November 2024 - openstack-discuss

[ironic] PTG Summary and Workitems for 2025.1
by Jay Faulkner 15 Nov '24

15 Nov '24

Hi all, The Ironic team, as usual, has whittled our PTG notes down into a document describing the items we'll be focusing on this cycle. The usual disclaimer applies: this is open source, and we're talking about what we /want/ to do, not making any promises about what /will/ be done. The document can be found at https://specs.openstack.org/openstack/ironic-specs/priorities/2025-1-workit…. I hesitate to include this in the same email; because that ^ document is semi-official and this is not; but I also made a short video describing these priorities if you prefer that medium; https://www.youtube.com/watch?v=oy25mATgbrY Thanks to everyone in the Ironic community for an awesome PTG, and a special thanks to our PTL Riccardo for doing an excellent job organizing and moderating it. Thanks, Jay Faulkner

1 0

[ec2-api][kuryr][murano][sahara][senlin][solum][winstackers] Move 2023.1 Antelope and older branches to End of Life
by Elõd Illés 15 Nov '24

15 Nov '24

Hi, The projects in $SUBJECT are retired projects without any activity or maintenance for some time now. That is why instead of moving their stable/2023.1 (antelope) branch to Unmaintained we plan to directly move them to End of Life. To be able to do that we also have to move older branches to End of Life and delete them too. We don't expect any objection, but if someone wants to step up and bring these projects back to life, then this is the last chance before we move them to End of Life. Thanks, Előd Illés irc: elodilles @ #openstack-release

2 1

[release] Release countdown for week R-19, Nov 18 - 22
by Elõd Illés 15 Nov '24

15 Nov '24

Development Focus ----------------- We are now past the Epoxy-1 milestone. Teams should now be focused on feature development and completion of goals [0]. [0] https://governance.openstack.org/tc/goals/selected/index.html General Information ------------------- Our next milestone in this development cycle will be Epoxy-2, on January 9th, 2025. This milestone is when we freeze the list of deliverables that will be included in the 2025.1 Epoxy final release, so if you plan to introduce new deliverables in this release, please propose a change to add an empty deliverable file in the deliverables/epoxy directory of the openstack/releases repository. Now is also generally a good time to look at bugfixes that were introduced in the master branch that might make sense to be backported and released in a stable release. If you have any question around the OpenStack release process, feel free to ask on this mailing-list or on the #openstack-release channel on IRC. Upcoming Deadlines & Dates -------------------------- Epoxy-2 Milestone: January 9th, 2025 2025.1 Epoxy final release: April 2nd, 2025 Előd Illés irc: elodilles @ #openstack-release

1 0

CLI or Steps to find installed OpenStack version
by Manohar Kundapura (makundap) 15 Nov '24

15 Nov '24

Hi All, Looking for the steps to find out which release/version of OpenStack from the below link, is installed on a Server. https://releases.openstack.org/ Thanks, Manohar

3 2

[horizon] SCSS compilers
by Radomir Dopieralski 15 Nov '24

15 Nov '24

The openstack dashboard uses scss for styling, and so far we have used a python scss library for compiling those files to css for use in the browser. However, that library (pyscss) had in recent years some problems, and is not developed or supported by the authors anymore. So it's time to switch to something with a better support. The django-compressor plugin that we use for handling the compilation of files allows us to use any command-line tool to do the compilation, so we can switch to an scss compiler that is supported and used by the wider community. However, there are several options we can choose from: sassc python-libsass pysassc ruby-sass ruby-sassc node-node-sass Each of those would introduce a non-python dependency on the dashboard. The question for the packagers and distro maintainer is which of those tools would be preferable in terms of support, packaging, and maintenance for the long term? Do you have opinions? Maybe there is another tool that would be even better that I didn't find?

