| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2023-11-13 | |||
| 17:42:05 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: WIP: Switch to use of 'Command.log' https://review.opendev.org/c/openstack/python-openstackclient/+/860478 | |
| 17:42:05 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: tests: Remove prints https://review.opendev.org/c/openstack/python-openstackclient/+/900813 | |
| 17:55:55 | M0weng[m] | > 0weng: Those failures are happening because neutron isn't configured with all of its extensions. That's not your fault: the tests... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/SyOAqVOyPSOsAlnjKAqDiPMt>) | |
| 17:56:06 | M0weng[m] | * > 0weng: Those failures are happening because neutron isn't configured with all of its extensions. That's not your fault: the tests... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/ZuRPTuSrIGdaJBQAYpDYhRpc>) | |
| 17:59:22 | M0weng[m] | I'm still getting two other kinds of failures; the first one is Authentication cannot be scoped to multiple targets. Pick one of: project, domain, trust, system or unscoped, and the other one is ResourceNotFound: 404: Client Error for url: http://10.248.246.143:9696/networking/v2.0/metering/metering-labels, The resource could not be found.; both of these seem to be happening for Tempest. (I'll paste another full log once all the | |
| 17:59:22 | M0weng[m] | tests are done running.) | |
| 17:59:42 | M0weng[m] | * I'm still getting two other kinds of failures; the first one is Authentication cannot be scoped to multiple targets. Pick one of: project, domain, trust, system or unscoped, and the other one is ResourceNotFound: 404: Client Error for url: http://10.248.246.143:9696/networking/v2.0/metering/metering-labels, The resource could not be found.; both of these seem to be happening on Tempest. (I'll paste another full log once all | |
| 17:59:42 | M0weng[m] | the tests are done running.) | |
| 18:02:59 | M0weng[m] | (no pressure to respond now because I know it's evening in GMT; just posting these here now :) ) | |
| 18:05:40 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: network: Add '--marker', '--limit' to most list commands https://review.opendev.org/c/openstack/python-openstackclient/+/802482 | |
| 18:14:25 | M0weng[m] | Here is the testing summary at the end of the output: https://paste.openstack.org/show/822336/ | |
| 18:16:00 | M0weng[m] | The rest of the log in parts: | |
| 18:16:02 | M0weng[m] | https://paste.openstack.org/show/822337/ | |
| 18:16:33 | M0weng[m] | https://paste.openstack.org/show/822338/ | |
| 18:17:06 | M0weng[m] | https://paste.openstack.org/show/822339/ | |
| 18:21:57 | M0weng[m] | Relating to the first error, here are my OpenStack-related environment variables (I didn't set them manually; I'm using clouds.yaml):... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/bOofNakCqWJmomGnVfIOSiBO>) | |
| 18:24:53 | M0weng[m] | And this is my clouds.yaml: https://paste.openstack.org/show/822340/ | |
| 18:29:46 | opendevreview | Merged openstack/osc-lib master: Fix pre-commit issues https://review.opendev.org/c/openstack/osc-lib/+/893261 | |
| 18:30:06 | opendevreview | Merged openstack/osc-lib master: Add pre-commit https://review.opendev.org/c/openstack/osc-lib/+/893262 | |
| 18:49:08 | opendevreview | Merged openstack/osc-lib master: Drop formal support for Python 3.6, 3.7 https://review.opendev.org/c/openstack/osc-lib/+/892787 | |
| #openstack-sdks - 2023-11-14 | |||
| 01:03:18 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented heat 'stack suspend' and 'stack resume' function https://review.opendev.org/c/openstack/openstacksdk/+/892156 | |
| 12:14:29 | stephenfin | M0weng[m]: Out of curiosity, do you have *both* `OS_*` environment variables set *and* a `clouds.yaml` file? | |
| 12:14:52 | stephenfin | M0weng[m]: if so, you should only use one or the other. I would unset all the `OS_*` prefixed things | |
| 14:20:45 | opendevreview | Merged openstack/openstacksdk master: Implemented heat 'stack suspend' and 'stack resume' function https://review.opendev.org/c/openstack/openstacksdk/+/892156 | |
