December 2024 - openstack-discuss

[nova]Intel RDT CAT Support Missing from OpenStack
by Junya Noguchi (Fujitsu) 20 Jan '25

20 Jan '25

To whom it may concern I have a question regarding the lack of Intel RDT CAT implementation in OpenStack We are planning to commit Resource Monitoring functionality to Openstack for ARM, and are exploring the implementation of Intel RDT on Openstack. We found the following blueprint for the Intel RDT CAT implementation in nova. https://blueprints.launchpad.net/nova/+spec/cat-support However, the Intel RDT CAT feature is not currently implemented in OpenStack. Has the implementation of Intel RDT CAT functionality in Openstack been rejected? Or is it just a development delay? Best regards, Junya Noguchi.

5 7

[glance][ops] proposal to remove the glance-scrubber
by Brian Rosmaita 20 Jan '25

20 Jan '25

Hello Glance users, The Glance project team wants to determine whether anyone is using the glance-scrubber utility [0]. If not, we are thinking of deprecating it this cycle to be removed in the appropriate future cycle. If you have not previously heard of the glance-scrubber, that's most likely because it's only useful for particular deployment configurations of which yours is not one. See the doc [0] for details if you are curious. Our suspicion is that the glance-scrubber may have outlived its usefulness for modern OpenStack configurations. (The last time I remember it being discussed at a PTG was during the snowstorm in Dublin, which I believe was the Rocky development cycle). On the other hand, the team also has some ideas for improving the scrubber, but there's no point doing that if it's not actually being used in the first place. So if removal of the glance-scrubber would negatively affect you, please respond to this email so we can get some data points. Thanks! [0] https://docs.openstack.org/glance/latest/cli/glancescrubber.html

2 2

[i18n][infra][horizon] Zanata translation job failures after Python 3.8 deprecation
by Ian Y. Choi 19 Jan '25

19 Jan '25

Hi all, During last week's PTG, the I18n SIG identified Zanata translation job failures for Horizon (and likely related plugins). The root cause appears to be Python 3.8 deprecation on Horizon [1]. Failed jobs have been observed on Zuul [2], and translation proposals were made on Gerrit [3]. Since translation sync on Horizon stable branches is functioning correctly, the error message seems to be related to Python 3.8 deprecation: ERROR: Cannot install -r doc/requirements.txt (line 3) due to conflicting package version dependencies. To address this, I'd like to raise two points: 1/ To resolve the issue, would you support updating translation sync jobs with Bionic-based Zuul jobs by either upgrading to Jammy or modifying the Bionic image to support Python >3.8? 2/ At least me from I18n SIG acknowledges that recognizing these translation job errors was delayed. Moving forward, would it be beneficial to implement periodic (e.g., monthly) reporting on translation job successes and failures? Thank you, /Ian [1] https://docs.openstack.org/releasenotes/horizon/2024.2.html#upgrade-notes [2] https://zuul.opendev.org/t/openstack/builds?job_name=propose-translation-up… [3] https://review.opendev.org/q/owner:proposal-bot+project:openstack/horizon+b…

2 2

[openstack][octavia] How clients can debug their LB
by Nguyễn Hữu Khôi 15 Jan '25

15 Jan '25

Hello. I would like to know how clients can debug their LB such as access log or monitor LB performance? Thank you. Regards Nguyen Huu Khoi

5 7

[cinder][kubernetes] pvc stuck in reserved state
by Eugen Block 10 Jan '25

10 Jan '25

Hi, we have a couple of k8s clusters in our openstack environment (Victoria) and are facing some problems when pods get moved to different worker nodes. The PVCs get stuck in "reserved" state, and they only get attached after resetting the volumes to "available" state. First I thought it was due to the missing multiattach volume type which seems to work fine in our customer's cluster. I also found the "duplicate connector" error messages in the nova-compute.log: > Unable to update attachment.(Invalid volume: duplicate connectors > detected on volume But after adding the multiattach volume type and retyping the affected volumes, the PVCs still get stuck in "reserved" state while I don't see any error messages on the hypervisors anymore. So it seems that something has changed/improved, but now I don't have a trace where to look next. Could this be a cinder-csi driver issue or something? Our cinder backend is ceph (Pacific). Not sure which information could help so please ask for anything I can provide. Any pointers are highly appreciated! Thanks! Eugen

1 2

[nova] How to gain approval for a new feature on unavailable hardware
by taketani.ryo＠fujitsu.com 08 Jan '25

