| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-07-31 | |||
| 21:08:20 | mriedem | dansmith: i think you just need to do: self.compute.manager.update_available_resource(ctxt) | |
| 21:08:30 | dansmith | mriedem: I know | |
| 21:08:43 | dansmith | mriedem: getting a failure on one of the conditions after doing that though | |
| 21:12:45 | openstackgerrit | Michael Still proposed openstack/nova master: Move execs of touch to privsep. https://review.openstack.org/489190 | |
| 21:14:02 | dansmith | yeah, no worky | |
| 21:14:05 | dansmith | it's running the code | |
| 21:15:17 | dansmith | and I'm running it for both of the fake hosts | |
| 21:17:34 | dansmith | heh | |
| 21:17:39 | dansmith | it's deleting the instance allocation | |
| 21:17:52 | dansmith | I wonder if this is just a different orientation of some races | |
| 21:20:40 | dansmith | yeah I think if I reorder the RT runs it fails in a different way | |
| 21:22:41 | cburgess | Anyone know if we have an API call to retrieve the contents of and entry in the block_device_mapping table? I can't find one but wanted to confirm before I went and did bad horrible ugly things. | |
| 21:23:24 | openstackgerrit | Jackie Truong proposed openstack/nova master: Add trusted_certs to Instance object https://review.openstack.org/489408 | |
| 21:23:26 | openstackgerrit | Merged openstack/nova master: Updated from global requirements https://review.openstack.org/488865 | |
| 21:25:29 | tonyb | Any chance someone that knows SSL can look at: https://review.openstack.org/#/c/489078/ Somethign like that is needed to unblock the glanceclient update in requirements | |
| 21:31:20 | openstackgerrit | Dan Smith proposed openstack/nova master: Test resize with placement api https://review.openstack.org/487958 | |
| 21:31:32 | dansmith | mriedem: jaypipes: pretty brutal brokenness when we explicitly run the periodics ^ :( | |
| 21:32:02 | dansmith | almost all of the checks are commented out and replaced with "assert there is no allocation at all" | |
| 21:33:14 | mikal | sdague: I am awake now | |
| 21:33:33 | mikal | sdague: I replied to your reno comment. I think we do them later as we remove things, not in that cleanup patch. | |
| 21:33:42 | mriedem | comments in ps9 | |
| 21:33:52 | mikal | sdague: I'd also like to get that series merged really early in Queens in case it destabilizes things in confusing ways. | |
| 21:35:32 | dansmith | mriedem: I had just removed those prints.. do you want them as log.info or is removed okay? | |
| 21:35:54 | mriedem | removed is fine, the should be in the mismatch error anyway | |
| 21:35:57 | mriedem | *they | |
| 21:39:43 | dansmith | mriedem: you wanna go through my latest rev quickly to see if you're okay with those changes and then I'll fix your redundancies? | |
| 21:40:19 | mriedem | i can | |
| 21:40:40 | mriedem | i thought you said the direct update_available_resource made everything go sideways though? | |
| 21:41:05 | dansmith | that was before I realized how fscked it was with them set | |
| 21:41:11 | mriedem | oh | |
| 21:41:28 | dansmith | letting them run in the sleep was causing them to run twice (well 1.5 times) and so the things that worked were hard-coded for that pattern | |
| 21:42:14 | dansmith | so it would update, delete, create in the span of the two seconds | |
| 21:42:20 | dansmith | running them each once leaves it at update, delete | |
| 21:42:27 | dansmith | so no allocation at all, target or source | |
| 21:42:36 | mriedem | ok new things looks fine | |
| 21:42:41 | dansmith | ack | |
| 21:44:39 | dansmith | boom ^ | |
| 21:44:41 | dansmith | dang | |
| 21:44:41 | openstackgerrit | Dan Smith proposed openstack/nova master: Test resize with placement api https://review.openstack.org/487958 | |
| 21:44:44 | dansmith | boom ^ | |
| 21:45:36 | mriedem | oh you sucker | |
| 21:45:59 | mriedem | DOUBLE BOOM! | |
| 21:46:03 | mriedem | SONIC BOOM?! | |
| 21:48:17 | tonyb | mriedem: https://www.youtube.com/watch?v=Fnmh7dF4c2U | |
| 21:48:20 | dansmith | ah, I didn't realize what you were pointing to in that | |
