- openstack-discuss - lists.openstack.org

[OVN][Neurton] Help in capturing metrics part of scale testing in OVN
by keerthi vassan 06 May '25

06 May '25

Hello Team Myself new to OVN with Openstack ( provider network based model) , as part of this I am trying to perform an iperf based network test to understand how traffic flow is happening. As part of that, how to validate how much time took for logical flow computation by services in hypervisor. For specific tests, looking for a way to figure out, how much services like ovn-controller, vswitchd & ovsdb server are contributing to it . Do we have any specific tooling to capture these metrics ? Can you please help with this ? Regards, Keerthivasan

2 1

[neutron] OVS Hardware Offloading Performance Issues
by Serhat Rıfat Demircan 06 May '25

06 May '25

Hello everyone, I've been testing OVS hardware offloaded ports for a while. I have ConnectX-5 and ConnectX-6 cards in my lab. Iperf tests are looking promising, but when I try to use hardware offloaded ports for krbd traffic, after some time, traffic falls back to the software path. Another interesting thing for me is without changing anything, it can also switch back to the hardware path. This issue mostly occurs while I am doing this test with 3 virtual machines on the same hypervisor. I'm not sure if this is a configuration issue or I'm hitting to cards limit. If any of you guys are experienced with this all ideas are more than welcome. Openstack version: Caracal Operation System: Ubuntu 22.04 Thank you, Serhat

1 1

[tc][all] OpenStack Technical Committee Weekly Summary and Meeting Agenda (2025.2/R-21)
by Goutham Pacha Ravi 05 May '25

05 May '25

Hello Stackers, Next week is milestone-1 of the 2025.2 ("Flamingo") release cycle [1]. This is typically a bug-targeting milestone across OpenStack project teams. During the last week, the OpenStack Technical Committee adopted a resolution to extend the purview of the OpenStack Vulnerability Management Team (VMT) to cover all OpenStack project teams [2]. This means that all of the user- and operator-facing deliverables within OpenStack will follow the Vulnerability Management process [3] consistently, and project teams no longer need to explicitly opt-in. To make this work in reality, project teams must ensure that their bug trackers are configured to explicitly share private security bugs with the VMT, and ensure that core security contacts are up to date. I'll share more on this in a dedicated post to this mailing list. Alongside this resolution, the TC also approved the creation of "grain-ui" as an OpenStack project under the telemetry team, and approved the retirement of the "openstack-helm-infra" project because the code is now part of the "openstack-helm" project. As you are aware, the OpenInfra Summit is coming to France this October [4]. The call for papers is open until 2025-06-13. Funding exists through the OpenInfra Foundation's Travel Support Program. If you're interested in applying, please do so by the deadline: 2025-07-09 [5]. === Weekly Meeting === The TC held its weekly meeting on 2025-04-29 in OFTC's #openstack-tc IRC channel [6]. We noted a change in affiliation with regard to Ghanshyam Maan (gmaan), who is now employed at Red Hat. This change does not negatively affect the TC's affiliation diversity requirement [7]. The TC caught up on several prior action items. It was noted that the base upgrade test jobs have been modified to now test upgrading from the 2025.1 ("Epoxy") release to the current release, and this change should have been reflected in all project-specific test jobs. SLURP (N-2→N upgrade) jobs are configured as voting by default but can be opted out of by projects, since the release under development, 2025.2 ("Flamingo"), isn't a SLURP release. We agreed that Sylvain Bauza (bauzas) and myself will serve as the VMT's TC liaisons. Our role will be to bridge communication between the VMT and project teams to coordinate security fixes. Discussing the i18n SIG's challenges with translation infrastructure provided us with a segue into OpenStack Special Interest Groups (SIG) Activity & Maintenance Processes. SIGs do not maintain any OpenStack deliverables; however, several of them maintain code repositories on OpenDev [8] that could be useful. In many cases, there isn't a clear process to onboard new contributors or maintainers, and communication breaks down if existing SIG chairs are unresponsive. We concluded that SIGs should have better documented succession and onboarding paths to avoid stagnation. We would like to do a health check of SIGs over the release and embrace any new ways to make SIGs sustainable. The next meeting of the OpenStack TC is today, 2025-05-06. This will be our monthly Video+IRC meeting. You're welcome to join any medium that you prefer. We'll take notes as usual and upload the meeting's recording to the TC's channel on YouTube [9]. Please find important meeting information such as the venue or the agenda on the meeting's wiki page [10]. I hope you'll be able to join us there. === Governance Proposals === ==== Merged ==== - Retire openstack-helm-infra repo | https://review.opendev.org/c/openstack/governance/+/947611 - Add openstack/grain-ui | https://review.opendev.org/c/openstack/governance/+/948252 - [resolution] Extend scope of VMT to cover all project teams | https://review.opendev.org/c/openstack/governance/+/944817 === Upcoming Events === - 2025-05-06: (Informal) OpenInfra Board discussion on AI policy: https://etherpad.opendev.org/p/board-informal-ai-contribution - 2025-05-13: OpenInfra Board Meeting: https://board.openinfra.org/ - 2025-05-22: OpenInfra UG Meetup, Sweden: https://www.meetup.com/openinfra-user-group-sweden/events/306139678/ - 2025-06-28: OpenInfra+Cloud Native Day, Vietnam: https://www.vietopeninfra.org/void2025 Thank you very much for reading! On behalf of the OpenStack TC, Goutham Pacha Ravi (gouthamr) OpenStack TC Chair [1] 2025.2 "Flamingo" Release Schedule: https://releases.openstack.org/flamingo/schedule.html [2] TC Resolution to extend the VMT's scope: https://governance.openstack.org/tc/resolutions/20250317-extend-scope-VMT-c… [3] VMT process: https://security.openstack.org/vmt-process.html [4] OpenInfra Summit 2025, Paris-Saclay: https://summit2025.openinfra.org/ [5] OpenInfra Summit Travel Support Program Application: https://openinfrafoundation.formstack.com/forms/openinfra_tsp [6] TC Meeting IRC Log 2025-04-29: https://meetings.opendev.org/meetings/tc/2025/tc.2025-04-29-17.00.html [7] OpenStack TC Affiliation Diversity Requirement: https://governance.openstack.org/tc/reference/charter.html#tc-diversity-req… [8] Repositories maintained by OpenStack SIGs: https://governance.openstack.org/tc/reference/sig-repos.html [9] OpenStack TC on YouTube: https://www.youtube.com/@openstack-tc [10] TC Meeting Agenda, 2025-05-06: https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting

