February 2022 - openstack-discuss

[kolla-ansible][nova]Problem with distribution of instance on servers
by Franck VEDEL 17 Feb '22

17 Feb '22

Hello, I seem to have a problem that I hadn't seen. I have 3 servers for my openstack, built with Kolla-ansible, I'm in Victoria version. I had simply put the 3 servers in the [compute] part of the multinode file, at first it worked, but for some time all the VMs are placed on server 1. The 3 servers are operational, identical. here are 3 screenshots to show it. (on the images, the instances on servers 2 and 3 are present because it worked correctly, but no more instances are created on these servers now) I tried to understand how the instances are distributed on the servers, but in my case, I don't understand why none are assigned to the 2nd and 3rd server. How to find the problem? It should be nova-scheduler . Do you have to do anything special? Go see if a parameter has a bad value? Thanks in advance if you can help me. Franck VEDEL

4 10

[manila][cinder][glance][nova][devstack][qa] Add fmount to devstack-plugin-ceph core
by Goutham Pacha Ravi 16 Feb '22

16 Feb '22

Hello, As you're aware, we're looking to adopt cephadm within devstack-plugin-ceph to align with the ceph community's recommendations on day0-day2 handling of ceph [1] in devstack. Francesco Pantano has worked a ton on the integration of ceph with tripleo. He has been our go-to person for deploying and using the new nfs daemon with manila. In light of this, to make reviews more productive, I would like to propose fpantano(a)redhat.com (fmount on launchpad and oftc) to the devstack-plugin-ceph-core group [2]. He's aware of our review policies, and will take care not to break existing testing. Any objections? Thank you, Goutham Pacha Ravi [1] http://lists.openstack.org/pipermail/openstack-discuss/2022-January/026778.… [2] https://review.opendev.org/admin/groups/8e51d8f39e33d932a40f03060b861db57a3…

5 5

[keystone] anyone using OpenID with Keystone?
by Francois 16 Feb '22

16 Feb '22

Hi Keystone users! I am wondering if anyone has experience with keystone openid integration. Initially I was using Keystone LDAP backend (using tripleo KeystoneLDAPDomainEnable and KeystoneLDAPBackendConfigs parameters) and it works! Users are able to log in through Horizon or through the cli, roles can be given per LDAP group, and you can click in Horizon and download a working openrc or clouds.yaml file (minus the root CA that has to be added) to authenticate with the cli (and your password ends as an OS_PASSWORD variable in your environment). I am now trying the Keystone Openid backend (using the enable-federation-openidc.yaml provided by tripleo - https://github.com/openstack/tripleo-heat-templates/blob/master/environment…) with a mapping like this: [{"local":[{"user":{"name":"{0}"},"group":{"domain":{"name":"Default"},"name":"federated_users"}}],"remote":[{"type":"HTTP_OIDC_EMAIL"}]}] The SSO works superb with Horizon, however - logging with the cli seems impractical. I see some doc here: https://docs.ukcloud.com/articles/openstack/ostack-how-use-api-sso.html where you need to provide a secret, I am skeptical I want to do that. The openrc file downloaded from Horizon is not usable as is and needs some tuning. And there is no SSO, and the password still ends up in the environment... - I don't see how I can grant roles to groups anymore. It seems I need an extra mechanism to grant permissions (as I used to do that using LDAP groups). I am wondering if anyone is willing to share their experience dealing with Keystone and OpenID. Thanks! Francois (frigo)

5 5

Openstack HPC Infiniband question
by Satish Patel 16 Feb '22

16 Feb '22

Hi all, I am playing with HPC on openstack cloud deployment and I have a Mellanox infiniband nic card. I have a couple of deployment questions regarding the infiniband network. I am new to ib so excuse me if i ask noob questions. I have configured Mellanox for sriov and created a flavor with the property pci_passthrough:alias='mlx5-sriov-ib:1' to expose VF to my instance. so far so good and i am able to see the ib interface inside my vm and its active. (I am running SM inside infiniband HW switch) root@ib-vm:~# ethtool -i ibs5 driver: mlx5_core[ib_ipoib] version: 5.5-1.0.3 firmware-version: 20.28.1002 (MT_0000000222) expansion-rom-version: bus-info: 0000:00:05.0 supports-statistics: yes supports-test: yes supports-eeprom-access: no supports-register-dump: no supports-priv-flags: yes I didn't configure any ipaddr on the ibs5 interface etc. For testing purposes I have compiled mpirun hello world program to POC my infiniband network between two instances and I am able to successfully run the mpi sample program. Somewhere i read about neutron-mellanox agent to setup IPoIB for segmentation etc but not very sure that how complex it and what are the advantage here over just using simple passthru of SR-IOV Is this the correct way to set up an HPC cluster using openstack or is there a better way to design HPC on openstack?

