September 2022 - openstack-discuss

[oslo][tooz] Feature Freeze exception request
by Alan Bishop 07 Sep '22

07 Sep '22

I'd like to request an exception for https://review.opendev.org/c/openstack/releases/+/856128. <https://review.opendev.org/710539> This patch proposes a new tooz release for the purpose of picking up [1], which fixes a bug [2] (it's not a feature, per se) that otherwise prevents tooz's etcd3gw driver from working with recent versions of etcd. The issue was initially observed back in 2020 [3]. [1] https://review.opendev.org/c/openstack/tooz/+/827492 [2] https://bugs.launchpad.net/python-tooz/+bug/1983668 [3] https://github.com/dims/etcd3-gateway/issues/41#issuecomment-673558108 Thank you for your consideration, Alan Bishop (abishop)

2 1

[election][Glance] PTL Candidacy for Antelope
by Pranali Deore 07 Sep '22

07 Sep '22

Hello everyone, I would like to let you know that I will be PTL of Glance project for the Antelope cycle. I have been working closely with the previous PTL, Abhishek, since last cycle and I am hoping to contribute back to the community as a PTL while following his excellent example. The following will be my goals for Antelope cycle: 1. Bridge gap between python-glanceclient and openstackclient The CLI support for all glance APIS is already there in GlanceClient and the similar CLI support we need to have in OpenstackClient for the missing Glance APIs in this upcoming cycle. 2. Default Glance to configure multiple stores Glance has deprecated single stores configuration since Stein cycle and will remove the support of the same in the upcoming cycle. 3. Secure RBAC In Xena we have managed to move all policy checks to API layer and implemented project scope of metadef APIs. As per the revised community goal, in zed cycle we have made all policy rules set scope_types=['project'] for all project resources that have already implemented scope. Now we need to have manager role support once it's fully implemented in Keystone. 4. Community priorities Whatever community goals put forward by TC, as a PTL of Glance project will want to adopt these as a priority for the Antelope cycle. 5. Other As we are very short team at this moment, I would like to put some efforts in attracting some contributors, help them to understand how Glance functions and encourage them to contribute for Glance. Looking at how Abhishek, Erno and Brian have steadied the ship, I know it's a big task and with their help I will do my best not to fail the expectations. Thank you for consideration, Pranali Deore (pdeore)

4 3

[cinder] Bug Report - 08-31-2022 - 09-07-2022
by Sofia Enriquez 07 Sep '22

07 Sep '22

This is a bug report from 08-31-2022 to 09-07-2022. Agenda: https://etherpad.opendev.org/p/cinder-bug-squad-meeting ----------------------------------------------------------------------------------------- Low - https://bugs.launchpad.net/cinder/+bug/1988317 "Remove IET iSCSI target." Unassigned. Wishlist - https://bugs.launchpad.net/cinder/+bug/1988803 "With image-volume cache enabled backend, volume creation for a smaller volume than image-volume cache has the same size of the image-volume cache." Unassigned. Cheers, Sofia -- Sofía Enriquez she/her Software Engineer Red Hat PnT <https://www.redhat.com> IRC: @enriquetaso @RedHat <https://twitter.com/redhat> Red Hat <https://www.linkedin.com/company/red-hat> Red Hat <https://www.facebook.com/RedHatInc> <https://www.redhat.com>

1 0

[DIB][diskimage-builder] Rocky Linux image build method
by Pierre Riteau 07 Sep '22

07 Sep '22

Hello, I am curious about the choice of providing only a rocky-container element in DIB, which works differently to the centos element, which uses cloud images. It makes it hard to produce working images for VMs or bare metal, as various packages that would normally be installed are missing, such as cloud-utils-growpart or openssh-server. See the kickstarts for reference [1] [2]. It seems to also occasionally cause complex failures such as the one that rendered Rocky Linux 8 images unbootable last week [3]. I am guessing this wouldn't have happened had the build been from a cloud image. Would the DIB community be open to also have a rocky element using GenericCloud images, like centos? Thanks, Pierre Riteau (priteau) [1] https://git.rockylinux.org/rocky/kickstarts/-/blob/r8/Rocky-8-Container-Bas… [2] https://git.rockylinux.org/rocky/kickstarts/-/blob/r8/Rocky-8-GenericCloud.… [3] https://review.opendev.org/c/openstack/diskimage-builder/+/851687

6 6

[election][Requirements] PTL candidacy
by Matthew Thode 06 Sep '22

06 Sep '22

I would like to announce my candidacy for PTL of the Requirements project. The project was recently updated to allow multiple PTLs this cycle and I imagine tony will be my better half. 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. 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. An image with all the python versions we want is likely needed. I look forward to continue working with you in this cycle, as your PTL or not. Thanks for your time, -- Matthew Thode

3 5

[election][Ironic] PTL Candidacy
by Jay Faulkner 06 Sep '22

06 Sep '22

Hi all, I wanted to let you know I'm running for PTL of Ironic for the Antelope cycle. You can read my candidate statement[1] for full information, but feel free to ask me any questions you'd like here if you have them. Thanks, Jay Faulkner [1] - https://governance.openstack.org/election/#antelope-ptl-candidates

1 0

[horizon] instance filtering dropdown default
by John Ratliff 06 Sep '22

06 Sep '22

When I load certain pages in horizon, like the compute instances page, the default filter is "Instance ID". How can I change the default to "Instance Name" instead? Thanks. --John Ratliff