2 3

[neutron] Drivers meeting agenda for Friday, November 15th
by Brian Haley 15 Nov '24

15 Nov '24

Hello Neutrinos: Sorry for the late notice but this is the agenda for tomorrow's drivers meeting: * Configurable Agent Termination on OVS Restart * https://bugs.launchpad.net/neutron/+bug/2086776 * There has been good discussion in this bug from Liu * L3 Agent Readiness Status for HA Routers * https://bugs.launchpad.net/neutron/+bug/2086794 * OVS Agent Synchronization Status for Container Environments * https://bugs.launchpad.net/neutron/+bug/2086799 - related to 2086794 * Add OVN Ganesha Agent to setup path for VM connectivity to NFS-Ganesha * https://bugs.launchpad.net/neutron/+bug/2087541 Unfortunately I will not be able to attend due to a last-minute conflict, so if someone else could lead the meeting I would appreciate it, otherwise we will postpone until next Friday. -Brian

1 0

[watcher] unmaintained branch cleanup
by Sean Mooney 14 Nov '24

14 Nov '24

Hi everyone, As some may now we are currently trying to revive the maintenance of the watcher project, i have volunteered to be the release liaison for the project. At the ptg a few weeks ago we briefly discussed the state of the unmaintained and stable branches watcher currently had 5 unmaintained branches from the oldest unmaintained/victoria <https://opendev.org/openstack/watcher/src/branch/unmaintained/victoria> to the newly created unmaintained/2023.1 <https://opendev.org/openstack/watcher/src/branch/unmaintained/2023.1> https://opendev.org/openstack/watcher/branches since the unmaintained branches have been create, they have not received any meaningful patches, including in some cases the auto generated patches from our openstack patch bots. during the ptg i proposed moving all unmaintained branches older than 2023.1 to end of life. recently https://review.opendev.org/c/openstack/releases/+/935041 has been proposed to "Transition zed-eom branches to EOL" which aligned to my proposed plan for watcher. Per the ptg discussion i committed to reaching out via the mailing list to see if there was any objection to this plan https://etherpad.opendev.org/p/r.79644f8f3cd04686c4f45c10aa35d8ac#L79 please respond here or on https://review.opendev.org/c/openstack/releases/+/935041 if you have an objection to the unmaintained branch transition. regards sean

1 0

[nova] mdev management for vgpu
by Mickael Razzouk 14 Nov '24

14 Nov '24

Hello, I am tying a few things with vGPU on nova and I struggle with the mdev management Nova does. My environment: - Nvidia A2 pGPU - openstack caracal What I did : - installed the nvidia grid host driver on the compute - enabled the virtual function of the pGPU - added in nova.conf the mdev type wanted and the pci address of the virtual functions - restarted nova - created a flavor with the property resources:VGPU=1 (basicaly followed this documentation : https://docs.openstack.org/nova/latest/admin/virtual-gpu.html) the only diference is that I did not put the pGPU pci address in the nova.conf but the address of its virtual functions, so in my case 0000:41:00.4 and not 0000:41:00.0 (the RP were not detected otherwise) From my understanding of this specs sheet (https://specs.openstack.org/openstack/nova-specs/specs/queens/implemented/a…) and some experimentation, When a vgpu is requested by the end user : - Nova looks at the mdevs already running and use them if available. - If there is no mdev available, Nova looks at the RP tree to find mdev capable pci devices that have the specific mdev type available. - If there is such pci device available, Nova create the mdev with a UUID and create the domain XML with this UUID inside. - If no such device is available, Nova returns no host available as an error. My questions are : - Mdev are not persistent across reboot, but when it happens, nova crash at boot because the mdevs are missing, just recreating mdev does not fix the issue as they need to have the same UUID. One could fix the issue by making mdev persistent using a tool like mdevctl, but by creating the mdev the first time I seem to me that Nova tries to manage mdev, is it normal for it to require manual intervention afterward ? Or am I missing something ? - In the nova configuration, is it normal to diverge from the documentation and enter the VF PCI addresses to detect the RF ? Thank you in advance for your help.

