August 2022 - openstack-discuss - lists.openstack.org

[openstack-ansible][ceph][yoga] wait for all osd to be up
by Father Vlasie 31 Aug '22

31 Aug '22

Hello everyone, I am running setup-infrastucture.yml. I have followed the ceph production example here: https://docs.openstack.org/openstack-ansible/latest/user/ceph/full-deploy.h… I have set things up so the compute and storage nodes are the same machine (hyperconverged). And the storage devices are devoid of any volumes or partitions. I see the following error: ------ FAILED - RETRYING: [compute3 -> infra1_ceph-mon_container-0d679d8d]: wait for all osd to be up (1 retries left). fatal: [compute3 -> infra1_ceph-mon_container-0d679d8d(192.168.3.145)]: FAILED! => {"attempts": 60, "changed": false, "cmd": ["ceph", "--cluster", "ceph", "osd", "stat", "-f", "json"], "delta": "0:00:00.223291", "end": "2022-08-22 19:36:29.473358", "msg": "", "rc": 0, "start": "2022-08-22 19:36:29.250067", "stderr": "", "stderr_lines": [], "stdout": "\n{\"epoch\":6,\"num_osds\":0,\"num_up_osds\":0,\"osd_up_since\":0,\"num_in_osds\":0,\"osd_in_since\":0,\"num_remapped_pgs\":0}", "stdout_lines": ["", "{\"epoch\":6,\"num_osds\":0,\"num_up_osds\":0,\"osd_up_since\":0,\"num_in_osds\":0,\"osd_in_since\":0,\"num_remapped_pgs\":0}”]} ------ I am not sure where to look to find more information. Any help would be much appreciated! Thank you, FV

2 2

[election][designate] PTL Candidacy for Antelope cycle
by Michael Johnson 31 Aug '22

31 Aug '22

Hello OpenStack community, I would like to announce my candidacy for PTL of Designate for the Antelope cycle. I would like to continue to support the Designate team for the Antelope release. As predicted, we have continued to focus on cleaning up technical debt and improving the test coverage. For a small team, we have accomplished a lot in the Zed release, though we could use more reviewers to help with the backlog of patches in need of review and comments. I expect the code cleanup will continue, but I hope we can also focus more on some of the proposed features that are in flight. Thank you for your support and your consideration for Antelope, Michael Johnson (johnsom)

1 0

[cinder] propose to EOL cinderlib train and ussuri
by Brian Rosmaita 30 Aug '22

30 Aug '22

At last week's cinder project midcycle [0], the team discussed recent fixes to keep the cinderlib CI functional in the oldest stable branches. At this point, stable/train is running the bare minimum of CI to keep the branch open, and stable/ussuri is running only one functional job in excess of the bare minimum. The only changes merged into either branch since each was tagged -em have been non-functional, so we believe that there is no demand for keeping the branches open. Thus, the following patches to EOL cinderlib train and ussuri have been posted: - https://review.opendev.org/c/openstack/releases/+/853534 - https://review.opendev.org/c/openstack/releases/+/853535 This email serves as notice of the intent of the cinder project to EOL the cinderlib train and ussuri branches. If you have comments or concerns, please reply to this email or leave a comment on the appropriate patch. [0] https://etherpad.opendev.org/p/cinder-zed-midcycles

1 1

[blazar][election] PTL candidacy for Antelope cycle
by Pierre Riteau 30 Aug '22

30 Aug '22

Hi, I would like to self-nominate for the role of PTL of Blazar for the Antelope release cycle. I have been PTL since the Stein cycle and I am willing to continue in this role. Though we had a quiet Zed cycle, we still have various improvements in the pipeline and I want to help the community deliver them to make Blazar more useful. Thank you for your support, Pierre Riteau (priteau)

1 0

[election][puppet] PTL Candidacy for Antelope cycle
by Takashi Kajinami 30 Aug '22