08 Jan '25

To Whom It May Concern, We are considering to propose and develop a new feature on OpenStack Nova for building Arm Confidential Computing Architecture (Arm CCA)-encrypted instances. I have some questions for doing it. * While specification of Arm CCA is already open(https://developer.arm.com/documentation/den0125/0300/Overview), No Arm CCA-enabled hardware is currently available to our knowledge. Therefore, we plan to conduct testing in an internal environment. * Would a proposal for a feature utilizing publicly unavailable hardware be acceptable to OpenStack? Alternatively, in this case, what steps or preparations are required for approval? * Should the blueprint and specification be submitted simultaneously? * The specification is currently incomplete. Building Arm CCA instances requires libvirt APIs, similar to AMD SEV instances(https://blueprints.launchpad.net/nova/+spec/amd-sev-libvirt-suppo…, but specification of these are not yet determined publicly. We are considering submitting the blueprint first to inform the community of our ongoing development efforts. Regards

3 5

Nova create vm with special ipv4 and ipv6
by Alex Song (宋文平) 07 Jan '25

07 Jan '25

Hi Nova team, does nova support create vm with special ipv4 and ipv6? As port create with ipv4 and ipv6 works by: openstack port create 111 --network 10851e96-8995-4988-b819-0514b38d7353 --fixed-ip subnet=a564ac90-3ff3-47d7-8ae5-20e2666eaa3c,ip-address=172.24.4.54 --fixed-ip subnet=afa3c858-c3d3-4784-8964-2c46840b714d,ip-address=2001:db8::183 but vm create with ipv4 and ipv6 with the cmd openstack server create --flavor 1 --image 62254c92-a579-4c8a-a2f4-f2624cef8ec1 --nic net-id=10851e96-8995-4988-b819-0514b38d7353,v4-fixed-ip=172.24.4.53,v6-fixed-ip=2001:db8::188 test raise exception: openstack server create: error: argument --nic: Invalid argument net-id=10851e96-8995-4988-b819-0514b38d7353,v4-fixed-ip=172.24.4.53,v6-fixed-ip=2001:db8::188; either 'v4-fixed-ip' or 'v6-fixed-ip' should be specified but not both. So why nova doesnot support?

2 3

Failing at TASK [glance : Copy over multiple ceph configs for Glance]
by Karera Tony 04 Jan '25

04 Jan '25

Hello Team, I am deploying Openstack release 2024.02 with External Ceph. However I am encountering the error below. But it is failing here TASK [glance : Copy over multiple ceph configs for Glance] Please assist Regards Tony Karera

2 1

hexchat irc channel
by Alex Song (宋文平) 31 Dec '24

31 Dec '24

Hello, I update hexchat and which network is available now? Thanks [cid:image001.png@01DB5782.29D8F150] [cid:image002.png@01DB5782.29D8F150] [cid:image003.png@01DB5782.29D8F150]

3 5

How to find the official documentation for installing OpenStack services and how to locate the historical documentation
by ゞPositive î 31 Dec '24

31 Dec '24

 https://docs.openstack.org/install-guide/openstack-services.html# I am installing the OpenStack zed version. How can I find the zed version document in this official document, as there are only the last four versions available ゞPositive ? 451526204(a)qq.com  

2 1

OpenStack zed version seeking help
by 451526204＠qq.com 30 Dec '24

30 Dec '24

Identity service Image service Placement service Compute service Networking service Dashboard Block Storage service In the official documentation, when installing OpenStack services, there are only the last four versions in the interface of the above services. So where can I find the zed version when installing it?

2 1

How to find the official documentation for installing OpenStack services and how to locate the historical documentation
by ゞPositive î 29 Dec '24

29 Dec '24

 https://docs.openstack.org/install-guide/openstack-services.html# I am installing the OpenStack zed version. How can I find the zed version document in this official document, as there are only the last four versions available ゞPositive ? 451526204(a)qq.com  

1 0

[ceilometer]Intel RDT CMT support on OpenStack
by Junya Noguchi (Fujitsu) 26 Dec '24

26 Dec '24

To whom it may concern I have a question about the support status of Intel RDT CMT in OpenStack. We are planning to commit Resource Monitoring functionality to Openstack for ARM, and are exploring the implementation of Intel RDT on Openstack. We found a document published by Intel on the Intel RDT CMT on OpenStack.[1] However, when I checked the source code for nova and ceilometer, it was modified to ignore the settings in the documentation[1].[2] In addition, ceilometer was written to perform resource monitoring by itself. Is resource monitoring with ceilometer supported? Or is it just old code that is not used? [1] https://www.intel.com/content/www/us/en/developer/articles/training/how-to-… [2] https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L… Best regards, Junya Noguchi.

2 2

[nova] force pinning instances to AZ
by Arnaud Morin 24 Dec '24

24 Dec '24

Hey all, I have a region with multiple AZ. If the server was not created in a specific zone then it is free to be moved to other zones. Is there any way in nova config to prevent this? Regards, Arnaud.