| 16:27:31 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: tests: Remove unused flag https://review.opendev.org/c/openstack/python-openstackclient/+/900917 | |
| 16:27:31 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: test: Ignore 'OS_' environment variables https://review.opendev.org/c/openstack/python-openstackclient/+/900918 | |
| 16:32:04 | opendevreview | Stephen Finucane proposed openstack/openstacksdk master: Implement heat 'stack event list' https://review.opendev.org/c/openstack/openstacksdk/+/892836 | |
| 17:36:28 | M0weng[m] | > Out of curiosity, do you have *both* `OS_*` environment variables set *and* a `clouds.yaml` file? | |
| 17:36:28 | M0weng[m] | I was previously setting both, but I noticed and I'm pretty sure I unset them - though it's possible that I didn't unset all of them, or I went back to a saved state that still had them, so I'll check again to see if that works. | |
| 17:37:02 | M0weng[m] | Oops, meant this as a reply to stephenfin | |
| 17:37:52 | stephenfin | M0weng[m]: If you have access to the environment right now, I would suggest checking the output of 'env | grep OS_' | |
| 17:38:22 | stephenfin | M0weng[m]: If you don't have anything set, try running the failing tox tests again. You can do that with e.g. 'tox -e functional -- --failing' | |
| 17:38:57 | stephenfin | (the '--' separates arguments for tox ('-e functional') from arguments for the called executables ('--failing')) | |
| 17:39:05 | stephenfin | (the called executable being 'stestr') | |
| 17:39:38 | stephenfin | M0weng[m]: I'm pretty sure that was your issue though. I've proposed https://review.opendev.org/c/openstack/python-openstackclient/+/900918 to eliminate that in the future | |
| 17:40:11 | M0weng[m] | > If you have access to the environment right now, I would suggest checking the output of 'env | grep OS_' | |
| 17:40:11 | M0weng[m] | This returns empty; it seems nothing is set right now | |
| 17:40:56 | M0weng[m] | Whoops haha, did this on the wrong machine XD | |
| 17:41:14 | M0weng[m] | I was on my DevStack machine instead of the dev one | |
| 17:41:38 | M0weng[m] | Ok, my dev environment does have OS_ variables set, so I'll try unsetting those now 👍️ | |
| 17:41:38 | M0weng[m] | Ok, my dev environment does have OS_ variables set, so I'll try unsetting those now �� | |
| 17:44:11 | M0weng[m] | Hmm, seems they're still failing after unsetting everything. | |
| 17:45:03 | stephenfin | and this is a standard minimally-configured devstack deployment? | |
| 17:45:21 | M0weng[m] | Yep | |
| 17:46:38 | stephenfin | That error message is coming from the authentication plugin in keystoneclient, which to me says this has to be a client-side issue | |
| 17:47:44 | M0weng[m] | Ah wait, this might be a different issue on my end; my network adapter reset so now the ip address is different | |
| 17:48:23 | stephenfin | ah, that would cause issues alright :) | |
| 17:52:42 | M0weng[m] | Is it necessary to rerun stack.sh for the new ip, or is changing the address in clouds.yaml enough? | |
| 18:02:26 | M0weng[m] | Never mind, seems rerunning is necessary (otherwise it redirects the new ip back to the old ip) | |
| 18:47:05 | M0weng[m] | stephenfin: Ok, I reran everything with the env vars unset (and disabled the openrc file that was setting them previously); here are the ones currently failing: https://paste.openstack.org/show/822351/ | |
| 21:32:19 | opendevreview | Merged openstack/openstacksdk master: requirements: Sort alphabetically https://review.opendev.org/c/openstack/openstacksdk/+/900162 | |
| 21:55:39 | opendevreview | Merged openstack/openstacksdk master: Implement heat 'stack event list' https://review.opendev.org/c/openstack/openstacksdk/+/892836 | |
| 22:02:24 | M0weng[m] | s/redirects/still/, s/the/uses/, s/new ip back to// | |
| #openstack-sdks - 2023-11-15 | |||
| 00:49:11 | opendevreview | SongSuheon proposed openstack/openstacksdk master: Implement heat 'stack resource signal' https://review.opendev.org/c/openstack/openstacksdk/+/894203 | |