4 4

[nova] Spec review day this Thursday 14 Nov 2024
by Sylvain Bauza 14 Nov '24

14 Nov '24

Hey folks, As agreed at the PTG, we'll have a spec review day this week on Thursday. Please mark your calendars for that day as we'll do a first round of spec reviews. Thanks all, -Sylvain

1 1

[OpenInfra Days NA] Recording/Slides from Workshop From zero to code
by Kaster, Jörn 13 Nov '24

13 Nov '24

Hello together, I have seen that on the OpenInfra Days NA there was a Session named 'From zero to code: A hands-on guide to making your first OpenStack code contribution' [1]. Unfortunately I can't attend the OpenInfraDays NA. Now the open question is the following: Is there any Recording/Slides or other Material available? Thank you for any information. [1]https://oidiu2024.sched.com/event/1fH6M/from-zero-to-code-a-hands-on-guid… Mit freundlichen Grüßen / Best Regards i.A. Jörn Kaster Cloud Platform Engineer Note: My work hours may not be your work hours. Please do not feel the need to respond during a time that is not convenient for you. EPG – Ehrhardt Partner Group EPX Ehrhardt + Partner Xtended GmbH Alte Römerstraße 3 56154 Boppard-Buchholz Germany Phone: (+49) 67 42 / 87 27 0 Fax: (+49) 67 42 / 87 27 50 E-Mail: info(a)epg.com Internet: www.epg.com __________________________________________ CEO: Marco Ehrhardt, Markus Derksen Commercial register Koblenz HRB 22546 Registered office: Boppard

2 1

[ironic] [unmaintained] Retire Bifrost UM? Retire Victoria?
by Jay Faulkner 13 Nov '24

13 Nov '24

Hi all, At the PTG, Ironic cores discussed the state of unmaintained branches in Ironic, this email serves as documentation of this conversation as well as a call to action for anyone who uses (some) unmaintained projects. During this discussion, it was very clear that we believe two things are ready for being fully end-of-life'd: - Bifrost; all unmaintained branches. Bifrost is more of a deployment utility than a service; it takes considerable effort to keep it working past the normal maintenance period. It's the consensus of Ironic cores that all current unmaintained branches for bifrost should be closed, and we should avoid opening them in the future. - Victoria: all projects. There are no longer Ironic core reviewers with vested interest in keeping victoria releases of Ironic (and related projects under BM governance) around anymore. This decision is also informed by the significant effort required to get patches tested against victoria during the recent Ironic OSSAs/CVEs. Since traditionally we have handled our own unmaintained branches, we're making a call for anyone who'd like to keep this alive -- including the current unmaintenance team. This is awkward, because if these were branches Ironic fully controlled we would simply end-of-life them. However, since the idea is for these branches to be able to be maintained without the core teams' interaction, we were going to make this call out. If you have interest in keeping maintenance for these branches, please let us know -- otherwise we will begin efforts to end of life them. Thanks, Jay Faulkner

3 2

[election][all][tc] A call for election officials for 2025.2
by Goutham Pacha Ravi 12 Nov '24

12 Nov '24

Hello Stackers, I'd like to express my gratitude to Ian Y Choi (ianychoi) and Slawek Kaplonski (slaweq) for their service as OpenStack's election officials for the 2025.1 term. A hallmark of our community has been its leadership and the guiding principles [1] we use when defining them. In Feb-Mar 2025, we're going to be electing the next set of OpenStack Project Team Leads as well as 5 members of the OpenStack Technical Committee. We've run a combined election for PTLs and TC seats in the past few years. We envision doing the same for this upcoming election. Most of the election process is automated, and election officials help execute and oversee this process [2]. If you're interested in being an election official, please reach out to me. If you've previously served as an election official, please do not hesitate to reach out. If we have enough volunteers, you may do us a great favor by mentoring those doing this for the first time. Thank you, Goutham Pacha Ravi (gouthamr) OpenStack TC Chair [1] https://governance.openstack.org/tc/reference/principles.html [2] https://governance.openstack.org/election/process.html

