March 2022 - openstack-discuss - lists.openstack.org

[kolla-ansible] OVS br-ex configuration for OVN SNAT gateway along with HAProxy
by Tony Liu 26 Mar '22

26 Mar '22

Hi, My goal is to have OVN SNAT gateway (with OVS bridge) and HAProxy co-exist. Here is the networking configuration (partial) before deployment. eth1, no address, neutron_external_interface eth1.40, external address, kolla_external_vip_interface After deployment, I'd like to see this. br-ex eth1 eth1.40 tag 40 eth1.40 with external address But actually, only eth1 is put on br-ex by Kolla Ansible. Need eth1.40 on br-ex, because the external network is also one of the provider networks. Need eth1 on br-ex, to support other provider networks (different VLANs). It works fine when deploy with TripleO with os-net-config. Is it supported by Kolla Ansible? It seems Kolla Ansible only puts one interface (neutron_external_interface) on br-ex? I know a few workarounds in case that's not supported, just want to confirm. Thanks! Tony

2 2

[neutron] Feedback on slides (driver's meeting topic)
by Rodolfo Alonso Hernandez 26 Mar '22

26 Mar '22

Hello Neutrinos (and Lajos on this topic): I would also add: - OVN BW scheduling aware (in progress during the last two cycles but almost ready). - OVN QoS GW IP (actually the patch was merged, so for sure we'll have this feature that closes the gap between OVN and OVS). Regards.

3 2

[all][tc] What's happening in Technical Committee: summary Mar 25th, 21: Reading: 10 min
by Ghanshyam Mann 25 Mar '22

25 Mar '22

Hello Everyone, Here is this week's summary of the Technical Committee activities. 1. TC Meetings: ============ * We had this week's meeting on Mar 24th. Most of the meeting discussions are summarized below (Completed or in-progress activities section). Meeting full logs are available @https://meetings.opendev.org/meetings/tc/2022/tc.2022-03-24-15.00.log.html * Due to OpenInfra episode and PTG, we have cancelled the next two meetings[1] and will be meeting on April 14th Thursday 15:00 UTC, feel free to add the topic on the agenda[2] by April 13th. 2. What we completed this week: ========================= * None for this week. 3. Activities In progress: ================== TC Tracker for Yoga cycle ------------------------------ * This etherpad includes the Yoga cycle targets/working items for TC[3]. 4 out of 9 items are completed and many are in progress. Open Reviews ----------------- * Five open reviews for ongoing activities[4]. Consistent and Secure Default RBAC -------------------------------------------- I sent the updates for Yoga timeline target on ML[5] and created the etherpad to track this goal for all the milestone[6]. The quick summary is, along with keystone, Nova has completed the work and neutron is in progress. Community-wide goals readiness checklist -------------------------------------------------- As we received the feedback in Yoga PTG TC+PTL interaction session, I have proposed the checklist which can help us to check if the goal is ready to start or not[7], please review and let us know if more items need to be added. Zed cycle Leaderless projects ---------------------------------- No updates on this. I am still waiting for Zaqar PTL +1 on the PTL appointment patch[8] and we will discuss Adjutant in PTG, hoping Braden will be ready with their company side permission[9]. PTG Preparation -------------------- Nothing much to share on this but please continue adding the topics in etherpad[10]. Dropping tags framework ------------------------------ Along with deprecation tag removal[11], we are removing the complete framework now[12]. Fixing Zuul config error ---------------------------- Requesting projects with zuul config error to look into those and fix them which should not take much time[13]. Project updates ------------------- * None for this week. 4. How to contact the TC: ==================== If you would like to discuss or give feedback to TC, you can reach out to us in multiple ways: 1. Email: you can send the email with tag [tc] on openstack-discuss ML[14]. 2. Weekly meeting: The Technical Committee conduct a weekly meeting every Thursday 15 UTC [15] 3. Ping us using 'tc-members' nickname on #openstack-tc IRC channel. [1] http://lists.openstack.org/pipermail/openstack-discuss/2022-March/027828.ht… [2] https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting [3] https://etherpad.opendev.org/p/tc-yoga-tracker [4] https://review.opendev.org/q/projects:openstack/governance+status:open [5] http://lists.openstack.org/pipermail/openstack-discuss/2022-March/027837.ht… [6] https://etherpad.opendev.org/p/rbac-goal-tracking [7] https://review.opendev.org/c/openstack/governance/+/835102 [8] https://review.opendev.org/c/openstack/governance/+/831123 [9] http://lists.openstack.org/pipermail/openstack-discuss/2022-March/027626.ht… [10] https://etherpad.opendev.org/p/tc-zed-ptg [11] https://review.opendev.org/c/openstack/governance/+/834327 [12] https://review.opendev.org/c/openstack/governance/+/834798/3 [13] https://etherpad.opendev.org/p/zuul-config-error-openstack [14] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-discuss [15] http://eavesdrop.openstack.org/#Technical_Committee_Meeting -gmann