1 0

Horizon
by Atharva Dhanwate 06 Sep '22

06 Sep '22

4 3

[neutron] Bug deputy report for week of August 29th
by Brian Haley 06 Sep '22

06 Sep '22

Hi, I was Neutron bug deputy last week. Below is a short summary about the reported bugs. BTW, you don't have to wait for me to be deputy to file all your bugs :) -Brian Critical bugs ------------- * https://bugs.launchpad.net/neutron/+bug/1988026 - Neutron should not create security group with project==None - Talked to Security Response Team and they decided to add a tracker to it since the quota system doesn't seem to limit how many random default security groups one can create. - https://review.opendev.org/c/openstack/neutron/+/855580 High bugs --------- * https://bugs.launchpad.net/neutron/+bug/1988037 - Neutron does not find an existing route - https://review.opendev.org/c/openstack/neutron/+/854986 * https://bugs.launchpad.net/neutron/+bug/1988296 - neutron-lib pep8 CI failing with pylint==2.15.0 - https://review.opendev.org/c/openstack/neutron-lib/+/855061 Medium bugs ----------- * https://bugs.launchpad.net/neutron/+bug/1988069 - neutron-dhcp-agent fails when small tenant network mtu is set - Could not reproduce issue reported with MTU as 70 for network, only saw that instances could not do DHCP with it so small. Looks like a potential kernel bug since this works on later versions. * https://bugs.launchpad.net/neutron/+bug/1988199 - https://review.opendev.org/c/openstack/neutron/+/855257 * https://bugs.launchpad.net/neutron/+bug/1988220 - large unused dependency ncclient + paramiko in os-ken - Assigned to Rodolfo since he made change to remove nose https://review.opendev.org/c/openstack/os-ken/+/850631 - closed (won't-fix) as ncclient is required for os-ken, just wasn't explicitly called-out in requirements. * https://bugs.launchpad.net/neutron/+bug/1988549 - Trunk status is down after a live-migration - Arnau Verdaguer took ownership Low bugs -------- * https://bugs.launchpad.net/neutron/+bug/1988298 - [neutron-lib] "registry.has_resource_extenders" is the same as "resource_extend.has_resource_extenders" - Needs owner * https://bugs.launchpad.net/neutron/+bug/1586731 - restart neutron ovs agent will leave the fanout queue behind - Old bug re-opened - Partial patch posted, asked commenter to submit to master Misc bugs --------- * https://bugs.launchpad.net/neutron/+bug/1988281 - neutron dhcp agent state not consistent with real status - After a restart, the agent status is UP even though it is still doing a full-sync operation, is there a way to signify this status? See bug for further discussion. * https://bugs.launchpad.net/neutron/+bug/1988382 - L3 agent(agent_mode=dvr_snat) restart, fip namespace removed rfp-port, resulting in fip not connecting - Failure is seen in code from 2019, asked submitter to try tip of stable/victoria and reproduce since there have been fixed in this area since then. Wishlist bugs ------------- * https://bugs.launchpad.net/neutron/+bug/1988077 - Noisy neutron-openvswitch-agent service - There are some deployments that wanted more logging which is why the agent is more verbose. Will have to have a further discussion to see if there is a way to address concern of log being noisy, besides using more aggressive log rotation as suggested. * https://bugs.launchpad.net/neutron/+bug/1988188 - Create an automated tool to retrieve the needed OVS version depending on the OVN one - Suggested by Rodolfo to help improve gate dependencies. * https://bugs.launchpad.net/neutron/+bug/1988542 - [RFE] OVN support for vnic type virtio-forwarder - Will need to add to Drivers meeting agenda

1 0

Two containers fail
by Derek O keeffe 06 Sep '22

06 Sep '22

Hi, I recently successfully installed OSA and I am now trying to install it again from scratch with all the configuration changes I needed in place. Unfortunately this time around it is failing on setup_hosts.yml here: TASK [lxc_container_create : Create container (dir)] ******************************************************************ok: [infra1_memcached_container-b92f4db7 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_rabbit_mq_container-d286a482 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_utility_container-05f756fc -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_galera_container-5beef059 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_aodh_container-639f5ad8 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_neutron_server_container-f67e972d -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_horizon_container-dfeaa0dc -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_placement_container-350e4ede -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_heat_api_container-82cbd764 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_nova_api_container-b30c38d1 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_repo_container-f16e50a9 -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_ceilometer_central_container-910a888f -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_keystone_container-32dcd68b -> infra1(CONTROLLER_IP_ADDRESS)]ok: [infra1_gnocchi_container-30ccb451 -> infra1(CONTROLLER_IP_ADDRESS)]fatal: [infra1_cinder_api_container-840a18bb -> infra1(CONTROLLER_IP_ADDRESS)]: FAILED! => {"changed": false, "error": "Failed to start container [ infra1_cinder_api_container-840a18bb ]", "lxc_container": {"init_pid": -1, "interfaces": [], "ips": [], "name": "infra1_cinder_api_container-840a18bb", "state": "stopped"}, "msg": "The container [ infra1_cinder_api_container-840a18bb ] failed to start. Check to lxc is available and that the container is in a functional state.", "rc": 1}fatal: [infra1_glance_container-b50e16a6 -> infra1(CONTROLLER_IP_ADDRESS)]: FAILED! => {"changed": false, "error": "Failed to start container [ infra1_glance_container-b50e16a6 ]", "lxc_container": {"init_pid": -1, "interfaces": [], "ips": [], "name": "infra1_glance_container-b50e16a6", "state": "stopped"}, "msg": "The container [ infra1_glance_container-b50e16a6 ] failed to start. Check to lxc is available and that the container is in a functional state.", "rc": 1} I don't understand how 14 containers are fine (can be attached to and commands run through lxc-attach) and 2 containers are failing. I have a deploy host, controller and two computes all with a fresh install of Ubuntu 20.04. The bootstrap_ansible script ran fine on the deploy host and the setup_hosts playbook ran through fully as after the above it skipped cinder and glance containers every where else. Can anybody please point me in the direction of getting more info or a possible cause. Thanks in advance.

2 2