30 Aug '22

Hello, I'd like to announce my candidacy for the PTL role in Puppet OpenStack, to continue my PTL role for the Antelope cycle. Over the past two cycles, we've successfully improved feature coverage, platform coverage and simplicity of our modules. I'd like to list up a few items which would be our priorities during the next cycle. * Add Ubuntu 22.04 support This would be the next major change after we've completed implementation of CentOS 9 Stream support. We already adapted to Ruby 3 as part of C9S support so I'm not aware of any huge challenges at this moment. * Complete migration to Puppet 7 Currently our modules still support both Puppet 6 and 7. However once we complete migration to Ubuntu 22.04, we complete migration from Ruby 2.x to 3.x. As Puppet officially supports Ruby 3.x since 7.7, this means we no longer maintain test coverage with Puppet 6. It's time to consider again complete migration to Puppet 7. * Improve scenario/component coverage by CI During Zed cycle we added OVN and Octavia to the integration jobs. We'll review a few remaining modules like Manila and will continue extending the component coverage. * Review unmaintained/unused modules In Puppet OpenStack projects we maintain number of modules to support multiple OpenStack components. However, some modules have not been really active and attracted no interest. In the past few cycles we have retired several modules but I'd like to continue reviewing our modules to consider retiring inactive ones. * Keep each module up to date and simple It's always important that we add support for the new features/parameters timely so that users can leverage the new capability via our modules. Thank you for your consideration. Thank you, Takashi

3 2

[election][cinder] PTL Candidacy for Antelope
by Rajat Dhasmana 30 Aug '22

30 Aug '22

Hello everyone, I would like to announce my candidacy for PTL of the Cinder project for the Antelope cycle. I served as the PTL for Cinder during the Zed development cycle and targeted some of the areas which were lacking attention and worked on improving them like: 1. Review Quality Proposing the Efficient Review Guidelines[1] to help new contributors improve the quality of reviews. 2. Security Vulnerabilities Triaged the open security issues during the first mid cycle[2]. There are still areas which could be improved upon like 3rd Party Compliance checks and Cinder driver documentation which are good candidates of work items for the next development cycle and I look forward to actively work on them. Apart from that, there were deadlines which could have been better handled which I would like to incorporate in my planning skills so we get things merged in time and we won't have to extend deadlines frequently. Overall, I feel it was a good learning experience being PTL for the first time and I'm thankful for all the support I've had from the team, not to mention the advice and help from Brian while making critical decisions, and Hopefully I will take the learnings to be a better PTL this time. [1] https://docs.openstack.org/cinder/latest/contributor/gerrit.html#efficient-… [2] https://etherpad.opendev.org/p/cinder-zed-midcycles#L136 Rajat Dhasmana (whoami-rajat)

1 0

[election][Release_Management] Proposal to change to Distributed PTL model
by Előd Illés 30 Aug '22

30 Aug '22

Hi, This is my 2nd cycle (Yoga + Zed) as PTL of the Release Management team and this is the time when usually a PTL role is handed over. We had a discussion in our team and decided that now that OpenStack moves from Zed to Antelope and the cycles starts again from the beginning of the alphabet + the new naming is going to be introduced (2023.1 Antelope) this is a good time to change the leadership model of the team to a group leadership model (or DPTL; distributed project team leadership). @TC: so this mail is to inform the TC that we want to move to this direction & let us know if there is something we need to sort out for the model change. Thanks, Előd Illés irc: elodilles

