| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2024-05-26 | |||
| 09:28:57 | ZonC | Hi, I have started working a bit with the openstackSDK and have run into a wall right now and was wondering if anyone could provide some help? The issue I am facing is in my code I want to do a offline migration to a specific host but the compute.migrate_server only take the server to migrate and not any other arguments. Is there anything I am missing in the documentation or do openstackSDK not have support for this as of | |
| 09:28:57 | ZonC | yet? | |
| 09:30:16 | ZonC | So basically I want to be able to target a host when offline migrate. Like the SDK does in the live_migrate_server. | |
| #openstack-sdks - 2024-05-27 | |||
| 07:03:21 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: WIP: Add support for default volume types https://review.opendev.org/c/openstack/openstacksdk/+/920464 | |
| 07:04:54 | opendevreview | Rajat Dhasmana proposed openstack/python-openstackclient master: WIP: Add support for default volume type commands https://review.opendev.org/c/openstack/python-openstackclient/+/920465 | |
| 10:48:16 | dtantsur | I wonder what the chances are that https://opendev.org/openstack/openstacksdk/commit/2b4aeff6d3972761da9c4a6a8b696103664a758b causes memory leaks in the code that worked before | |
| 10:57:56 | frickler | dtantsur: do you actually see some memory leak or is this just a theoretical question? | |
| 10:58:34 | dtantsur | frickler: I do see ironic-inspector leaking memory. Taking periodical snapshows from Python's tracealloc led me here: https://paste.opendev.org/show/b9HRCbkxqtPrZ3yU0TCH/ | |
| 10:58:48 | dtantsur | but after looking at lines, I'm not longer sure it's related to __del__ | |
| 10:59:28 | dtantsur | I'm now wondering if we somehow make CONF grow infinitely | |
| 11:01:22 | dtantsur | At the very least, it looks like we're modifying global defaults.. | |
| 11:01:40 | dtantsur | mmm, or not really, it has copy() everywhere | |
| 11:09:37 | frickler | dtantsur: which version are you using? there was a leak fixed a short time later https://review.opendev.org/c/openstack/openstacksdk/+/890781 | |
| 11:10:09 | dtantsur | frickler: it's possible that I don't have the newest version, let me check | |
| 11:10:54 | dtantsur | alas, I have this change already | |
| 12:45:56 | opendevreview | Jens Harbott proposed openstack/openstackclient master: zuul: Drop retired repos from required-projects https://review.opendev.org/c/openstack/openstackclient/+/920512 | |
| 13:26:28 | opendevreview | Dmitry Tantsur proposed openstack/openstacksdk master: Do not create a class in runtime on a potentially hot path https://review.opendev.org/c/openstack/openstacksdk/+/920541 | |
| 13:26:40 | dtantsur | Okay, this is the first, unfortunately by far the smallest, part of the puzzle ^^ | |
| 14:01:06 | dtantsur | frickler, so, is now calling close() on a Connection mandatory? I don't think it was always the case. | |
| 14:04:57 | dtantsur | Honestly, using atexit smells badly. Are you sure you guys considered long-running users of openstacksdk? | |
| 14:10:06 | frickler | dtantsur: I've so far successfully avoided to dig into performance relevant details of python implementations, will defer to gtema and stephenfin | |
| 14:10:54 | gtema | dtantsur - token caching currently relies on calling close | |
| 14:11:08 | dtantsur | gtema: it's off by default though? | |
| 14:11:12 | dtantsur | I mean, the caching? | |
| 14:11:27 | gtema | well, yes - it is config option | |
| 14:11:34 | gtema | and off by default | |
| 14:11:34 | dtantsur | At the very least, you need a very, VERY ugly warning in the release notes | |
| 14:11:58 | dtantsur | I don't think we're the only people who don't call close() | |
| 14:12:24 | gtema | currently the cli does so (since that is the main user of caching) | |
| 14:13:20 | dtantsur | gtema: related question: are Connections long lived and (green) thread safe? Can I just cache one for the lifetime of ironic-inspector? | |
| 14:13:55 | gtema | what is this lifetime? I have certain usecases with connections living multiple months | |
| 14:14:12 | dtantsur | gtema: it's an OpenStack service. Weeks, months, in a very unlikely case years :) | |
| 14:14:13 | gtema | should be thread safe | |
| 14:15:46 | dtantsur | in any case, it's a pity that the change went in without a release note (or am I not looking carefully enough?) | |
| 14:16:00 | gtema | this is there since years | |
| 14:16:15 | dtantsur | last year, to be precise | |
| 14:17:02 | gtema | ok, could be, time is relative | |
| 14:17:07 | gtema | :) | |
| 14:19:15 | dtantsur | gtema: also for your attention: https://review.opendev.org/c/openstack/openstacksdk/+/920541 | |
| 14:19:37 | gtema | ack | |