1 0

[magnum] kube_cluster_deploy timing out
by Chris DiLorenzo 25 Mar '22

25 Mar '22

Hello, I'm trying to deploy a kubernetes cluster using CoreOS 33. I have also tried with CoreOS 35/. 1 master, 1 worker. Ths VMs are deploying without issue. However, the stack is timing out on kube_cluster_deploy. It appears the kubernetes API isn't running: [root@tes-001-pkjawhk66slm-master-0 log]# podman ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES a08cbea59a57 docker.io/openstackmagnum/heat-container-agent:wallaby-stable-1 /usr/bin/start-he... 21 minutes ago Up 21 minutes ago heat-container-agent [root@tes-001-pkjawhk66slm-master-0 log]# ss -ln | grep 8080 -> nothing [root@tes-001-pkjawhk66slm-master-0 heat-config-script]# tail -f 8d8fa128-bca4-485e-946f-90732ebb4cf0-tes-001-pkjawhk66slm-kube_cluster_config-bwkfqwjxzk36.log + sleep 5 ++ kubectl get --raw=/healthz The connection to the server localhost:8080 was refused - did you specify the right host or port? + '[' ok = '' ']' + sleep 5 ++ kubectl get --raw=/healthz The connection to the server localhost:8080 was refused - did you specify the right host or port? + '[' ok = '' ']' + sleep 5 ++ kubectl get --raw=/healthz Not sure where else to look. Thanks Chris

1 0

[kolla] [rabbitmq] Another RMQ config question
by Albert Braden 25 Mar '22

25 Mar '22

We’re running kolla-ansible Train and implementing durable queues. I was able to make it work by adding these lines to etc/kolla/config/global.conf: [oslo_messaging_rabbit] amqp_durable_queues = True But we want to do it one cluster at a time, so I tried commenting that out and adding line: amqp_durable_queues: "True" to group_vars/<region>/vars.yml but that seems to have no effect. Where can I put this value to enable durable queues per-cluster?

2 1

[tripleo] CI jobs broken due to #1966382
by Chandan Kumar 25 Mar '22

25 Mar '22

Hello, The tripleo-ci jobs are broken due to the following bug. https://bugs.launchpad.net/tripleo/+bug/1966382 - Multiple tripleo jobs are failing because of "error: Multiple top-level packages discovered in a flat-layout" The latest release of setuptools 61.0 caused it and more info is added here: https://github.com/pypa/setuptools/issues/3197 The team is working on fixing it https://review.opendev.org/q/topic:lp%252F1966382. Till then please hold the rechecks and avoid workflowing patches. Thanks, Chandan Kumar

2 1

[all][tc][goals] : Consistent and Secure Default RBAC: Yoga Timeline updates
by Ghanshyam Mann 25 Mar '22