1 0

[OpenStack-Ansible] Major Upgrade from Yoga to Zed/Antelope
by Roby G 12 Nov '24

12 Nov '24

Hello everyone, Current environment: - Ubuntu 20.04 - OpenStack-Ansible Yoga - no LXC, services are installed bare metal - `install_method` is `distro` My goal is to upgrade this environment to Zed, or skip Zed and upgrade directly to Antelope, since Zed is also in Unmaintained state. What would you recommend? Which order should I follow - upgrade operating system or OpenStack first? The decision is a bit difficult as all 3 releases (Yoga, Zed and Antelope) support different Ubuntu versions in install_method distro and not the same one.

2 1

[kolla-ansible]: guidance needed for neutron setup
by turkey automation 12 Nov '24

12 Nov '24

Hey folks, I am new to Openstack and want to try it on KVM setup with 1 controller and 2 compute VMs. In my first attempt, I was able to install Openstack using kolla-ansible(2024.1 version) and also run-initonce completed without any issues. Now, I want to create instances that can access my back-bone i.e. KVM virtual network(192.168.121.x/24) but either floating IP concept or extending my KVM network to openstack(I don't know if it's feasible!) Some info about my infra: a. Controller 1: Interface eth0 has 192.168.121.5 IP address. eth1 has no IP. b. compute 1: Interface eth0: 192.168.121.10, eth1 has no IP. c. compute 2: interface eth0: 192.168.121.15, eth1 has no IP. What configuration do I need to enable in globals.yml and what all I need to do post deployment of openstack to ensure I have virtual instances on openstack that have IPs from 192.168.121.x network(dhcp can be controlled by KVM network?) and are accessible from other KVM based VMs that I created outside of openstack. I am a learner to please be elaborate if possible. Thanks in advance to the community. - Best regards!

3 6

Upstream Oslo IRC Meetings Are Back.
by Daniel Bengtsson 12 Nov '24

12 Nov '24

Hello everyone, I hope you are all doing well! I sent an email a while ago to reestablish the Upstream Oslo IRC meetings. With the removal of Eventlet, this makes even more sense. I have received two positive responses, and we also mentioned it during the PTG last week. I propose to hold these meetings every other Monday, starting today, at 4:00 PM UTC. Does this work for everyone? Thank you for your feedback! -- Daniel Bengtsson Software engineer

6 18

Oslo Meeting day and agenda.
by Daniel Bengtsson 12 Nov '24

12 Nov '24

Hi everyone, I hope you’re doing well! As a reminder, we’ve chosen the day and time for the Oslo meetings. They will take place every Wednesday with the following schedule: • Even weeks: Europe/US • Odd weeks: Europe/Asia I’ve proposed a patch[1] to update the meeting times. We’ll also have a meeting tomorrow at 13:00 UTC. The topic for tomorrow will be the proposed removal of Eventlet[2] in Oslo service. To help track our meetings, I’ve created an Etherpad document[3], which you can view here: Oslo Etherpad. I used the release structure as inspiration, keeping only the essentials for Oslo. Thanks, and see you tomorrow! [1] https://review.opendev.org/c/opendev/irc-meetings/+/934648 [2] https://review.opendev.org/c/openstack/oslo-specs/+/927503 [3] https://etherpad.opendev.org/p/epoxy-oslo-meeting-tracking -- Daniel Bengtsson Software engineer

1 0

#eventlet-removal - OFTC IRC meeting time slot poll - Express your choice!
by Herve Beraud 12 Nov '24