2 2

[all][elections][tc] TC Election Campaigning Kickoff
by Ghanshyam Mann 16 Feb '22

16 Feb '22

Community Members, The TC Election Campaigning Period has now started [1]. During the next couple days, you are all encouraged to ask the candidates questions about their platforms [2], opinions on OpenStack, community governance, and anything else that will help you to better determine how you will vote. This is the time to raise any issues you wish the future TC to consider, and to evaluate the opinions of the nominees prior to their election. Candidates, Each of you has posted a platform [2], and announced your nomination to the developers. From this point, you are encouraged to ask each other questions about the posted platforms, and begin discussion of any points that you feel are particularly important during the next cycle. Even though there are no runoffs, still it is good to have campaigning discussion. [1] https://governance.openstack.org/election/ [2] https://governance.openstack.org/election/#z-tc-candidates -gmann & The OpenStack Technical Election Officials

1 0

[all][elections][ptl][tc] Combined PTL/TC Zed cycle Election Nominations End
by Ghanshyam Mann 16 Feb '22

16 Feb '22

Hello Everyone, The PTL and TC Nomination period is now over. The official candidate lists for PTLs [0] and TC seats [1] are available on the election website. -- PTL Election Details -- There are 17 projects without candidates, so according to this resolution[2], the TC will have to decide how the following projects will proceed: Adjutant, Freezer, Heat, Magnum, Masakari, Monasca, Murano, OpenStack_Charms, Openstack_Chef, Senlin, Skyline, Solum, Trove, Venus, Watcher, Zaqar, Zun There are 0 projects that will have elections. -- TC Election Details -- The 5 seats were filled and no runoff is required. Now begins the campaigning period where candidates and electorate may debate their statements. Thank you, [0] https://governance.openstack.org/election/#z-ptl-candidates [1] https://governance.openstack.org/election/#z-tc-candidates [2] https://governance.openstack.org/resolutions/20141128-elections-process-for… -gmann & The OpenStack Technical Election Officials

1 0

Create Image in queue State in XENA Version
by Adivya Singh 16 Feb '22

16 Feb '22

Hi Team, We have lately upgraded to Xena , but when users trying to upload any image, it is always in queue state, i did not see any error regarding any issue with glance configuration file. Any thoughts on this Regards Adivya Singh

2 1

[tooz] [mistral] Tenacity >= 8.0.1 issues (in Debian)
by Thomas Goirand 16 Feb '22

16 Feb '22

Hi, It seems Tenacity has some incompatible changes leading to this: https://bugs.debian.org/1005636 (tooz) https://bugs.debian.org/1005467 (mistral) Note that both Manila and Heat were fixed already for this last version of Tenacity, so these would likely be the only remaining blockers to upgrade to 8.0.1. As Tenacity 8.0.1 is already in Debian Unstable, I'd prefer if we fixed these rather than pinning to a lower version. Can anyone help? Cheers, Thomas Goirand (zigo)

2 2

[cinder] final reviews needed for os-brick yoga release
by Brian Rosmaita 16 Feb '22

16 Feb '22

Hello Argonauts, We are now aiming to release on Tuesday 15 Feb at 20:00 UTC. The list of patches needing review and some comments on their status is here: https://etherpad.opendev.org/p/os-brick-yoga-release We're in mostly good shape, just need a final sprint to close this out. reviewers: please review! patch authors: please keep an eye on your patches and respond quickly and make sure your CI has responded with success thanks, brian

