May 2023 - openstack-discuss - lists.openstack.org

[cinder] cinder-backup volume stuck in creating
by Satish Patel 27 Jul '23

27 Jul '23

Folks, I have configured nova and cinder with ceph storage. VMs running on ceph storage but now when i am trying to create a backup of cinder volume its getting stuck on creating and doing nothing. Logs also do not give any indication of bad. My cinder.conf [DEFAULT] enabled_backends = rbd-1 backup_driver = cinder.backup.drivers.ceph.CephBackupDriver backup_ceph_conf = /etc/ceph/ceph.conf backup_ceph_user = cinder-backup backup_ceph_chunk_size = 134217728 backup_ceph_pool = backups backup_ceph_stripe_unit = 0 backup_ceph_stripe_count = 0 restore_discard_excess_bytes = true osapi_volume_listen = 10.73.0.181 osapi_volume_listen_port = 8776 Output of "openstack volume service list" showing cinder-backup service is up but when i create a backup it's getting stuck in this stage and no activity. I am not seeing anything getting transferred to the ceph backups pool also. Any clue? or method to debug? # openstack volume backup list --all +--------------------------------------+------+-------------+----------+------+ | ID | Name | Description | Status | Size | +--------------------------------------+------+-------------+----------+------+ | bc844d55-8c5a-4bd3-b0e9-7c4c780c95ad | foo1 | | creating | 20 | +--------------------------------------+------+-------------+----------+------+

5 7

Re: [all] [oslo.messaging] Interest in collaboration on a NATS driver
by Tobias Urdin 22 Jul '23

22 Jul '23

Hello Kai, Great to hear that you’re interested! I’m currently just testing this as a POC directly in devstack so there is real-world testing (not should it yet I think, needs more development first). Best regards Tobias > On 30 Aug 2022, at 18:32, Kai Bojens <kb(a)kbojens.de> wrote: > > Am 29.08.22 um 15:46 schrieb Tobias Urdin: > >> If anybody, or any company, out there would be interested in collaborating in a project to bring this support and maintain it feel free to >> reach out. I’m hoping somebody will bite but atleast I’ve put it out there for all of you. > > Hi, > I am very much interested to take a closer look at your work and maybe contribute to it. Although I'm working with OpenStack for my employer at the moment, I'd do this in my spare time as I'm not sure that I can convince him to add another staging system with a full OpenStack installation just for developing a RabbitMQ replacement. That's not on our agenda as we are mostly users and not developers of OpenStack. > > As I'm pretty new to the messaging topic and had just heard of NATS and your idea, I'll first would have to dive into NATS and then I would take a closer look at your code and maybe try to create some documentation or tests. So, I can't promise anything but as I said I'm very interested in your approach as I also see the massive load from RabbitMQ. > > This brings me to the most important question: Where do you run and test your code? > > Greetings, > Kai

8 30

[ptl] Need PTL volunteer for OpenStack Vitrage
by Ghanshyam Mann 11 Jul '23

11 Jul '23

Hi Eyal, I am reaching out to you as you were PTL for OpenStack Vitrage project in the last cycle. There is no PTL candidate for the next cycle (2023.2), and it is on the leaderless project list. Please check if you or anyone you know would like to lead this project. - https://etherpad.opendev.org/p/2023.2-leaderless Also, if anyone else would like to help leading this project, this is time to let TC knows. -gmann

3 3

Experience with VGPUs
by Oliver Weinmann 22 Jun '23

22 Jun '23

Dear All, we are planning to have a POC on VGPUs in our Openstack cluster. Therefore I have a few questions and generally wanted to ask how well VGPUs are supported in Openstack. The docs, in particular: https://docs.openstack.org/nova/zed/admin/virtual-gpu.html explain quite well the general implementation. But I am more interested in general experience with using VGPUs in Openstack. We currently have a small YOGA cluster, planning to upgrade to Zed soon, with a couple of compute nodes. Currently our users use consumer cards like RTX 3050/3060 on their laptops and the idea would be to provide VGPUs to these users. For this I would like to make a very small POC where we first equip one compute node with an Nvidia GPU. Gladly also a few tips on which card would be a good starting point are highly appreciated. I know this heavily depends on the server hardware but this is something I can figure out later. Also do we need additional software licenses to run this? I saw this very nice presentation from CERN on VGPUs: https://indico.cern.ch/event/776411/contributions/3345183/attachments/18516… In the table they are listing Quadro vDWS licenses. I assume we need these in order to use the cards? Also do we need something like Cyborg for this or is VGPU fully implemented in Nova? Best Regards, Oliver

17 36

[ironic] ARM Support in CI: Call for vendors / contributors / interested parties
by Jay Faulkner 14 Jun '23

14 Jun '23

Hi stackers, Ironic has published an experimental Ironic Python Agent image for ARM64 ( https://tarballs.opendev.org/openstack/ironic-python-agent-builder/dib/file…) and discussed promoting this image to supported via CI testing. However, we have a problem: there are no Ironic developers with easy access to ARM hardware at the moment, and no Ironic developers with free time to commit to improving our support of ARM hardware. So we're putting out a call for help: - If you're a hardware vendor and want your ARM hardware supported? Please come talk to the Ironic community about setting up third-party-CI. - Are you an operator or contributor from a company invested in ARM bare metal? Please come join the Ironic community to help us build this support. Thanks, Jay Faulkner Ironic PTL