| 21:48:57 | mriedem | tonyb: exactly | |
| 21:49:21 | openstackgerrit | Dan Smith proposed openstack/nova master: Test resize with placement api https://review.openstack.org/487958 | |
| 21:49:41 | dansmith | turBOOM | |
| 21:51:32 | mriedem | dansmith: the commit message is a bit old now | |
| 21:51:41 | dansmith | gdi riedeman | |
| 21:52:11 | mriedem | well you could just leave it | |
| 21:52:24 | mriedem | 2 n's btw | |
| 21:52:29 | mriedem | double-n as my dad would say | |
| 21:52:42 | mriedem | karl with a k, double-n | |
| 21:53:14 | dansmith | gdi riedermann | |
| 21:53:21 | mriedem | double r is more like it | |
| 21:53:25 | dansmith | heh | |
| 21:53:32 | mriedem | riderman as my first manager at ibm would say | |
| 21:53:35 | mriedem | and write | |
| 21:54:00 | openstackgerrit | Dan Smith proposed openstack/nova master: Test resize with placement api https://review.openstack.org/487958 | |
| 21:54:05 | dansmith | hah, riderman | |
| 21:55:31 | mriedem | so i pulled it down and ran just the test class and both tests failed | |
| 21:55:46 | dansmith | they pass for me... | |
| 21:56:03 | dansmith | fail how? | |
| 21:56:43 | mriedem | i could be picking up a stale branch with the git review -d | |
| 21:56:56 | mriedem | no that's not it | |
| 21:57:37 | dansmith | okay I see fails if I run them in tox in parallel | |
| 21:58:03 | dansmith | they pass in isolation | |
| 21:59:03 | mriedem | tox -e functional -- nova.tests.functional.test_servers.ServerMovingTests | |
| 21:59:03 | mriedem | i'm just doing: | |
| 21:59:08 | mriedem | this time they both passed | |
| 21:59:27 | dansmith | ugh | |
| 22:00:22 | dansmith | I just ran singles a bunch of times and saw it fail | |
| 22:01:20 | dansmith | and it fails to assert the usage of the target, which should be stable since I run the periodics in the same order each time | |
| 22:01:45 | dansmith | brb | |
| 22:01:57 | mriedem | same here | |
| 22:07:22 | mriedem | wonder if it has something to do with the _FAKE_NODES in the fake virt driver | |
| 22:10:27 | dansmith | I think we're racing with other threads | |
| 22:10:40 | dansmith | I see when it fails that we delete the allocation, but PUT it right after, then the test fails | |
| 22:10:53 | dansmith | because it expects it to be gone, since the periodic on the second node should delete it | |
| 22:16:09 | dansmith | seems to fail more often than not when I run it from tox | |
| 22:16:38 | dansmith | but passes every time when I run it with subunit.run | |
| 22:16:46 | jaypipes | dansmith: have you pushed up another gibi patch? do I need to rebase? | |
| 22:16:58 | dansmith | jaypipes: yes, but it's not stable and I'm not sure why | |
| 22:20:00 | jaypipes | dansmith: that whole time.sleep(1) and manipulating the fake.set_nodes() globals is probably the culprit... | |
| 22:20:43 | melwitt | dansmith: I think subunit.run doesn't run tests in parallel but tox does (via testr underneath) | |
| 22:20:55 | jaypipes | dansmith: you could try adding a time.sleep(1) after the second self.start_service() call... | |
| 22:24:08 | mriedem | i'm not sure why the first time.sleep(1) is needed after the first service starts | |
| 22:24:52 | dansmith | melwitt: I'm running one test with tox, so should be the same | |
| 22:25:03 | melwitt | oh, one test | |
| 22:26:30 | dansmith | jaypipes: doesn't help | |
| 22:29:07 | dansmith | oh, | |
| 22:29:19 | dansmith | I was thinking he was forcing to host1 on initial boot each time | |
| 22:29:21 | dansmith | but he's not | |
| 22:29:37 | dansmith | so maybe it's just based on which it initially lands on and then moves to | |
| 22:30:02 | dfisher | nova-compute doesn't use etcd3, does it? (pike b3) | |
| 22:30:06 | dansmith | because I'm running periodics in a set order, but the actual stuff will be reversed | |
| 22:39:28 | dansmith | yeah I think that makes it repeatable | |
| 22:43:02 | mriedem | dansmith: yeah it's random | |
| 22:43:16 | mriedem | so toggle the update_rt call based on which host i guess? | |