1 1

[election][Requirements] PTL candidacy for Zed
by Matthew Thode 15 Feb '22

15 Feb '22

I would like to announce my candidacy for PTL of the Requirements project for the Zed cycle. The following will be my goals for the cycle, in order of importance: 1. The primary goal is to keep a tight rein on global-requirements and upper-constraints updates. (Keep things working well) 2. Un-cap requirements where possible (stuff like prettytable). 3. Fix some automation of generate contstraints (pkg-resources being added and setuptools being dropped) 4. Audit global-requirements and upper-constraints for redundancies. One of the rules we have for new entrants to global-requirements and/or upper-constraints is that they be non-redundant. Keeping that rule in mind, audit the list of requirements for possible redundancies and if possible, reduce the number of requirements we manage. 5. introduce better python version range support into constraints. I look forward to continue working with you in this cycle, as your PTL or not. Thanks for your time, -- Matthew Thode

1 0

[election][swift] PTL candidacy for Zed
by Timothy Burke 15 Feb '22

15 Feb '22

I intend to continue serving as Swift PTL for the Zed cycle. We have seen Swift scale in spectacular ways. Hard drives have grown from 2-4 terabytes when Swift was started to 10-20T. The idea of a "large cluster" has gone from dozens of petabytes across hundreds of nodes to hundreds of petabytes across thousands of nodes. We will continue to see (and address) new challenges in scaling Swift. We anticipate hitting the bounds of our current ring format. We see operators struggle to balance the prioritization of cluster expansion and data durability. We see the need to improve sharding, so that clients are less impacted by what's going on in the backend. All of these have seen progress in the last cycle, and will continue to be advanced in the next. Just as important as scaling the core storage platform, however, is building up the ecosystem of ancillary services that users have come to expect out of their object storage. When a container grows to have billions of objects, it is impractical to perform thousands of listing requests to analyze its contents; instead, users expect a periodic inventory that's easy to pull into their existing data pipelines. Users should be able to examine their own access logs rather than needing to involve operators. Users want to manage their data movement and retention en masse rather than object-by-object. I look forward to seeing the balance we strike between enhancing the core storage platform and building out the data services surrounding it over the coming years. Tim Burke

1 0

[election][Release_Management] PTL candidacy for Z cycle
by Előd Illés 15 Feb '22

15 Feb '22

Hello OpenStack Community, I would like to announce my candidacy for the PTL role of the Release Management team for the Z cycle. I am core member in the Release Management team since Wallaby, but I followed the team's work even before that, and I also have been the PTL of the team for Yoga cycle for some months now. (Though, from release management perspective this is not a long period, since the most laborous part will only start in the coming weeks ;)) Furthermore, as in the past, beside release management duties, my focus is still on stable maintenance as well, since I am stable-maint-core member, too. The next cycle's challenges will be to prepare our tooling to alphabet roll-over after the Z cycle and to adjust to any other changes in our processes. And of course, parallel to this, to release OpenStack Z :) Thanks for your consideration, Előd Illés (elodilles)

1 0

[nova][placement] Implementation review day on Thursday Fev 17
by Sylvain Bauza 15 Feb '22

15 Feb '22

Hey folks, Last minute item, but we agreed during today's meeting to run a feature review day on Thursday February the 17th. If you're one of the owners of the blueprints that are marked in [1], you're eligible to ensure that this etherpad is up to date based on your last efforts and that you're somehow around either on IRC or asynchronously through Gerrit on that day to answer comments and questions about your patches. Cores, I'd appreciate if you can dedicate this Thursday to feature reviews as we still have a large set of blueprints that aren't merged yet. Thanks, -Sylvain [1] https://etherpad.opendev.org/p/nova-yoga-blueprint-status

1 0

[all][tc] OpenStack next release name is final- "OpenStack Zed"
by Ghanshyam Mann 15 Feb '22

15 Feb '22

Hello Everyone, We have the final name selected for the OpenStack next release name. Its "OpenStack Zed". As per the release name criteria[1], selection of release name has gone through three steps: Step1: Name nomination from community members[2]. Step2: TC voted on the nominated names[3]. Step3: Trademark checks by Foundation on winners from TC votes. In trademark check, “Zed” is finalized as the best bet for this release. Thanks community members for participating in name nomination, TC to vote and Foundation for performing the trademark checks. [1] https://governance.openstack.org/tc/reference/release-naming.html#release-n… [2] https://wiki.openstack.org/wiki/Release_Naming/Z_Proposals [3] https://civs1.civs.us/cgi-bin/results.pl?id=E_693bd8bbe63ca52f -gmann