1 0

[Cinder] : Group API query
by Harsh Ailani 05 May '25

05 May '25

Hello team, In one of our customer requirements we came across a use-case to move a volume from one group to another directly. Currently user needs to first remove the volume from the current group and add it to the new group. Wanted to know, if there is one direct cinder API or method to directly move the volume from one group to another ? Thanks and Regards, Harsh Ailani

3 9

Student Opportunity - CMU Summer Semester 2025
by Kendall Nelson 05 May '25

05 May '25

Hello All! Are any of you interested in mentoring students to work on any projects you have on the backlog? This is what I have from one of the Profs of the course: ------- To remind you of the ask from last year (and *call out differences for this year*): - We are looking for projects that a team of 4-5 students could tackle together with at least two mentors for each project. As we saw last year, mentors can certainly overlap projects if appropriate and they have the time. (Life happens and having the built-in mentor redundancy helps. I’ve had mentors get laid off, change jobs, and take summer vacation.) - Mentors are expected to meet student teams once a week for an hour (via any video conference setup folks want to use), and to be available by email during the rest of the week to answer any urgent questions. - *This summer we are running the class from 13 May to 31 July (11 weeks)*. - *We want to try teaching concurrently in both campuses Doha, Qatar (GMT+3) and Pittsburgh (GMT-5), USA. The entire course will be taught virtually this year, without a classroom.* I certainly did something similar a few years ago when I was teaching at Hopkins (20 students) with another group in Galway (16 students). The morning class in Pitt will be the afternoon in Doha. - *We likely have 15-20 students in each location, so if you had on the order of 4-5 team projects with mentors that fit the format, like last year, that would be fantastic. * - *We are considering going so far as to choose the teams across time zones to get them working remotely from the start.* Last year, after six weeks together in class and daily stand-ups, the students scattered home away from Doha, and all of them worked remotely the last four weeks. They proved they could work remotely together. Of course, the relationships with mentors have always been remote. The profs in Doha and Pitt want to try remote from the beginning. (I have a few concerns but I’m also always up to experiment on students.) - We post the projects on the first day of class and will organize the teams in that first couple of days, so student teams are introduced to their mentors in the first week of class and expected to organize that first meeting to begin the project learning curve. That’s when mentors point students at any tutorials and bootstrap materials, recommended getting started materials, etc. - We have set the expectations with the students that they will be spending 20-40 hours of time per week on the project. It is an internship-like experience. - *Two co-teachers run classes on three days a week for 50 minutes, and I will guest lecture a collection of classes. *(Last year, there was just the real professor and I.) - The three of us will provide a coaching session with each team to ensure they are working with the mentors well. - Students generally have Windows or Mac laptops, but we have teaching assistants on each site that we can start to prep any other access to resources they might need. - As with last year, mentors have a lot of freedom to experiment. Some have run joint sessions if they are mentoring more students for learning curves. Some have run Slack or Discord channels. ------ So, who is interested in working with students? Please let me know if you have any additional questions. -- Kendall Nelson *Senior Upstream Developer Advocate* The OpenInfra Foundation <https://openinfra.dev/>