25 Mar '22

Hello Everyone, As you know, "Consistent and Secure Default RBAC" is one of the currently selected community-wide goals and we have divided it into multiple milestones (cycle)[1]. Our first milestone was Yoga release[2] and in this email, I am going to summarize the progress of Yoga timeline targets. With Lance's role change in his company and we finalized the direction in between of Yoga cycle, there is less progress than what we targeted. Gerrit Topic: https://review.opendev.org/q/topic:%2522secure-rbac%2522+(status:open+OR+st… Tracking: https://etherpad.opendev.org/p/rbac-goal-tracking Completed: ========= * Keystone is already implemented the RBAC as per the new direction. * Nova completed the RBAC as per the new direction[3]. In-progress: ========= * Neutron has started the policy update as per the new direction[4]. Yoga Targets and their progress: ======================== 1. Keystone implements a new default role called manager: Spec merged[5], implementation yet to start. 2. Keystone implements a new default role called service: Spec under discussion[6] 3. Keystone enforces scope by default: Not yet done. We can do it at the start of the Zed cycle. 4. Services start implementing Phase 1: Done. The target was to do at least in one of the services and Nova completed it. 5. OpenStack-wide Personas Documentation: Not yet done. [1] https://governance.openstack.org/tc/goals/selected/consistent-and-secure-rb… [2] https://governance.openstack.org/tc/goals/selected/consistent-and-secure-rb… [3] https://blueprints.launchpad.net/nova/+spec/policy-defaults-refresh-2 [4] https://review.opendev.org/c/openstack/neutron/+/821208 [5] https://review.opendev.org/c/openstack/keystone-specs/+/818603 [6] https://review.opendev.org/c/openstack/keystone-specs/+/818616 -gmann

1 0

[release] Release countdown for week R-0, Mar 28-30
by Előd Illés 25 Mar '22

25 Mar '22

Development Focus ----------------- We will be releasing the coordinated OpenStack Yoga release Wednesday next week, on March 30th. Thanks to everyone involved in the Yoga cycle! We are now in pre-release freeze, so no new deliverable will be created until final release, unless a release-critical regression is spotted. Otherwise, teams attending the (virtual) PTG should start to plan what they will be discussing there, by creating and filling team etherpads. You can access the list of PTG etherpads at: http://ptg.opendev.org/etherpads.html General Information ------------------- On release day, the release team will produce final versions of deliverables following the cycle-with-rc release model, by re-tagging the commit used for the last RC. A patch doing just that will be proposed. PTLs and release liaisons should watch for that final release patch from the release team. While not required, we would appreciate having an ack from each team before we approve it on the 30th, so that their approval is included in the metadata that goes onto the signed tag. Upcoming Deadlines & Dates -------------------------- Final Yoga release: March 30th, 2022 PTG: April 4 - 8, 2022 (virtual) OpenInfra Summit: June 7-9, 2022 (Berlin) ElődIllés irc: elodilles

1 0

[requirements][all] We are now branched for stable/yoga
by Matthew Thode 25 Mar '22

25 Mar '22

As the title states, openstack/requirements just branched. This means that master can start moving forward with upper-constraints and global-requirements changes. 1. If you have not branched yet you may start seeing dependency changes for the 'Z' release in your development branch (Yoga if you've not branched it yet). 2. If you have changes going into requirements' devepment branch but want to back port to yoga, remember that you'll have to create a backport patch for yoga (and other previous branches). -- Matthew Thode

1 0

[cinder][drivers][stable-policy] propose adding DataCore driver to train
by Karthik Viswanath 24 Mar '22

24 Mar '22

Hi OpenStack community, I am the product manager for SANsymphony, a software defined SAN storage offering, at DataCore<http://www.datacore.com/>. A little bit of history on the current request. DataCore had a Cinder driver in Queen’s branch few years back. Due to organizational restructuring and change in priorities we had to scale back on our efforts towards OpenStack solutions. Because of this we were unable to be active in the community and as a result we got ousted out of Rocky branch due to lack of CI. Details here<https://review.opendev.org/c/openstack/cinder/+/638029>. Currently we have multiple open opportunities with our European customers requiring us to certify RHOSP Cinder driver for both iSCSI and FC protocols. All our current customers are on RHOSP 16.x and would continue to be on that release for another 12-14 months. As a storage solutions provider, we will have to support our customers on 16.x release. As per OpenStack’s Cinder driver policy, this change is considered as *new* driver request and we are unable to backport the changes directly. Given that this has direct revenue impact for DataCore and is time sensitive, we request the community to approve our request to add Cinder driver patch to 16.x maintenance releases ( mainly for Trains branch). Thanks! Karthik Viswanath Sr Product Manager, DataCore Cell: 9886032686 [https://www.datacoreassets.com/email-signatures/logo-email-signature.png]<https://www.datacore.com/?utm_source=signature&utm_medium=email&utm_content…> DataCore Software · The Authority on Software-Defined Storage 1901 Cypress Creek Road Ste. 200 Ft. Lauderdale, FL 33309 [cid1612899890*image002.png(a)01D83EF4.5833C600]

5 4