4 6

[election][Kuryr] PTL candidacy for the Z cycle
by Maysa De Macedo Souza 15 Feb '22

15 Feb '22

Hello OpenStack community, I would like to announce my candidacy for the Z cycle. I have been serving as Kuryr PTL since the Wallaby and I would like to continue doing that. In the Yoga cycle we have achieved the following: * Extended features and improvements: addition of events to Kubernetes resources, reduction on the workload Kuryr puts on Neutron, limiting the number of Neutron Ports creation happening in parallel, continuation on the load-balancer resources reconciliation with k8s Services, and many more. * CI improvements: Made Devstack installations to use CRI-O and improved some tempest tests. For the next cycle, I propose the following goals to Kuryr: * Continue improvements for Kuryr scalability with Neutron, OpenStack resources reconciliation with Kubernetes and Kuryr controller health checks. Thanks for your consideration. Best regards, Maysa Macedo. IRC: maysams

1 0

[election][tc] TC candidacy for Z release
by Slawek Kaplonski 15 Feb '22

15 Feb '22

Hi, I want to propose my candidacy for the Technical Committee. I have been an OpenStack contributor since around 2015. For most of that time I have contributed to the Neutron project where I am core reviewer and member of the Neutron drivers team, which is the subset of the Neutron core team for steering the Neutron project. I also served as the Neutron PTL for 4 cycles. Currently I work for Red Hat as software engineer and OpenStack Neutron developer and I would like to bring that developer’s perspective to the TC. Before I joined Red Hat, I worked in the OVH which is a Public Cloud provider. That gave me experience as an OpenStack operator as well. I have experience with the Neutron CI and I would like to expand it wider, for OpenStack as a whole. In the past I cooperated with TC members to reduce CI resources usage by the Neutron project and we made great improvements there. I understand that this new role will require a lot of learning and I will be very happy to have that opportunity. I’m willing to learn and contribute to the TC as much as possible. -- Slawek Kaplonski Principal Software Engineer Red Hat

1 0

[election][tc] TC candidacy for Z release
by Arne Wiebalck 15 Feb '22

15 Feb '22

Hello Everyone! I would like to declare my candidacy for the Technical Committee. I joined the CERN OpenStack team in 2014, initially to help with the operations of the storage components (Cinder, later Manila), and since 2016 I am looking after the overall operations of our cloud infrastructure service here at CERN. As part of this operational role, I have contributed upstream to several projects [1], recently mostly to Ironic which I help as a core member since about three years. I chair the Bare Metal SIG [2] since its inception, and together with other community members I am trying to establish it as a forum to promote the use of OpenStack for bare metal use cases [3] and to help new users to get started. Presentations at several summits [4] and regular blog posts [5] are examples of how I share the operational challenges and the solutions we apply in our cloud, in the hope these are helpful to other deployers in the same way we profit from them sharing their experiences. The combination of running an in-production OpenStack deployment downstream while interacting with core developers and the overall community upstream is something that I've enjoyed _a lot_ since several years now. I see how real-life input can be essential for upstream projects to move forward, set priorities, or maintain momentum. Equally, interaction with a responsive community where feedback is turned into tangible improvements is a valuable asset for any deployment. This positive experience of being an operator/developer hybrid in the OpenStack community is something I would like to bring to the TC, assisting with the ongoing (but continuous) effort to engage operators and help when global technical issues need to be sorted. I am sure being a member of the TC and helping with its work comes with a learning curve, but I'd be happy to contribute where I can. Thanks for reading and your consideration! Arne Wiebalck IRC (OFTC): arne_wiebalck [1] https://www.stackalytics.io/?metric=commits&release=all&user_id=arne-wiebal… [2] https://etherpad.opendev.org/p/bare-metal-sig [3] https://www.youtube.com/playlist?list=PLKqaoAnDyfgoBFAjUvZGjKXQjogWZBLL_ [4] https://www.openstack.org/videos/search?search=wiebalck [5] https://techblog.web.cern.ch/techblog/

