | dansmith | cdent: but someone can send a mail to the list if there's anything | |
| 21:01:39 | cdent | dansmith: I can move stuff around stick around now if there's something specific that I could help with? | |
| 21:01:40 | mriedem | i'm looking through vladikr's vlan metadata stuff again | |
| 21:01:54 | mriedem | after that it's jaypipes' resource class ironic stuff | |
| 21:02:03 | mriedem | cdent: i think we're mostly just watching the gate | |
| 21:02:04 | mriedem | and rechecking | |
| 21:02:05 | dansmith | cdent: unless you staring at the gate will make it go faster than me staring at the gate. otherwise, I'd go | |
| 21:02:07 | dansmith | heh | |
| 21:02:55 | cdent | my experience is that when I look at the gate it fails, so I'm going to go. but yeah, if there's utc morning stuff to do, leave another of those public handoff emails | |
| 21:03:09 | cdent | good luck, break a leg or whatever it takes | |
| 21:03:44 | bauzas | meh | |
| 21:04:06 | bauzas | the old compute looks good | |
| 21:04:31 | _gryf | do we need a saw | |
| 21:04:33 | _gryf | ? | |
| 21:04:33 | bauzas | but I still don't understand why we're not updating its hostState on the sheduler side | |
| 21:04:55 | _gryf | oh man. wrong channel | |
| 21:09:08 | bauzas | dansmith: the only possibilty I see with the grenade-multi issue is that we don't have the right service version | |
| 21:10:19 | bauzas | checking whether we are calling placement | |
| 21:13:29 | mriedem | +W on that bottom server groups test patch | |
| 21:14:36 | bauzas | erf, we don't follow-up passing the req-id to the placement API | |
| 21:14:45 | bauzas | we regenerate a new one | |
| 21:14:58 | bauzas | so good luck for guessing | |
| 21:15:11 | mriedem | bauzas: we want a reno on this right? https://review.openstack.org/#/c/425806/ | |
| 21:15:22 | mriedem | heads up - your computes won't start if you upgrade to ocata but don't have placement configured | |
| 21:15:33 | mriedem | bauzas: yeah cdent is working on that | |
| 21:15:35 | bauzas | mriedem: I did that in the top patch | |
| 21:15:46 | mriedem | bauzas: ok | |
| 21:15:47 | bauzas | mriedem: but I should have splitted the note | |
| 21:16:00 | mriedem | for shame | |
| 21:16:21 | bauzas | mriedem: that said, if we are unable to merge the top one in time, we can just add the note in a follow-up before RC1 | |
| 21:16:33 | bauzas | unless you strongly disagree | |
| 21:16:55 | mriedem | i think we're ok | |
| 21:16:58 | mriedem | we'll have a reno either way before rc1 | |
| 21:17:09 | bauzas | ok | |
| 21:18:01 | claudiub | mriedem: hi. just an fyi, both moshele and sfinucan gave their blessings on the Hyper-V PCI passthrough patch. :D https://review.openstack.org/#/c/420614/ | |
| 21:18:16 | mriedem | yeah i saw | |