September 2024 - openstack-discuss

[ceilometer] What's the general status / recommendation for Telemetry in OpenStack
by Waines, Greg 26 Sep '24

26 Sep '24

Hey ... I am working in OpenStack StarlingX which supports a Kubernetes cluster and a containerized OpenStack cluster (via openstack-helm). Our OpenStack offering is fairly simple ... only running Nova, Cinder, Neutron, Keystone, Glance, Horizon and Heat ... @ Antelope. In a predecessor of StarlingX, maybe in the Queens(?) timeframe ... we had a Telemetry solution that consisted of: - Ceilometer (for basic data polling, notifications and collection) - Panko (for event DB) - Gnocchi (for time-series DB) - Aodh (for alarming based on metric thresholds, etc.) HOWEVER ... today (Caracal) ... - I think Gnocchi is still alive and well ... https://github.com/gnocchixyz/gnocchi * but maintained outside of OpenStack - I think OpenStack Panko was deprecated in ~ Wallaby ? * and don't think(?) it is maintained outside of OpenStack ? - The OpenStack Landscape at https://www.openstack.org/software/ * shows only Ceilometer and Aodh MY QUESTIONs : ----------------------- - so what are the current recommendations for OpenStack Telemetry ? - is it just Ceilometer and Aodh ? * Does Ceilometer provide DBs ? (like it used to a long time ago) - is Gnocchi (although maintained outside of OpenStack) still supported in the OpenStack environment as a DB option ? - what is used for event DB ? ... now that panko is gone ... Gnocchi ? Ceilometer ? - are there actually multiple options/solutions for OpenStack Telemetry ? e.g. * Ceilometer / Gnocchi / Aodh * Ceilometer / External Elastic Solution / Aodh * Others ? Greg.

5 5

[powerflex][glance]Dell PowerFlex: After setting the Glance backend to PowerFlex Cinder, uploading an image and creating an image volume does not result in the image volume being created correctly.
by cuiye.liu＠dell.com 26 Sep '24

26 Sep '24

Description of customer infrastructure (Environment) - MANDATORY PowerFlex 3.6.2, 4.5.2 2023.2(Bobcat) Issue/problem detail and customer impact (Description, Impact, Steps to reproduce the issue, How often it happens) - MANDATORY Provide a clear description of the issue After setting the Glance backend to PowerFlex Cinder, uploading an image and creating an image volume does not result in the image volume being created correctly. Clear statement of customer impact unable to use PowerFlex as the backend for Glance, so additional storage is needed Steps to reproduce the issue i.e. commands #openstack volume create --size 8 --type powerflex --image cirros_flex --bootable test_image_volume2 Frequency of the issue everytime The error information is as following: 2024-09-19 07:03:46.320 24 DEBUG cinder.volume.flows.manager.create_volume [None req-c0a6a1fb-e26e-4925-a338-a4c9a68952f1 9c7543d67e184d51a6a77d5046c9e8a1 379ea5a5d4824c7f85e51e77d7c12add - - - -] Volume 37a4ef47-93b5-4395-8cdc-d04498fde0a0: re-scheduling SchedulerAPI.create_volume attempt 3 due to Failed to download image 87ea548b-80ff-4f7b-802b-042debaae696, reason: IOError: 32 Corrupt image download. Hash was cf83e1357eefb8bdf1542850d66d8007d620e4050b5715dc83f4a921d36ce9ce47d0d13c5d85f2b0ff8318d2877eec2f63b931bd47417a81a538327af927da3e expected 1b03ca1bc3fafe448b90583c12f367949f8b0e665685979d95b004e48574b953316799e23240f4f739d1b5eb4c4ca24d38fdc6f4f9d8247a2bc64db25d6bbdb2 _reschedule /var/lib/kolla/venv/lib64/python3.9/site-packages/cinder/volume/flows/manager/create_volume.py:162 It is similar to this bug that is submmited in 2021: 1936314 – Glance image download fails with the error [Errno 32] Corrupt image download. (redhat.com) https://bugzilla.redhat.com/show_bug.cgi?id=1936314 Does anyone knows if it is fixed?

1 0

Problems with the Placement service
by Ron Gage 26 Sep '24

26 Sep '24

4 6

Openstack deployment on a single node server
by Gyan Dev Mishra 25 Sep '24

25 Sep '24

Hi, I am a newbie to OpenStack deployment, we are looking for a deployment strategy for OpenStack to test all our cloud-based apps to save on testing expenses. I tried it with microstack where I faced some issues related to Available Zones which was always empty. So, is it possible to use only one node to deploy and use complete OpenStack or do we need to follow some different strategy. I am having one server machine on which I want to deploy the complete cloud solutions. Kindly provide me with the necessary insights. I really appreciate any help you can provide. Gyan dev, Sr. Software Developer at eCosmos Solutions DISCLAIMER: This communication may contain confidential or copyright information. If you are not an intended recipient, you must not keep, forward, copy, use, save or rely on this communication. Any such action is unauthorized and prohibited. If you have received this communication in error please delete it. Reasonable precautions have been taken to ensure no viruses are present in this email, the company cannot accept responsibility for any loss or damage arising from the use of this email or attachments.