6 11

[storlets] Proposal to make Train/Ussuri/Victoria EOL
by Takashi Kajinami 13 Jun '23

13 Jun '23

Hello, Currently we have multiple stable branches open but we haven't seen any backport proposed so far. To reduce number of branches we have to maintain, I'd like to propose retiring old stable branches(train, ussuri and victoria). In case you have any concerns, please let me know. Thank you, Takashi Kajinami

1 1

[nova] Nova Spec review day next Tuesday 6 June
by Sylvain Bauza 06 Jun '23

06 Jun '23

Hey Nova community, As a reminder, you can find the Nova agenda for the 2023.2 Bobcat release here https://releases.openstack.org/bobcat/schedule.html As you can see, we plan to do a round of reviews for all the open specifications in the openstack/nova-specs repository by next Tuesday June 6th. If you're a developer wanting to propose a new feature in Nova and you have been said to provide a spec, or if you know that you need to create a spec, or if you already have an open spec, please make sure to upload/update your file before next Tuesday so we'll look at it. Also, if you can look again on your spec during the day, it would be nice as we could try to discuss your spec during this day. Thanks, -Sylvain

2 1

[puppet] puppet module improvements
by Benedikt Trefzer 06 Jun '23

06 Jun '23

Hi all I use the openstack puppet modules to deploy openstack. I'd like to suggest some improvements to the modules and like to know what the community thinks about: 1.) use proper types for parameters Parameter validation is done with 'validate_legacy...' instead of defining the class/resource parameter with a proper type all over the code. I cannot imaging of any advantage not using proper type definitions. Instead using typed parameters would be more efficient an code readability would increase. 2.) params.pp This is the legacy way to define parameter defaults for different OS's. In the modern puppet world a module specific hiera structure is used, which eliminates the need of params.pp class (with inheritance and include). The usage of hiera improves readability and flexibility (every parameter can be overwritten on request, eg. change of packages names etc.) This also eliminate the restriction that the modules can only be used by certain OS'es (osfamily 'RedHat' or 'Debian'). 3.) Eliminate "if OS=='bla' {" statements in code These statements make the code very inflexible. It cannot be overruled if necessary (eg. if I use custom packages to install and do not need the code provided in the if statement). Instead a parameter should be used with a default provided in hiera. Since there is lot of code to change I do not expect this to be done in a single commit (per module) but in steps probably in more than one release cycle. But defining this as best practice for openstack puppet modules and start using above in new commits would bring the code forward. Finally: These are suggestions open for discussion. In no way I like to critic the current state of the puppet modules (which is quite good, but a bit legacy) or the people working on the modules. This is just a feedback/suggestion from an operator using the modules on a daily basis. Regards Benedikt Trefzer

4 4

swift container retention
by Geert Geurts 05 Jun '23

05 Jun '23

Hallo, I've gotten the assignment to implement a backup solution using openstack swift containers. What we want is that the files of customers are kept in the customers own container and for those files to retain their versions on updates of the file. So far so good, this is all possible, and extensively documented online. Now, the second part is more complicated it appears where I thougth this would be a quite common usage example of object storage. I cannot find how I can automagically delete versions of files that are older then X days. There is the X-delete-after header, but this header needs to be added for each transaction. What I want is for all objects in a container to get deleted after X days without having to add a header once the object is uploaded. Does anyone have suggestions on how I could implement this requirement? Please find below a grafical repesentation of what I want Versioning disabled Automated add headers delete-after Versioning enabled Usage on the account of LF Usage on the account of client │ │ │ │ │ │ ▼ ▼ ┌────────────────────────┐ ┌────────────────────────┐ │ │ │ │ │ Versions-container │ │ Current-container │ │ │ │ │ ├────────────────────────┤ ├────────────────────────┤ │ │ │ │ │File1 v123456 │ │ File1 │ │File1 v567899 │ │ File2 │ │ │ │ File3 │ │File3 v234567 │ │ ... │ │ │ │ FileN │ └────────────────────────┘ └─────────────┬──────────┘ ▲ │ │ │ └────────────────────────────────────────┘ Modifications cause current to be moved to Versioned-container Best regards, Geert

2 1

Kubernetes Conformance 1.24 + 1.25
by Kendall Nelson 05 Jun '23

05 Jun '23

Hello All! First of all, I want to say a huge thanks to Guilherme Steinmuller for all his help ensuring that OpenStack Magnum remains Kubernetes Certified [1]! We are certified for v1.24! However, in digging into the set of conformance tests this round, it appears that we will not be able to get 1.25 certified with the current Magnum. This means that in May when we have to re-cert (every 3 months they reach an EOL because that's k8s release cadence) we will not be able to maintain this which is the other part of why I wanted to write this email - to raise awareness and ask for help digging into what needs to happen to make sure we can retain this. Tangentially related, Guilherme has posted a patch with documentation for how to run the conformance tests for yourself and it could use some reviews [2]. And one last thing, I have begun conversation with the opendev folks about what it might take to setup jobs to run the conformance tests periodically so that we don't have to manually run them when it's time to recert. I am planning on submitting a forum topic for this for those interested :) Long live LOKI! - Kendall Nelson [1] https://github.com/cncf/k8s-conformance/pull/2446 [2] https://review.opendev.org/c/openstack/magnum/+/849156

5 10