12 Nov '24

Hey Stackers, We propose to schedule a one hour OFTC IRC meeting (#opentack-eventlet-removal) each first and third week of each month. Let's find together the right UTC time slot for our one hour OFTC IRC meeting. Once the vote is done, meetings will be scheduled each first and third <selected time slot in the poll> of each month. Please, vote here: https://framadate.org/rUghO3j6ZkRhKD7m Votes will be considered as closed by Nov. 8th First meeting will be forced to 4pm UTC on Tuesday Nov 5th and then moved to the selected time slot for all the following meetings. See the proposed time slots in the poll link above. Join us on the #opentack-eventlet-removal OFTC IRC channel. Thanks in advance for your votes -- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/

1 2

OpenStack Cinder - Volume-ID in cinder DB
by Harsh Ailani 12 Nov '24

12 Nov '24

Hello contributors, I am working on clone operation for a volumegroup and in that the volume ID auto-generated by OpenStack is not the one with which the clone volumes are being created on my IBM storage backend. This is as per design that IBM Storage backend will take the source-volume ID in the source-volumegroup and add a suffix to it while cloning. Now since there is a mismatch in the IDs between cinder and storage backend, because of which I am unable to perform any volume operations on this cloned volume. So, I was wondering if I can change the volume ID in cinder DB? I know that it’s a unique key and there is a limit of 36 characters for this field in cinder DB, but since I cannot change the volume in storage backend, I wanted to check if there is a way I can alter the volume ID in cinder DB? Kindly share your thoughts/suggestions on this matter. Thanks and Regards, Harsh Ailani

2 4

[OpenInfra Days NA] Recording/Slides from Workshop From zero to code
by Kaster, Jörn 12 Nov '24

12 Nov '24

Hello together, I have seen that on the OpenInfra Days NA there was a Session named 'From zero to code: A hands-on guide to making your first OpenStack code contribution' [1]. Unfortunately I can't attend the OpenInfraDays NA. Now the open question is the following: Is there any Recording/Slides or other Material available? Thank you for any information. Mit freundlichen Grüßen / Best Regards i.A. Jörn Kaster Cloud Platform Engineer Note: My work hours may not be your work hours. Please do not feel the need to respond during a time that is not convenient for you. EPG – Ehrhardt Partner Group [EPG_Logo_withClaim_rgb_transparent-background.png]<https://www.epg.com/de/> __________________________________________ EPX Ehrhardt + Partner Xtended GmbH Alte Römerstraße 3 56154 Boppard-Buchholz Germany Phone: (+49) 67 42 / 87 27 0 Fax: (+49) 67 42 / 87 27 50 E-Mail: info(a)epg.com Internet: www.epg.com __________________________________________ CEO: Marco Ehrhardt, Markus Derksen Commercial register Koblenz HRB 22546 Registered office: Boppard

1 0

[tc][all] OpenStack Technical Committee Weekly Summary and Meeting Agenda (2025.1/R-20)
by Goutham Pacha Ravi 12 Nov '24