| 05:04:53 | opendevreview | SongSuheon proposed openstack/openstacksdk master: Implement heat 'stack resource signal' https://review.opendev.org/c/openstack/openstacksdk/+/894203 | |
| 09:44:33 | stephenfin | M0weng[m]: Okay, I see the metering label issue also. Do you want to try fix that one yourself? | |
| 09:45:29 | stephenfin | M0weng[m]: The issue is that there's a missing extension for the networking service related to metering or labelling a meter. You can use my previous patch as a basis for how to address that scenario https://review.opendev.org/c/openstack/python-openstackclient/+/900607 | |
| 09:46:35 | stephenfin | M0weng[m]: Some more hints. Unlike most other services, neutron stores its API definitions in another project, neutron-lib. You can find the source for that here https://github.com/openstack/neutron-lib Have a look through there and see if you can find the *alias" of the extension you're looking for | |
| 09:48:28 | stephenfin | M0weng[m]: (Extensions have names and aliases - the aliases should be a slug, or lowercase string of ASCII characters and dashes, and are what we search for rather than the name. For example, 'standard-attr-segment' is the alias of the "Standard Attribute Segment Extension" extension | |
| 09:49:03 | stephenfin | Once you find the alias, look at my previous patch again for how to skip tests if an extension is missing | |
| 09:50:15 | stephenfin | M0weng[m]: I think I said this previously, but the reason you're seeing this is because we configure DevStack in a specific manner in our CI jobs. You could of course redeploy DevStack to match one of those configurations, but this is a nice gap in our testing that you've found which we should fix. | |
| 09:52:36 | stephenfin | M0weng[m]: Last point. I also see the other failures, so we obviously need to add more extension checks than just the metering label one. The technique will be the same for all. You can do it all in one patch or separately, whatever is easier | |
| 09:54:02 | stephenfin | M0weng[m]: With one exception: openstackclient.tests.functional.network.v2.test_network.NetworkTests.test_network_dhcp_agent is skipped for me with "SKIPPED: No dhcp_agent_scheduler extension present". We might debug that one together later today | |
| 09:57:51 | stephenfin | M0weng[m]: Wait, never mind, I figured it out. My DevStack is an older deployment. I'm likely missing this change https://review.opendev.org/c/openstack/neutron/+/897528 I will talk to the neutron folks about it | |
| 10:06:45 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: Blacken code https://review.opendev.org/c/openstack/osc-lib/+/893263 | |
| 10:06:45 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: pre-commit: Integrate black https://review.opendev.org/c/openstack/osc-lib/+/893264 | |
| 10:06:46 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: Ignore black changes https://review.opendev.org/c/openstack/osc-lib/+/893265 | |
| 10:06:46 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: tox: Use pre-commit for style checks https://review.opendev.org/c/openstack/osc-lib/+/893266 | |
| 10:06:47 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: Remove lower-constraints.txt https://review.opendev.org/c/openstack/osc-lib/+/893267 | |
| 11:33:49 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: compute: Address bug in shelve offload logic https://review.opendev.org/c/openstack/python-openstackclient/+/901019 | |
| 11:33:50 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: Use CommandError, not SystemExit, to exit https://review.opendev.org/c/openstack/python-openstackclient/+/901020 | |
| 11:33:50 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: Remove unnecessary file https://review.opendev.org/c/openstack/python-openstackclient/+/901021 | |
| 11:33:51 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: parseactions: Use ArgumentError, not ArgumentTypeError https://review.opendev.org/c/openstack/python-openstackclient/+/901022 | |