1 1

[barbican] Weekly meeting canceled 5-5-2025
by Grzegorz Grasza 05 May '25

05 May '25

Hi all, Due to conflicts and time off, we have to cancel the Barbican weekly meeting today (5-5-2025). Please reach out via IRC if you need anything. Thanks! / Greg

1 0

[kolla][OpenStackAnsible] 2025.1 Epoxy release for cycle-trailing projects
by Elõd Illés 05 May '25

05 May '25

Hi teams, This is a gentle reminder that the following projects with cycle trailing release model haven't done their 2025.1 Epoxy release yet: [kolla] ansible-collection-kolla kayobe kolla-ansible kolla [OpenStackAnsible] openstack-ansible-roles openstack-ansible Note, that next week we'll ask for cutting stable/2025.1 branches, as kind of a Feature Freeze, 3 weeks prior the 2025.1 Epoxy Cycle-Trailing Release Deadline on June 5th, 2025 [1] (exactly one month away). [1] https://releases.openstack.org/flamingo/schedule.html#f-cycle-trail Thanks in advance, Előd Illés irc: elodilles @ #openstack-release

1 0

[tacker] Skip next IRC meeting
by Yasufumi Ogawa 04 May '25

04 May '25

Hi Tacker team, Due to a holiday in Japan and most of us will not join, I would like to skip IRC meeting tomorrow. Thanks, Yasufumi

1 0

[manila] Share Encryption Collab Review meeting on 2025-05-08
by Goutham Pacha Ravi 02 May '25

02 May '25

Hello Zorillas and interested stackers, We're hosting a discussion on meetpad during the weekly Manila Team meeting slot [1][2] next week, 2025-05-08 (Thursday) to further discuss the design of share encryption. The discussion will follow a proposed specification [3], and seeks to align a number of folks that are implementing the API/UX portions, back end storage driver pieces and functional tests. The meetpad discussion will be at 15:30 UTC. We'll have our weekly IRC meeting as usual at 1500 UTC. Thanks, and hope to see you there, Goutham Pacha Ravi (gouthamr) [1] Manila Team Meeting: https://meetings.opendev.org/#Manila_Team_Meeting [2] ICS for this collab review: https://gist.github.com/gouthampacha/0eaaad7bb2eec343dd56cf15f2f87f9a [3] Design Spec for share encryption:

1 0

OVN Routers don't seem to be highly available
by Ryan Belgrave 02 May '25

02 May '25

I have Openstack dalmatian deployed with OVN. All networking seems to work as expected, however OVN routers don't seem to be highly available. I've followed this guide https://docs.openstack.org/neutron/latest/install/ovn/manual_install.html and have enable-chassis-as-gw set on all the compute nodes. I have a very simple network setup with 1 self-service network connected to a router and that router connected to a provider network. Looking in the OVN SB database at the port_binding for the router, it has 1 chassis listed and a ha_chassis_group. There are multiple compute nodes listed in the ha_chassis_group when I lookup that ID in the table. When I fail the server that is listed as the chassis for the port binding it never gets assigned to a different chassis and the instances in the self-service network stop being able to connect to things in the provider network. I've tried various different ways to fail the server, unplugging the self-service nic makes the port_binding chassis show as "[]" and it never gets assigned to another, hard powering off the server keeps the chassis as the old one and it never gets assigned to another. Waiting multiple minutes to 1+ it never gets reassigned. As soon as I bring back the failed compute node, it gets assigned back to that one and everything works. The only way I can seem to get a failover to another compute node is to stop the ovn-controller service on the server that is listed as the chassis for the port binding. In that case another chassis does get assigned and routing works as expected. According to https://docs.openstack.org/neutron/latest/admin/ovn/routing.html under Failover (detected by BFD) this should "just work". I did check the tunnels and BFD status and BFD does show as down as expected, but nothing else happens. Any ideas? I'm not too sure what to look at to debug this, all my configs seem fine as everything else works as expected.

2 2