7 10

[Nova] Nova-compute service flapping on Antelope
by Jakub Darmach 24 Dec '24

24 Dec '24

Hello, Recently I encountered an interesting issue - nova-compute service started to temporarily lose Rabbit connectivity to regain it after a few seconds on Antelope. The issue was replicated on the test environment, without specific steps to replicate though - it just starts to happen after some time. I pasted a log example in the bug I opened [1]. First we can see rabbit logging closing AMQP connection, followed by nova-compute reporting rabbit server being unreachable, with next message being successful reconnection. Restarting nova-compute services helps temporarily, the issue starts to manifest after a few days. Initially disconnects once every few hours, each time in shorter intervals, to the point it happens every minute or so. Did anyone encounter something similar? [1] https://bugs.launchpad.net/nova/+bug/2092297 Pozdrawiam / Best regards, *Jakub Darmach*

2 1

[tacker] Cancelling the next two weekly meetings
by Yasufumi Ogawa 24 Dec '24

24 Dec '24

Hi all, Due to the holidays, I'd skip the next two meetings. So, the next our meeting is Jan 14th. Thanks, Yasufumi

1 1

ERROR openstack [-] Resource OS::TripleO::Network::Ports::ControlPlaneVipPort maps to type OS::Neutron::Port and the Neutron service is not available when using ephemeral Heat.| Openstack tripleo wallaby version
by Swogat Pradhan 24 Dec '24

24 Dec '24

Hi, I am currently trying to deploy openstack wallaby using tripleo arch. I created the network jinja templates, ran the following commands also: #openstack overcloud network provision --stack overcloud --output networks-deployed-environment.yaml custom_network_data.yaml # openstack overcloud network vip provision --stack overcloud --output vip-deployed-environment.yaml custom_vip_data.yaml # openstack overcloud node provision --stack overcloud --overcloud-ssh-key /home/stack/sshkey/id_rsa overcloud-baremetal-deploy.yaml and used the environment files in the openstack overcloud deploy command: (undercloud) [stack@hkg2director ~]$ cat deploy.sh #!/bin/bash THT=/usr/share/openstack-tripleo-heat-templates/ CNF=/home/stack/ openstack overcloud deploy --templates $THT \ -r $CNF/templates/roles_data.yaml \ -n $CNF/workplace/custom_network_data.yaml \ -e ~/containers-prepare-parameter.yaml \ -e $CNF/templates/node-info.yaml \ -e $CNF/templates/scheduler-hints.yaml \ -e $CNF/workplace/networks-deployed-environment.yaml \ -e $CNF/workplace/vip-deployed-environment.yaml \ -e $CNF/workplace/overcloud-baremetal-deployed.yaml \ -e $CNF/workplace/custom-net-bond-with-vlans.yaml Now when i run the ./deploy.sh script i encounter an error stating: ERROR openstack [-] Resource OS::TripleO::Network::Ports::ControlPlaneVipPort maps to type OS::Neutron::Port and the Neutron service is not available when using ephemeral Heat. The generated environments from 'openstack overcloud baremetal provision' and 'openstack overcloud network provision' must be included with the deployment command.: tripleoclient.exceptions.InvalidConfiguration: Resource OS::TripleO::Network::Ports::ControlPlaneVipPort maps to type OS::Neutron::Port and the Neutron service is not available when using ephemeral Heat. The generated environments from 'openstack overcloud baremetal provision' and 'openstack overcloud network provision' must be included with the deployment command. 2022-04-19 13:47:16.582 735924 INFO osc_lib.shell [-] END return value: 1 Can someone tell me where the mistake is? With regards, Swogat Pradhan

4 4

[all] Installation guide for openSUSE and SUSE Linux Enterprise
by Takashi Kajinami 23 Dec '24

23 Dec '24

Hello, As you may know, we have been maintaining installation guide for three linux distributions below. - oepnSUSE and SUSE Linux Enterprise - Red Hat Enterprise Linux and CentOS - Ubuntu However, as far as I know, the packages for openSUSE and SLES haven't been built/updated for some time. Also openSUSE was removed from tested runtimes in wallaby and hasn't been tested. So I doulbt that users can use these operating systems to install OpenStack nowadays. Can/Should we remove the guide for openSUSE/SLES Now ? I'm not very sure if we can maintain the guide for these with no details available nowadays. Thank you, Takashi -- Takashi Kajinami irc: tkajinam github: https://github.com/kajinamit launchpad: https://launchpad.net/~kajinamit