| 14:36:21 | opendevreview | Merged openstack/openstackclient master: zuul: Drop retired repos from required-projects https://review.opendev.org/c/openstack/openstackclient/+/920512 | |
| 21:00:16 | opendevreview | Merged openstack/openstackclient master: Remove clients of retired projects https://review.opendev.org/c/openstack/openstackclient/+/919857 | |
| #openstack-sdks - 2024-05-28 | |||
| 12:31:42 | opendevreview | Merged openstack/python-openstackclient master: Remove admin only text from 'port create' help text https://review.opendev.org/c/openstack/python-openstackclient/+/920421 | |
| 14:03:38 | opendevreview | Lajos Katona proposed openstack/openstacksdk master: BGP: return list for get_dragents and get_bgp_speakers_hosted_by_dragent https://review.opendev.org/c/openstack/openstacksdk/+/920670 | |
| 18:57:13 | opendevreview | Merged openstack/openstacksdk master: Allow project switching for Designate API https://review.opendev.org/c/openstack/openstacksdk/+/899027 | |
| #openstack-sdks - 2024-05-29 | |||
| 16:25:29 | stephenfin | M0weng[m]: Took a look at that error message: it's not keystoneclient's fault but rather the fault of '_process_identity_and_resource_options' in 'openstackclient/identity/v3/role.py' | |
| 16:26:31 | stephenfin | M0weng[m]: It's an if-else ladder and there's a missing fallthrough/else case | |
| 16:28:57 | stephenfin | M0weng[m]: IMO, it would be better if we started nesting and had three cases: if user: {do stuff}; elif project: {do stuff}; else {fail} | |
| 16:30:37 | stephenfin | and then another one for if system: {do stuff}; elif project: {do stuff}; elif domain; {do stuff}; else {fail} | |
| 16:30:56 | stephenfin | not sure if the else in that second one is necessary or not. I'll let you figure that out | |
| 17:23:16 | opendevreview | Antonia Gaete proposed openstack/python-openstackclient master: identity: Migrate 'service' commands to SDK https://review.opendev.org/c/openstack/python-openstackclient/+/906395 | |
| 21:46:10 | opendevreview | Rajat Dhasmana proposed openstack/python-openstackclient master: Add support for default volume type commands https://review.opendev.org/c/openstack/python-openstackclient/+/920465 | |
| #openstack-sdks - 2024-05-30 | |||
| 07:38:48 | mikal | Hey, twice now openstacksdk has been broken for python3.8 by mypy changes. This seems to be happening because there isn't any CI for that python version upstream, despite python 3.8 being in the support matrix for 2024.1. I get that the universe is imperfect, but I am mostly catching these by pure luck in a downstream project at the moment. Apart from playing whackamole when I find them, is there something w | |
| 07:38:48 | mikal | e can do to make this less of a thing? | |
| 07:42:43 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: Add support for default volume types https://review.opendev.org/c/openstack/openstacksdk/+/920464 | |
| 08:14:52 | frickler | mikal: are you referring to the stable/2024.1 branch or master? and sure you can do something, create reviews adding testing jobs for what you think is missing | |
| 08:15:48 | mikal | Master is definitely broken for python3.8. I need to rehydrate my state on what pythons OpenStack says they support for 2024.2 to be honest. | |
| 08:16:10 | mikal | I think the issue with upstream CI is I'm not sure there's any distro that has python 3.8 in the supported list any more. | |
| 08:16:20 | mikal | I'm using Rocky 9 out of tree to find these issues. | |
| 08:17:29 | frickler | doesn't that have python3.9 by default? in general, support for py3.8 has been dropped, see https://governance.openstack.org/tc/reference/runtimes/2024.2.html | |
| 08:17:42 | frickler | so testing that would need to be specific to sdk/osc | |
| 08:17:54 | mikal | Yeah, just found the same page. | |
| 08:18:07 | mikal | Definitely the error I see states that you need python 3.10+ for the syntax that is being used. | |
| 08:18:30 | mikal | Let me spin up a quick test VM and double check what version of python it has by default. | |
| 08:24:10 | mikal | Sorry, you're right, Rocky 9 is indeed python 3.9. However, that is still too old for this syntax. | |
| 08:35:04 | frickler | so that's certainly a bug then, that should be fixed | |
| 08:36:44 | mikal | Yeah, I am waiting for my CI to finish running before uploading a fix. I just want to make sure it actually works first. | |
| 08:48:39 | opendevreview | Michael Still proposed openstack/openstacksdk master: Use mypy syntax compatible with older pythons. https://review.opendev.org/c/openstack/openstacksdk/+/920814 | |
| 09:04:35 | frickler | gtema: ^^ iirc we've seen more regressions like this earlier. maybe need a mypy py39 job? (cc stephenfin who has voiced interest in also keeping support for py38) | |
| 09:07:11 | gtema | Have I told how I love modern py madness? | |
