June 2024 - openstack-discuss - lists.openstack.org

[neutron] Please hold off on merging patches
by Brian Haley 07 Jun '24

07 Jun '24

Hi Neutrinos, It was noticed today by Mohammed Naser that the neutron gate was only running the docs and pep8 jobs. We were able to track it down to the RE2 patches that were recently merged [0]. Until the reverts land please don't +W anything since it could merge untested and would just need to be reverted. Other projects might have similar issues if they updated code using the 'negate' directive for regex recently as we found it doesn't actually work with RE2 when using negative lookahead [1]. Thanks, -Brian [0] https://review.opendev.org/q/topic:"files-negate" [1] https://github.com/google/re2/issues/156

4 8

oslopolicy-policy-generator: what is the namespace to use to get Mgnum policy?
by Michel Jouvin 07 Jun '24

07 Jun '24

Hi, I'd like to see the default Magnum (Oslo) policy used with Magnum. With other services, I use oslopolicy-policy-generator and use the service name as the namespace but for Magnum, "--namespace magnum" returns an error saying that there is no "magnum" namespace. Does anybody know what is the namespace to use with Magnum? Or how to display the currently active policy, preferably in YAML format? Thanks in advance. Cheers, Michel

2 6

[all][tc][ptls] Eventlet: Python 3.13 - Attach your seatbelt
by Herve Beraud 07 Jun '24

07 Jun '24

Hi all, As Python 3.13 is close to become a reality, new concerning incompatibilities between this new CPython version and Eventlet also become a reality: - https://github.com/eventlet/eventlet/issues/964 Fedora Python maintainers proposed a patch, but for now the problem remains unsolved. Also, many people reported bugs related to incompatibility between python3.12/dnspython/eventlet/pip-system-cert. There is an example who summarize well the problem https://github.com/eventlet/eventlet/issues/957 Do you have some visibility concerning the adoption of Python 3.13 into Openstack? Turbulences ahead... -- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/

2 1

[swift] clean up orphan segment parts
by Rafael Tavares Oliveira 07 Jun '24

07 Jun '24

Hey there! We've been using swift and we came across this bug https://bugs.launchpad.net/swift/+bug/1813202. We decided to work it out and have an ongoing patch to solve this issue alongside ronaldo.maia(a)luizalabs.com, we would like to share and get some feedback before we submit it. To solve this, we took inspiration in obj/expirer.py: In S3 controllers we register task objects in an internal account, each task object points to an orphaned slo object, which can be created in two situations, when a multipart-upload is not completed or when a multipart object is overwritten. These tasks will be checked by the daemon (a new one we made called "s3-slo-expirer") after some time and deleted from the base accounts. Rafael

2 2

[all][tc] Eventlet migration: Call to action
by Jay Faulkner 07 Jun '24

07 Jun '24

Hi all, As you may have read in Goutham's TC summary, the existing eventlet goal[0] is unlikely to gain consensus to merge. This appears to be due to the difference in needs between OpenStack projects, and belief that it's unlikely a single solution would serve them all. Our current situation is: 1) Eventlet usage must be discontinued. The temporary maintainers are just that -- temporary -- and we should not expect eventlet to continue to be a valid path moving forward. 2) We have basic agreement that shared tooling, such as oslo libraries, should not dictate a specific threading model. In lieu of specific technical leadership and guidance from the TC, this leaves it to individual projects to decide what threading model to use instead of eventlet monkey patching. The unmerged spec[0] may be a good starting point if you're wondering where to start. Importantly: do not let this work wait! It is a significant change and will take a long time to complete. Thanks, Jay Faulkner 0: https://review.opendev.org/c/openstack/governance/+/902585

6 7

[neutron] Drivers meeting cancelled tomorrow
by Brian Haley 06 Jun '24

06 Jun '24

Hello Neutrinos: Due to the lack of agenda, Friday's drivers meeting is cancelled. Have a nice weekend! -Brian

1 0

[ironic] Failed when writing image to the device file /dev/sda
by James Leong 06 Jun '24

06 Jun '24

Hi all, I have been exploring ironic, with a simple setup. However, I get the following error messages whenever I attempt to deploy an ISO image to my baremetal node. 1. An agent token generation request is being refused as one is already present for the node 2. Writing image to device /dev/sda failed with exit code 32. stdout: . stderr: lsblk: /dev/sda2: not a block device 3. Invalid or missing agent_token received for node I am not quite sure what is required to be set on my node's property other than vendor and cpu_arch. Any help would be appreciated. Best, James

1 0

[kolla-ansible][Antelope][Keystone] Redelegation Not Working with Heat
by Devin Patterson 06 Jun '24

06 Jun '24

Hello! We are attempting to use role redelegation with Heat and Keystone to deploy a stack that deploys Zaqar subscriptions among other resources. When attempting to scale up the stack, we receive this error: TypeError: '<' not supported between instances of 'NoneType' and 'int') This appears to be sourced from this check in the code in the Keystone core.py file. # expiry times trust_expiry = trust.get('expires_at') redelegated_expiry = redelegated_trust['expires_at'] if trust_expiry: # redelegated trust is from backend and has no tzinfo if redelegated_expiry < trust_expiry.replace(tzinfo=None): raise exception.Forbidden( _('Requested expiration time is more ' 'than redelegated trust can provide')) The issue appears to be related to 'None' values for what is passed into 'expires_at' for 'redelegated_expiry'. We have this config in keystone.conf: [trust] allow_redelegation = True And this in heat.conf: reauthentication_auth_method = trusts allow_trusts_redelegation = True This is how we are scaling the stack: openstack stack resource signal <stack_name> scale_up_policy Any suggestions on how to fix this issue? Thanks, Devin

1 0

[masakari] DB migration failure during upgrade to 2024.1
by Dmitriy Rabotyagov 06 Jun '24

06 Jun '24

Hey folks, Just wanted to let you know, that upgrade of Masakari to 2024.1 is most likely borked due to issue in Alembic migration, where masakari considers taskflow DB as its own, as masakari-manage does not pass correct namespace for alembic. Upgrade path is not tested in CI, which made it possible to slip the issue from the radar. There's a patch existing [1] that fixes the root cause of the issue, though needs to be reviewed and backported. Thus, I wanted to call out Masakari Maintainers for some attention to this issue as well as give operators some awareness about the issue. [1] https://review.opendev.org/c/openstack/masakari/+/920034

1 0

[stackalytics] no "neutron" module for Commits
by Ihar Hrachyshka 06 Jun '24

06 Jun '24

Hi, I noticed that "neutron" is not an option in the Module list when Commits are selected, and no neutron commits are registered since ~Oct 2023. On the other hand, other Neutron program repos (e.g. neutron-lib and neutron-tempest-plugin) are available in the list and do register commits. Also, when Reviews are selected, neutron module is back as an option, and review stats are correctly updated. I checked governance yaml and it doesn't strike me as incorrect: https://opendev.org/openstack/governance/src/commit/5f93b99a3c90b452f6c7368… Any ideas what may have happened? Thanks, Ihar

4 7