| 11:34:16 | opendevreview | Stephen Finucane proposed openstack/osc-lib master: parseactions: Use ArgumentError, not ArgumentTypeError https://review.opendev.org/c/openstack/osc-lib/+/836204 | |
| 11:41:19 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: test: Ignore 'OS_' environment variables https://review.opendev.org/c/openstack/python-openstackclient/+/900918 | |
| 11:54:53 | opendevreview | Merged openstack/osc-lib master: Blacken code https://review.opendev.org/c/openstack/osc-lib/+/893263 | |
| 12:29:48 | opendevreview | Merged openstack/osc-lib master: pre-commit: Integrate black https://review.opendev.org/c/openstack/osc-lib/+/893264 | |
| 12:30:04 | opendevreview | Merged openstack/osc-lib master: Ignore black changes https://review.opendev.org/c/openstack/osc-lib/+/893265 | |
| 12:30:06 | opendevreview | Merged openstack/osc-lib master: tox: Use pre-commit for style checks https://review.opendev.org/c/openstack/osc-lib/+/893266 | |
| 12:30:07 | opendevreview | Merged openstack/osc-lib master: Remove lower-constraints.txt https://review.opendev.org/c/openstack/osc-lib/+/893267 | |
| 17:01:30 | opendevreview | Stephen Finucane proposed openstack/python-openstackclient master: tests: Check for DHCP agents first in DHCP test https://review.opendev.org/c/openstack/python-openstackclient/+/901054 | |
| 17:03:44 | diablo_rojo | stephenfin: gtema coming to the meeting? | |
| 17:03:55 | stephenfin | yup, zoom is being a pain | |
| 17:04:26 | diablo_rojo | stephenfin: its totally one of those days | |
| 17:04:32 | gtema | yupp, sorry | |
| 17:24:27 | stephenfin | gtema: You haven't pushed the updates to any of that fake generator stuff yet, I assume? | |
| 17:24:41 | stephenfin | I've seen the previous stuff but nothing new recently | |
| 17:24:42 | gtema | nope | |
| 17:25:30 | gtema | I am working hard on getting openapi generated from sources and it 1) pulls lot of energy 2) requires small changes into the generators which I group for later | |
| 17:25:48 | stephenfin | ack okay, I'll wait and watch in that case | |
| 19:46:01 | M0weng[m] | Is it possible to use is_extension_enabled() from a classmethod rather than with self? (Trying to skip the meter_rule test if the metering extension isn't enabled.) | |
| #openstack-sdks - 2023-11-16 | |||
| 10:21:50 | stephenfin | M0weng[m]: You can only use other classmethods and staticmethod (plus external functions, obviously) from classmethods, so if 'is_extension_enabled' is a classmethod then you can use it from another classmethod | |
| 10:22:32 | stephenfin | M0weng[m]: I should rephrase that: classmethods are only allowed to call other classmethods and staticmethods. They can't use instance methods. | |
| 16:13:41 | opendevreview | Merged openstack/os-service-types master: Add Python 3.8 and 3.9 to supported runtimes https://review.opendev.org/c/openstack/os-service-types/+/843266 | |
| 18:26:03 | opendevreview | Artem Goncharov proposed openstack/openstacksdk master: WIP: Implement auto-code generation https://review.opendev.org/c/openstack/openstacksdk/+/882326 | |
| 18:27:21 | M0weng[m] | stephenfin: Thanks! I didn't realize there were two different `is_extension_enabled()`s; I was looking at the non-classmethod one by mistake, no wonder things didn't seem to make sense haha | |
| 19:38:01 | frickler | gtema: regarding the sdk cache issue, looking at the implementation it seems that there is no file locking or atomic updates happening, so the errors that yatin found seems very plausible to me https://github.com/jaraco/keyrings.alt/blob/main/keyrings/alt/file_base.py | |
| 19:38:31 | frickler | not sure if other backends would work better or maybe the whole library isn't suited for this usecase | |
| 19:40:30 | gtema | frickler: from the failed jobs I haven't seen a single one where cache seemed like an issue, there were always unrelated issues. So I would be interested to see whether after disabling it failure rate decrease | |
| 19:42:03 | frickler | gtema: did you look at the cache file contents that were posted in the bug report? those looked pretty convincing to me | |