3 2

Retire sqlalchemy-migrate
by Thomas Goirand 25 Sep '24

25 Sep '24

Hi, As much as I know, there's no active OpenStack project using sqlalchemy-migrate anymore. Can someone retire the project, or at least push some indication in the README.rst that the project is retired? When looking at: https://opendev.org/x/sqlalchemy-migrate nothing says so... Cheers, Thomas Goirand (zigo)

3 3

[trove][zaqar][vitrage] What to do with projects not supporting SQLAlchemy 2.x ?
by Thomas Goirand 25 Sep '24

25 Sep '24

Hi, A number of project still don't have support for SQLAlchemy 2.x, which is currently in Debian testing/unstable, and which will be the version in Debian 13 (aka: Trixie). If these project don't get quickly support for SQLAlchemy 2.x, they are going to drop (well, in fact, they were already kicked out of Debian Testing). Are these project still maintained? IMO, no support for SQLAlchemy 2.x is a reason good enough to declare they aren't... Cheers, Thomas Goirand (zigo)

2 2

Network MTU setting in ovn/ovs
by Łukasz Chrustek 25 Sep '24

25 Sep '24

Hi, I have a question about increasing MTU for tenant network. Now it set to 1442 by default on geneve baked tenant network. Is there any drawbacks or it is some kind of bad practice to increase this MTU to 1500 bytes or higher ? We had an issue with one migrated instance from older version of openstack (where was linuxbrigde configured), there mtu was set to 1500, so the settings for i.e. docker inside the vm was set respectively to the higher value. -- Best regards Lukasz Chrustek

4 11

[oslo.db][magnum] QueuePool hangs in Caracal
by Mohammed Naser 25 Sep '24

25 Sep '24

Hi there! The community has identified a bug (https://bugs.launchpad.net/magnum/+bug/2067345) which makes Magnum unusable / entirely broken in Caracal. There is a proposed workaround here: https://bugs.launchpad.net/magnum/+bug/2067345/comments/35 The majority of the issue seems to be around when the changes to remove autocommit were done, I've attempted to switch to enginefacade to prevent these issues (same as Ironic) but with no avail: https://review.opendev.org/c/openstack/magnum/+/927446 There is also a greenthread stacktrace there which shows where/how it's hanging, would it be possible to have someone who might have a bit more insight to help point us in the right direction? I'm more than happy to take this on and keep pushing it forward but I'm not sure if I know where the fix can go at this point. Thanks, Mohammed

2 1

[release][zun][kuryr-libnetwork][kuryr-kubernetes] Immediate action required for 2024.2 (Dalmatian) release
by Dmitriy Rabotyagov 25 Sep '24

25 Sep '24

Hi, As you might know, R-2 week where Release Candidates for all projects following cycle-with-rc cadence should be created has already passed. Nevertheless there are a couple of projects, which do have broken Zuul Jobs[1], which prevents them from being released (tagged with RC1 and create a stable/2024.2 branch). There is a hard deadline by the end of this week (September 27, 2024) to merge required fixes into projects and update corresponsive release patches [2][3][4] in order to be included into the Dalmatian release. Otherwise the Technical Committee will have to take action on the mentioned projects, which can include preventing them from being released and marking them as Inactive ones. So everyone interested in the future of these projects - please take a look at the gate health and help core teams out fixing outstanding issues. Core teams - please prioritise review of proposed patches or come up with fixes during this week. [1] https://review.opendev.org/q/topic:dalmatian-rc1-deadline+is:open [2] https://review.opendev.org/c/openstack/releases/+/928531 [3] https://review.opendev.org/c/openstack/releases/+/928532 [4] https://review.opendev.org/c/openstack/releases/+/928575

1 0

[manila] Propose Gireesh Awasthi to manila core
by Carlos Silva 25 Sep '24

25 Sep '24

Hello everyone, I would like to add Gireesh Awasthi to the OpenStack Manila core team. Gireesh started his OpenStack journey a couple of cycles ago, and caused great impact from the get go while working on some complex features that impacted all of the manila code base. He's also been consistently participative in the latest PTGs. With the knowledge acquired in the code base, he then started contributing more to reviews, and became a go-to person for providing feedback on the NetApp ONTAP driver changes, and/or third party driver changes. It is important to note that this is yet another step into maintaining a healthy community with people from different affiliations sharing the responsibility of maintaining an open source project. In case there are no objections, I will add him to the core team in 1 week from now. [1] https://bit.ly/manila-reviews-90-days Thank you! Regards, carloss

2 2