12 Nov '24

Hello Stackers, This week is milestone-1 of the 2025.1 ("Epoxy") release cycle [1]. We've been hard at work implementing the action items from the recent Virtual Project Teams Gathering (PTG) [2]. This past week, we didn’t pursue any new governance changes, though several remain in draft as follow-ups from the PTG. We’re also excited about the extended deadline for the Call for Proposals for the OpenInfra Days event at SCaLE 2025. The new deadline is Nov 15, 2024 [3]. === Weekly Meeting === Last week, the Technical Committee met concurrently on video [4] and IRC [5]. Allison Price (aprice) and Jimmy McArthur (jimmymcarthur) from the OpenInfra VMware Migration Working Group [6] joined us to discuss the group's charter and activities so far. The working group’s goal is not only to address gaps and workarounds but also to influence the future development of OpenStack software. The group has published a whitepaper [7] for operators planning to migrate cloud workloads. The TC expressed interest in collaboration between the working group and the current maintainers of the VMware hypervisor driver in OpenStack Nova. We also highlighted low contributor numbers in several critical project teams mentioned in the whitepaper, including those maintaining OpenStack Compute High Availability (Masakari), OpenStack Storage Backup and Restore (Freezer), OpenStack Resource Optimization (Watcher), and OpenStack Database Service (Trove). The working group plans to hold regular meetings soon and will share details via this mailing list. The TC reviewed the maintenance status of the OpenStack documentation styling extension for Python Sphinx, openstackdocstheme. This library is outdated, as it depends on older versions, but there aren’t enough maintainers to review changes, which could impact documentation across all projects. Ideally, maintainers of this library should have expertise in Sphinx and JavaScript, making the TC a less-than-ideal group to support it. We encourage migrating documentation away from this custom tool, as Sphinx styling extensions have advanced and may now suffice without heavy customization. While there’s no immediate migration plan, we’re looking for volunteers to lead this effort. In the meantime, we’ll raise concerns with the Oslo project team about the lack of reviews for openstackdocstheme changes. The next meeting of the OpenStack Technical Committee is today (2024-11-12) on OFTC's #openstack-tc channel. You can find the agenda on the Meeting wiki [8]. I hope you'll be able to join us. === Governance Proposals === ==== Open for Review ==== Propose to select the eventlet-removal community goal | https://review.opendev.org/c/openstack/governance/+/931254 Add Cinder Huawei charm | https://review.opendev.org/c/openstack/governance/+/867588 Thank you very much for reading! On behalf of the OpenStack TC, Goutham Pacha Ravi (gouthamr) OpenStack TC Chair [1] 2025.1 "Epoxy" Release Schedule: https://releases.openstack.org/epoxy/schedule.html [2] Superuser compilation of PTG summaries: https://superuser.openinfra.dev/articles/october-2024-openinfra-ptg-summary/ [3] OpenInfra Days CFP: https://lists.openinfra.dev/archives/list/foundation@lists.openinfra.dev/me… [4] TC Meeting Video Recording, 2024-11-05: https://youtu.be/o_w3OutGEfM [5] TC Meeting IRC Log 2024-11-05: https://meetings.opendev.org/meetings/tc/2024/tc.2024-11-05-18.04.log.html [6] OpenStack VMware Migration Group: https://www.openstack.org/vmware-migration-to-openstack/ [7] VMware Migration to OpenStack White Paper: https://www.openstack.org/vmware-migration-to-openstack-white-paper [8] TC Meeting Agenda, 2024-11-12: https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting

1 0

[openstack-ansible] Victoria - enable neutron plugin Subnet_dns_publish_fixed_ip
by Rambo Rambo 11 Nov '24

11 Nov '24

Hi Team I am running openstack-ansible Victoria release and would need to activate automatic dns records creation when a new instance is created. So, I am looking for the way to enable neutron plugin Subnet_dns_publish_fixed_ip. After which I can enable dns_publish_fixed_ip in the subnet properties Regards Rambo

2 3

[telemetry] Event Pipeline Config and Use Cases
by Fernandes Figueiredo, Alex 11 Nov '24

11 Nov '24

Hello Openstack Telemetry Team, I'm currently working on re-enabling the Openstack Telemetry services for the StarlingX Openstack (STX-O) project. Openstack telemetry was integrated into STX-O long time ago, but it has been disabled for the past few years. Since the last time telemetry was used by STX-O, it seems that important updates happened, including: * Ceilometer API deprecation * Ceilometer Collector deprecation * Panko (event storage) deprecation My questions are mainly related to the last one (Panko deprecation): * Since Panko is deprecated and there is no actual replacement for event storage [1], is there any other possible telemetry use case to justify keeping the event pipeline enabled and generating data? * Are there any aodh alarming mechanisms that relies on events generated by Ceilometer event pipelines? If so, which "sinks.publishers" should be used to ensure the integration between the Ceilometer event pipeline and the aodh alarm service? * If event pipeline is enabled, what are the publishers currently supported for events? * The current event pipeline configuration (event_pipeline.yaml) in STX-O defines the "sinks.publishers" as "gnocchi://". However, based on Ceilometer doc, it looks like Gnocchi only supports samples/meters, it doesn't support events. Is my interpretation of the documentation correct or does Gnocchi support events and can be really used as a publisher for the event pipeline? [1] OpenInfra summit Vancouver 23 - Openstack Observability revamped<https://youtu.be/AhQJEBR5BUg?si=JbBuLG2iaaDZW7Iu&t=1590> Thank you in advance for any possible help! I'm looking forward to getting Openstack telemetry up and running on StarlingX Openstack. Best regards, Alex Figueirêdo.