1 0

OpenInfra Live - February 17 at 1500 UTC
by Allison Price 15 Feb '22

15 Feb '22

Hi everyone, This week’s OpenInfra Live episode is brought to you by the OpenInfra Foundation. Organizations like AT&T, CERN, China Mobile, China Telecom, Verizon, Vodafone, and Yahoo have adopted the OpenInfra Standard and run it in production today. Similar to how LAMP stack became the standard for deploying web applications, LOKI helps operators identify successful patterns and combinations of technologies to build production infrastructure. This OpenInfra Live episode will discuss this emerging standard and why organizations have adopted these open source technologies together. Episode: Linux OpenStack Kubernetes Infrastructure: The OpenInfra Standard Date and time: Thursday, February 17 at 9am CT (1500 UTC) You can watch us live on: YouTube: https://www.youtube.com/watch?v=pnYB7eYYiPg <https://www.youtube.com/watch?v=pnYB7eYYiPg> LinkedIn: https://www.linkedin.com/video/event/urn:li:ugcPost:6899060799814418432/ <https://www.linkedin.com/video/event/urn:li:ugcPost:6899060799814418432/> Facebook: https://www.facebook.com/104139126308032/posts/4852599164795314/ <https://www.facebook.com/104139126308032/posts/4852599164795314/> WeChat: recording will be posted on OpenStack WeChat after the live stream Participants: Host: Mark Collier, COO at the OpenInfra Foundation Maria Bracho, Manager Product Management at Red Hat Philip Williams, Product Leader at Canonical Have an idea for a future episode? Share it now at ideas.openinfra.live <https://openinfrafoundation.formstack.com/forms/openinfralive>. Thanks, Allison

1 0

[glance] Zed PTG Planning
by Abhishek Kekane 15 Feb '22

15 Feb '22

Hello Everyone!!! As you are already aware that Zed virtual PTG will be held between April 4th to 8th [1]. Please use below etherpad to add the topics you would like to discuss during PTG. https://etherpad.opendev.org/p/zed-ptg-glance-planning <https://etherpad.opendev.org/p/yoga-ptg-glance-planning> [1] https://openinfra-ptg.eventbrite.com/ Thanks & Best Regards, Abhishek Kekane

1 0

oslo.messaging integration development
by liyang 15 Feb '22

15 Feb '22

Hi everyone, I find Apache Pulsar is very popular in MQ area, especially in clound native area. Is there any body integrate this MQ to oslo.messaging? Said to the truth. RabbitMQ developed in Erlang has very little source code level material, Qpid has very little operation practice in the internet. Kafka only support notification in oslo.messaging. I think the Apache Pulsar is clound native based. It may enforece the big scale in extension about Openstack deployment.

5 9

[election][tc] TC candidacy for Z
by Brian Rosmaita 15 Feb '22

15 Feb '22

Hello Everyone, I am declaring my candidacy for the Technical Committee. I've been involved in OpenStack for a while now (my first commit was into Folsom). I've been a very active member of the community as a core member of the Glance, Cinder, and Searchlight teams. I've attended many PTGs and presented at many Summits. I've also served as PTL for Glance (Ocata through Queens) and Cinder (Ussuri through Yoga). You may have seen my clever irc nick, 'rosmaita', mostly in #openstack-cinder and #openstack-glance, but occasionally making a pest of myself in other channels. The TC is tasked with providing technical leadership for OpenStack as a whole and enforcing the OpenStack ideals (the Four Opens [0], the Guiding Principles [1]). While the TC's primary mission is to provide technical leadership, promoting the ideals calls for a broader membership than just technical people actively developing OpenStack. So I think it's appropriate that TC membership includes operators, distributors, and cultural members of the OpenStack community. But it's also important that the TC include highly technical people who contribute to the projects it oversees [2]. I believe that I'm well-suited to represent that constituency. I don't have an agenda other than to represent active contributors and participate in the TC in a manner consistent with its charter [3]. If you ask people who've worked with me, you'll find that I'm good about making sure discussions are done in the open, and that all parties concerned are listened to. I'll bring these qualities to the TC. Finally, I should probably mention who pays my bills. I'm a Principal Software Engineer at Red Hat. As you're hopefully aware, Red Hat has an "upstream first" philosophy [4], which is pretty evident when you look at code contributions to OpenStack, but which also extends to the company encouraging us, when participating in an open source community, to make decisions based on the needs and goals of the community rather than those of Red Hat. (That sounds very noble, but the idea is that a strong community benefits Red Hat in the long run. My key point here, though, is that I'm empowered to represent the OpenStack community's interest, not Red Hat's, if I'm elected to the TC.) Thanks for reading this far, and thank you for considering my candidacy for the OpenStack Technical Committee. - Brian Rosmaita (rosmaita) [0] https://governance.openstack.org/tc/reference/opens.html [1] https://governance.openstack.org/tc/reference/principles.html [2] I appropriated this line from Dan Smith. [3] https://governance.openstack.org/tc/reference/charter.html [4] https://opensource.com/article/16/12/why-red-hat-takes-upstream-first-appro…