2 1

PTG October 2022 Teams List
by Kendall Nelson 30 Aug '22

30 Aug '22

Hello Everyone! The October 2022 Project Teams List is official! Projects + Teams: - Computing Force Network (CFN) Working Group - Diversity& Inclusion Working Group - Environmental Sustainability Working Group - Kata Containers - Magma - OpenInfra Edge Computing Group - OpenStack Teams - - Ansible OpenStack Modules - Barbican - Blazar - Cinder - CloudKitty - Designate - First Contact SIG - Glance - Heat - Horizon - Interop WG - Ironic - Keystone - Kolla - Kuryr - Manila - Neutron - Nova/Placement - Octavia - OpenStack Charms - OpenStack Helm - OpenStack Operators - OpenStack SDK/CLI - OpenStack Security SIG - OpenStack Technical Committee - Openstack-Ansible - OpenStack-Helm - QA - Release Management - Swift - Tacker - Telemetry - TripleO - StarlingX If your team was planning to meet and isn’t in this list, please contact ptg(a)openinfra.dev IMMEDIATELY. Soon I will be contacting moderators to sign up for time via the PTGBot[1] once I have it configured for the teams that are signed up to attend the event. Otherwise, please don't forget to register[2] — its FREE! As usual, feel free to let us know if you have any questions. Thanks! -Kendall (diablo_rojo) [1] PTGBot Docs: https://github.com/openstack/ptgbot#open-infrastructure-ptg-bot [2] PTG Registration: https://openinfra-ptg.eventbrite.com

1 0

[election][Neutron] PTL candidacy for Antelope cycle
by Rodolfo Alonso Hernandez 30 Aug '22

30 Aug '22

Hello all: I would like to announce my candidacy for the role of PTL of Neutron for the Antelope cycle. Let me introduce myself first. I started contributing to OpenStack in Liberty release (2015). I've contributed to several projects, mainly Neutron, neutron-lib, os-ken and os-vif, in addition to Nova and devstack. During the last two years I've been focused on the ML2/OVN integration, QoS and Placement related features and helping on the Neutron CI improvement and stabilization. For this new release, those are the main goals I would like to focus on: * Take care and focus on the approved and merged Neutron specs [1]. Those RFEs should be actively attended by the community, from the spec proposal to the code review. In order to increase the attention of the community on these new RFEs, new ways of tracking them should be proposed (a topic that should be discussed during the PTG). For example, having a core reviewer “godfather” for each RFE. * Continue with the improvement of CI stability. The job done during the last 2 or 3 releases has been impressive, probably the hardest and the most continued effort on the CI in the Neutron community ever. We *must* continue with this effort and the current processes to track the healthiness of the CI. * Start working on the smart NIC / hardware offload testing. The number of backends (ML2/OVS, ML2/OVN, ML2/SR-IOV) make non-viable to test any possible combination. But at least, depending on the available hardware, we would be able to test the stability of those backends with the newest hardware offload NICs. Note: that will imply an external CI support. * Work with users and operators, providing an active channel with them. The goal is to attract customers to be actively involved in the community, participating in the Neutron meetings (team meeting, CI meeting, drivers meeting) or even creating a specific meeting with them in order to capture new needs or issues. * Live migration improvement, specially in ML2/OVN. This feature is still being tested and has not proven to be very stable. There are several core OVN and Neutron efforts right now but we still need to make this feature stable enough to be delivered to customers. A part from those main goals, we should always keep an eye on: * The SQLAlchemy 2.0 migration and any possible issue detected. * The Neutron's stadium ecosystem alive and healthy: this is a permanent goal in any release. The Neutron ecosystem is wide and diverse and it is maintained with few resources. We should focus on those active projects and communities that support their respective repositories. * Still closing the ML2/OVS - ML2/OVN feature gap, that is smaller every cycle. Thank you in advance. Rodolfo Alonso Hernandez (ralonsoh, ralonsoh(a)redhat.com)

1 0

[TripleO] quay.io namespace local mirror using harbor. Quay.io do not list repos/images
by Ruslanas Gžibovskis 30 Aug '22

30 Aug '22

Hi all, I am trying to make a local repo/copy to be able to deploy OSP onsite behind firewall. I have found out from harbor, that Quay.io do not return repos/image anonymously also I have created account in quay.io with my RH account. Later reset quay.io password, so quay.io connection verify works. I can get it as a proxy, or exact image, but not full offline installation. Could you help/suggest any tool that could do a local repo? I have also tried foreman, but it only can sync repo in namespace, not whole namespace. Thank you for your advice and comments. -- Ruslanas Gžibovskis +370 6030 7030

2 1