November 2022 - openstack-discuss

Neutron : Routed Provider Networks (no subnets declared in compute placement API: consequences?)
by Andrew West 09 Nov '22

09 Nov '22

Hi neutron experts Have a cloud built with multiple network segments (RPN) . (kolla-ansible, openstack ussuri), things are running OK (on the network level): networks have DHCP agents ok (*os network agent list --agent-type dhcp --network $networkID* ) all network segments are listed in Host Aggregates BUT if I run through all the existing segments , NONE have a (compute service) inventory declared for each segment IPv4 subnet i.e *os resource provider inventory list $segmentID * returns no output. (official doc on RPN says this should exist) What feature may not function if this inventory is missing ? I don't quite understand what role this IPv4 subnet compute service inventory plays here (during placement? port declaration ?) thanks Andrew

2 1

[largescale-sig] Next meeting: Nov 9, 15utc
by Thierry Carrez 09 Nov '22

09 Nov '22

Hi everyone, The Large Scale SIG will be meeting this Wednesday in #openstack-operators on OFTC IRC, at 15UTC. Note that most countries are no longer in DST! You should doublecheck how that UTC time translates locally at: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20221109T15 Feel free to add topics to the agenda: https://etherpad.opendev.org/p/large-scale-sig-meeting Regards, -- Thierry Carrez

1 1

[oslo] Moving back oslo deliverables to the cycle with intermediary release model
by Herve Beraud 09 Nov '22

09 Nov '22

Hey Osloers, Months ago we moved a couple of oslo deliverables to the independent release model [1][2], however it led us to issues with backports [3]. Backports are not an option for those deliverables. During the previous release management team' PTG we discussed this topic and we decided [4] to move back the oslo deliverables to the cycle-with-intermediary model. You can follow the transition to the CWI model through this patch https://review.opendev.org/c/openstack/releases/+/864095 Do not hesitate to react directly to the patch. Thanks for your time. [1] https://lists.openstack.org/pipermail/openstack-discuss/2020-November/01852… [2] https://opendev.org/openstack/releases/commit/5ecb80c82ed3ab0144c8e5860ee62… [3] https://lists.openstack.org/pipermail/openstack-discuss/2022-September/0306… [4] https://etherpad.opendev.org/p/oct2022-ptg-rel-mgt -- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud

4 5

Need assistance
by Gk Gk 09 Nov '22

09 Nov '22

Hi All, I have a OSA setup. I am trying to trace the control flow of nova-api using pdb in the file "/openstack/venvs/nova-20.2.1/lib/python3.6/site-packages/nova/objects/instance.py". My goal is to trace the flow for "nova list --all" command. I am launching the nova-api service manually from the command line as follows: #/openstack/venvs/uwsgi-20.2.1-python3/bin/uwsgi --ini /etc/uwsgi/nova-api-os-compute.ini --workers 1 I am executing "nova list --all" command in another terminal. I have inserted pdb in instance.py as follows: ---- @base.remotable_classmethod def get_all(cls, context, expected_attrs=None): import pdb; pdb.set_trace() """Returns all instances on all nodes.""" db_instances = db.instance_get_all( context, columns_to_join=_expected_cols(expected_attrs)) return _make_instance_list(context, cls(), db_instances, expected_attrs) --- But when I fire the nova list --all command, I see no pdb prompt being shown in the nova-api window. Can anyone help me how to use the pdb to trace the flow of control for "nova list --all" command ? Thanks Kumar

3 3

[cinder] Bug report from 10-26-2022 to 11-08-2022
by Sofia Enriquez 09 Nov '22

09 Nov '22

This is a bug report from 10-26-2022 to 11-08-2022. Agenda: https://etherpad.opendev.org/p/cinder-bug-squad-meeting ----------------------------------------------------------------------------------------- Low - https://bugs.launchpad.net/cinder/+bug/1995204 "Fail to unify allocation_capacity_gb values among multiple Active-Active Cinder-Volume services ." Unassigned. - https://bugs.launchpad.net/python-cinderclient/+bug/1995883 "cinderclient against wallaby fails to create a snapshot." Fix proposed to master. - https://bugs.launchpad.net/cinder/+bug/1995863 "Failed to create multiple instances with boot volumes at the same time in version 20.0.2.dev11." Unassigned. - https://bugs.launchpad.net/cinder/+bug/1996049 " Backup fails with VolumeNotFound but not set to error." Fix proposed to master. Incomplete - https://bugs.launchpad.net/cinder/+bug/1995838 "Reimage results are stuck in downloading state." Unassigned. - https://bugs.launchpad.net/cinder/+bug/1996039 "Volume State Update Failed After Backup Completed." Assigned to wanghelin. 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

[kolla] Weekly meeting 9th Nov 22 cancelled
by Michał Nasiadka 08 Nov '22

08 Nov '22

Hola Koalas, The weekly meeting today is cancelled (I’m on full day workshops) - let’s talk next week. Michal

1 0

(Openstack-Keystone)Regarding Authentication issue of one user while login to Open Stack using AD password
by Adivya Singh 08 Nov '22

08 Nov '22

Hi Team, There is one issue , where a user is getting " Authenticated Failure" all of a sudden, and this user is the only user who is facing this problem. I tried to disable and enable the project if, Check the logs but do not found anything related to Keystone authentication Delete the Project id and Create it again , Results are same , Any insights what i can do more to fix this issue Regards Adivya Singh

2 8

[all][tc] Technical Committee next weekly meeting on 2022 Nov 9 at 1600 UTC (PLEASE NOTE: MEETING DAY/TIME CHANGED)
by Ghanshyam Mann 08 Nov '22

08 Nov '22

Hello Everyone, The technical Committee's weekly meeting day and time have changed to every Wed, 16 UTC. The next weekly meeting is scheduled for 2022 Nov 9, at 1600 UTC. If you would like to add topics for discussion, please add them to the below wiki page by Wednesday, Nov 8 at 2100 UTC. https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting -gmann

1 1

[kolla] Kolla-ansible plays with --limit failing if a compute host is down.
by Aidan Collins 08 Nov '22

08 Nov '22

Hello, It seems that the kolla-ansible plays reconfigure, prechecks, bootstrap-servers and deploy all fail when using limit if any compute host is down, even if it is not the one being specified by limit. Is there any way to configure gather-facts in these plays to not fail if this is the case? Due to the size of our plant we sometimes need to take down a compute host for maintenance and still provisiion new ones. We are using Victoria. Thanks a lot -aidan

2 2

[keystone][cache] How to tune role cache
by Eugen Block 08 Nov '22

08 Nov '22

Hi *, one of our customers has two almost identical clouds (Victoria), the only difference is that one of them has three control nodes (HA via pacemaker) and the other one only one control node. They use terraform to deploy lots of different k8s clusters and other stuff. In the HA cloud they noticed keystone errors when they purged a project (cleanly) and started the redeployment immediately after that. We did some tests to find out which exact keystone cache it is and it seems to be the role cache (default 600 seconds) which leads to an error in terraform, it reports that the project was not found and refers to the previous ID of the project. The same deployment seems to work in the single-control environment without these errors, it just works although the cache is enabled as well. I already tried to reduce the cache_time to 30 seconds but that doesn't help (although it takes more than 30 seconds until terraform is ready after the prechecks). But the downside of disabling the role cache entirely leads to significantly longer response times when using the dashboard or querying the APIs. Is there any way to tune the role cache in a way so we could have both a reasonable performance as well as being able to redeploy projects without a "sleep 600"? Any comments or recommendations are appreciated! Regards, Eugen

1 1