2 1

[neutron] Bug deputy Dec 16 - Dec 22
by Rodolfo Alonso Hernandez 23 Dec '24

23 Dec '24

Hello Neutrinos: This is the bug list from last week: HIGH * api slow due to missing LRP index ( https://bugs.launchpad.net/neutron/+bug/2092133) Assigned and patch proposed: https://review.opendev.org/c/openstack/ovsdbapp/+/938004 MEDIUM * [RFE] OVN: allow to configure activation-strategy (change it to a different strategy from rarp; or disable completely) ( https://bugs.launchpad.net/neutron/+bug/2092250) Assigned and patch proposed: https://review.opendev.org/c/openstack/neutron/+/938106 LOW * [OVN] Migrate the OVN L3 scheduler to use HA_Chassis_Group ( https://bugs.launchpad.net/neutron/+bug/2092271) Assigned. * Deprecate and remove vlan_transparent config option ( https://bugs.launchpad.net/neutron/+bug/2092174) *NOT ASSIGNED*. Regards.

1 0

[freezer][freezer-dr] Proposal to retire Freezer DR project
by Dmitriy Rabotyagov 23 Dec '24

23 Dec '24

Hi folks, As an ongoing effort to return Freezer project back to the list of Active projects and recovering gate health for repositories owned by Freezer, I've come across freezer-dr[1] Freezer DR is a separate from freezer/freezer-api instance-ha project which is highly pluggable by design with the main goal to evacuate instances from the compute node once service detects that the serving compute node is down. And that is what is currently achieved by the Masakari project. While pluggable design for the service, can be treated as a good benefit in favor of it. Another relatively valuable benefit of freezer-dr, is that it uses nova service state to detect if the compute is down, rather then require 3rd party cluster tooling to independently evaluate hypervisor health. Regardless, I don't see enough value to keep maintenance of that project and instead it's better to keep focus on Masakari maintenance/development instead. With that being said, I suggest the retirement of the freezer-dr project. If someone feels in favor of keeping maintenance of a separate project for instance evacuation - please feel free to reply in thread until December 10, 2024. After that date I will initiate the project retirement process. [1] https://opendev.org/openstack/freezer-dr

2 2

[magnum][cluster api] Heads-up - Octavia OVN driver and latest CAPO
by Stig Telfer 20 Dec '24

20 Dec '24

Hi Magnum folks - Heads up - we wanted to make people aware of a recent upstream CAPO change that could break your Magnum setup if using Cluster API and the Octavia OVN driver. What we've seen is that a recent contribution to CAPO tightens the security group rules for NodePorts to the local tenant network. This works fine for Octavia's Amphora driver, in which HAproxy becomes the source for all back-end API requests. The Octavia OVN driver doesn't have that behaviour, which results in inbound requests getting filtered. The CAPO release is v0.11.0 (https://github.com/kubernetes-sigs/cluster-api-provider-openstack/releases/…) The issue is described here: https://github.com/kubernetes-sigs/cluster-api-provider-openstack/issues/23… The Magnum drivers will pin CAPO driver versions so it's more of a developer / power-user concern than for conventional operators. Cheers, Stig

1 0

[masakari] is this project still active and can be used in production environment
by luckydog xf 20 Dec '24

20 Dec '24

Hi, list, I saw this project hasn't been active for months. So I'm wondering if it is still actively developed and can it be used in production? Are there any use cases ? Thanks.

4 8

Openstack ansible Upgrades
by engineer2024 20 Dec '24

20 Dec '24

Is it possible to upgrade openstack ansible from one major release to another, along with the os distribution upgrade all in one flow ? I mean, as of now we are doing the upgrade as two step process. first we are upgrading the os of the hosts and then the os of the lxc containers and then re running the existing openstack version playbooks. Next , we are upgrading to major release of OSA. This two step process is doubling our effort and time to upgrade to a major version. Is it possible to download and run the major release playbooks soon after the os upgrades ? instead of rerunning the existing playbook again and then download and running next version playbooks ?

2 1

[openstack][magnum] user who created cluster was deleted from keystone
by Nguyễn Hữu Khôi 20 Dec '24

20 Dec '24

Hello. I have a case: What will we do if the user who created the cluster was deleted from the keystone? In this case, we cannot add or remove node groups or use autoscale and autoheal. I tried to create a new application credential and patch my cluster. Then I can add or remove node groups but new node groups cannot join the cluster(nodes not ready) Thank you. Regards Nguyen Huu Khoi

2 2

[cinder] Next 2 Meetings cancelled
by Jon Bernard 19 Dec '24

19 Dec '24

Hello all, the next two weekly meetings (Dec. 25 and Jan 1) are cancelled for the holidays. I hope everyone has a nice break! Catch you at the Jan 8. meeting. Take care, -- Jon

1 0