| 17:54:19 | opendevreview | Antonia Gaete proposed openstack/python-openstackclient master: identity: Migrate 'service' commands to SDK https://review.opendev.org/c/openstack/python-openstackclient/+/906395 | |
| 23:48:03 | M0weng[m] | <stephenfin> "0weng: Took a look at that error..." <- Ahh I see; I'll take a look at that. Should I fix this in the same patch as the SDK migration or make a separate one? | |
| #openstack-sdks - 2024-06-01 | |||
| 12:28:09 | opendevreview | Tomas Rinblad proposed openstack/openstacksdk master: Compute, migrate: Add support for targeting host. https://review.opendev.org/c/openstack/openstacksdk/+/920948 | |
| 16:51:11 | opendevreview | Tomas Rinblad proposed openstack/openstacksdk master: Compute, migrate: Add support for targeting host. https://review.opendev.org/c/openstack/openstacksdk/+/920948 | |
| #openstack-sdks - 2024-06-03 | |||
| 17:14:15 | opendevreview | Antonia Gaete proposed openstack/python-openstackclient master: identity: Migrate 'service' commands to SDK https://review.opendev.org/c/openstack/python-openstackclient/+/906395 | |
| 18:27:23 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: Increase interval between volume GET calls https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 19:15:37 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: Add support for default volume types https://review.opendev.org/c/openstack/openstacksdk/+/920464 | |
| #openstack-sdks - 2024-06-04 | |||
| 13:36:26 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: WIP: Increase interval between volume GET calls https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 13:38:13 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: Add support for default volume types https://review.opendev.org/c/openstack/openstacksdk/+/920464 | |
| 13:39:57 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: WIP: Increase interval between volume GET calls https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| #openstack-sdks - 2024-06-05 | |||
| 07:52:34 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: WIP: Increase interval between volume GET calls https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 10:49:03 | dtantsur | TheJulia: mmm, our openstacksdk job fails on RBAC stuff now.. | |
| 10:49:30 | dtantsur | I mean, the functional one | |
| 14:16:45 | opendevreview | Merged openstack/openstacksdk master: baremetal: Decode 'config_drive' argument to 'set_provision_state' https://review.opendev.org/c/openstack/openstacksdk/+/889690 | |
| 15:00:52 | TheJulia | dtantsur: if you can point me to, happy to take a look | |
| 15:25:38 | dtantsur | TheJulia: https://zuul.opendev.org/t/openstack/build/62accb8ae9ce47dabdcbfc2846f9e93c caught my attention | |
| 15:26:28 | TheJulia | lets put a bug into launchpad and tag ironic on it | |
| 16:24:14 | opendevreview | Antonia Gaete proposed openstack/python-openstackclient master: identity: Migrate 'service' commands to SDK https://review.opendev.org/c/openstack/python-openstackclient/+/906395 | |
| 18:42:40 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: [func test] Fix race between attachment delete and server delete https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 20:50:28 | opendevreview | Oria Weng proposed openstack/openstacksdk master: Identity: Add support for `options` property for roles https://review.opendev.org/c/openstack/openstacksdk/+/921411 | |
| 21:02:19 | M0weng[m] | stephenfin: Quick question about the role commands - should `inherited` no longer be an allowable argument for the `role add` commands in openstackclient then? | |
| 21:13:18 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: [func test] Fix race between attachment delete and server delete https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 22:47:41 | opendevreview | Oria Weng proposed openstack/python-openstackclient master: Identity: Migrate 'role assignment' commands to SDK https://review.opendev.org/c/openstack/python-openstackclient/+/910664 | |
| #openstack-sdks - 2024-06-06 | |||
| 08:06:05 | opendevreview | Rajat Dhasmana proposed openstack/openstacksdk master: [func test] Fix race between attachment delete and server delete https://review.opendev.org/c/openstack/openstacksdk/+/921062 | |
| 18:00:32 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: title : Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921468 | |
| 18:02:00 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921468 | |
| 18:03:27 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921469 | |
| 18:04:31 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921469 | |
| 18:08:32 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921471 | |
| 18:19:20 | opendevreview | Baek Seung Ju proposed openstack/openstacksdk master: Implemented snapshot-related features of the stack. https://review.opendev.org/c/openstack/openstacksdk/+/921471 | |