1 0

[manila][election][ptl] PTL Candidacy for Z
by Carlos Silva 15 Feb '22

15 Feb '22

Hello, Zorillas and dear stackers, I would like to submit my candidacy to be the PTL of Manila for the Z cycle. I have been a contributor to OpenStack since the Stein release. It has been an awesome experience. During these years, I had the opportunity to contribute with new features, fixes, reviews and other exciting enhancements for Manila. I can tell you that I've faced thrilling challenges, and I learned a lot with the community that made this journey enjoyable day after day. I'm tremendously happy to see the state Manila is at. I'd like to keep the momentum for the project by continuing to focus in some areas we have been focusing, but also start tackling other issues: - Continue increasing the number of contributors and active reviewers through mentoring and engaging them in the community, teaching the OpenStack way and making it collaboratively as we did in the past cycles, promoting hackatons, bug squashes and collaborative review sessions. - Encouraging the involvement of Manila contributors in important programs/events such as GHC, Outreachy and the other opportunities we've been able to contribute in the past. - Pursue the completion of the Manila implementation on the openstackclient/openstacksdk; - Continue pushing Manila to cover the tech debt areas we identified over the past cycles; - Enforcing maintainers to collaboratively cover the lack of documentation we have on third party CI setups for Manila, helping potential new vendors to quickly setup up their CI systems; - Continue discussing with community and making the necessary enhancements for HA and Edge; Thank you for your consideration! Carlos da Silva IRC: carloss

4 3

[election][cloudkitty] PTL candidacy for Zed
by Rafael Weingärtner 15 Feb '22

15 Feb '22

Hello guys, I would like to self-nominate for the role of PTL of CloudKitty for the Zed release cycle. I am the current PTL for the Xena cycle and I am willing to continue in this role. I will keep working with existing contributors to maintain CloudKitty for their use cases and encourage the addition of new functionalities. Thank you for your support, -- Rafael Weingärtner

1 0

[neutron][ci] Agenda for the CI meeting - Tuesday 15.02.2022
by Slawek Kaplonski 15 Feb '22

15 Feb '22

Hi, Agenda for today's CI meeting is at [1]. Please remember that our today's meeting will be video meeting on [2]. See You there at 1500 UTC today. [1] https://etherpad.opendev.org/p/neutron-ci-meetings [2] https://meetpad.opendev.org/neutron-ci-meetings -- Slawek Kaplonski Principal Software Engineer Red Hat

1 0

[election][tacker][ptl] PTL candidacy for Z
by Yasufumi Ogawa 15 Feb '22

15 Feb '22

Hi, I'd like to propose my candidacy for Tacker PTL in Z cycle. I have been a PTL from Victoria and made an effort to implement the latest ETSI-NVF standard features to enable operators to deploy their services. In Yoga release, we are going to provide several container supports for deploying CNFs, multi-version APIs for mixed environment of legacy and cutting edge-products provided by several vendors and more. We also had several collaborative sessions with ETSI NFV for accelerating each activity of standardization and deployment of required features. In Z release, I would like to expand our target to VNFs for network acceleration other than proceed to implement VM/container management features which has been introduced in the previous releases. It has became one of the key requirements of the recent commercial use cases such as 5GC. Best regards, Yasufumi (yasufum)

1 0