3 4

[nova][neutron] Persistent PCI topology for network interfaces
by jan＠uponu.com 11 Nov '24

11 Nov '24

Hi, we have come came across the issue that some appliances rely on the PCI topology to assign network interfaces and not the mac address. This is the a problem, because the PCI topology changes, when an instance (the libvirt config) is recreated. Is there a way to make the network interface order persistent in the PCI topology and if not can someone give me a hint on the steps that would need to be done in nova? I think this could be an important feature in general. Thanks Jan ------ E-Mail: jan(a)uponu.com

2 1

[neutron] Bug deputy report (week starting on November 4)
by Elvira Garcia Ruiz 11 Nov '24

11 Nov '24

Hi folks! Here you have the list of bugs reported for Neutron during last week (04 Nov - 10 Nov) Many new RFEs this week. High ------ - https://bugs.launchpad.net/neutron/+bug/2086740 Deadlock when metadata agent starts Fix by Jakub https://review.opendev.org/c/openstack/neutron/+/934188 [Merged] - https://bugs.launchpad.net/neutron/+bug/2087824 random router related test failures since wsgi switch Unassigned Medium ----------- - https://bugs.launchpad.net/neutron/+bug/2087822 When security group rules have the same normalized CIDR, neutron-ovn-db-sync-util has a wrong log message. Unassigned - https://bugs.launchpad.net/neutron/+bug/2086787 Neutron fails with postgres on agents Unassigned Low ------ - https://bugs.launchpad.net/neutron/+bug/208678 ovn_db_revision_check is not required anymore Patch proposed by Sharat Sharma https://review.opendev.org/c/openstack/neutron/+/934241 Wishlist ---------- - https://bugs.launchpad.net/neutron/+bug/2086776 [RFE] Configurable Agent Termination on OVS Restart Unassigned - Reporter has updated the RFE description - https://bugs.launchpad.net/neutron/+bug/2086794 [RFE] L3 Agent Readiness Status for HA Routers Unassigned - https://bugs.launchpad.net/neutron/+bug/2086799 [RFE] OVS Agent Synchronization Status for Container Environments Unassigned - https://bugs.launchpad.net/neutron/+bug/2087541 [RFE] Add OVN Ganesha Agent to setup path for VM connectivity to NFS-Ganesha Assigned to Amir Nikpour Best regards, Elvira

1 0

[openstacksdk] Flakey unit test makes assumptions about the speed of API results
by Michael Still 09 Nov '24

09 Nov '24

So I've spent a day or so debugging an occasionally flakey unit test that I see in the py3 test suite for openstacksdk. I'm at the point that I want to check that I actually understand what it's doing before proposing a fix. Specifically, openstack.tests.unit.test_stats.TestStats.test_servers_no_detail periodically fails to find a statsd value: self.assert_reported_stat( 'openstack.api.compute.GET.servers.200', value='0', kind='ms' ) This assertion is effectively saying that a GET to the servers compute API should return a HTTP 200, and take 0 milliseconds to execute. The unit test is failing because sometimes it doesn't take 0 milliseconds. In my debugging for example, it takes 4 milliseconds in my environment every now and then: Considering key openstack.api.compute.GET.servers.200, value 4.000000|ms Value 4.000000, kind ms Value 0 != 4.000000 Am I understanding this correctly? It seems to me that perhaps we should be asserting the call took less than a threshold (5 milliseconds) instead of just insisting it always be instantaneous? Michael

1 0