August 2023 - openstack-discuss - lists.openstack.org

Kayobe VLAN & bond configuration
by BPt-FM BBS IaaS 10 Aug '23

10 Aug '23

Hi! I am trying to set up Openstack with Kayobe and right now I’m very stuck at VLAN configuration with LACP bonds on my two network nodes. Last Friday the network even came to a hold when activating Openvswitch because it started a broadcast storm or something in the whole company even though I have 2 VLANs just for this project. So I am trying to do it like this: #bond configuration dummy_net_interface: bond0 dummy_net_bootproto: static dummy_net_bond_slaves: - eno12399np0 - eno12409np1 dummy_net_bond_mode: 802.3ad dummy_net_bond_miimon: 100 dummy_net_bond_lacp_rate: fast #configure mgmt interface mgmt_net_interface: brmgmt mgmt_net_bridge_ports: - bond0 #external net work interface and bridge configuration #ext_net_bridge_ports: ext_net_interface: "brbond{{ ext_net_vlan }}" ext_net_bridge_ports: - "bond0.{{ ext_net_vlan }}" Part of my networks.yml … external_net_names: - ext_net … ext_net_cidr: 172.31.246.0/24 ext_net_no_ip: true ext_net_vlan: 705 Last time this worked, when I ran a “kayobe overcloud host configure” (the node had network connection) but then “kayobe overcloud service deploy” killed the network when it started the Openvswitch. Now (today) this configuration doesn’t even allow to reach the network/(to be)neutron nodes (they lose network connection) when the playbook restarts system-networkd. It already creates OVS PHY bridges and lots of stuff even though I ran “overcloud service destroy” before that. Should it do that at this stage? Are there any examples to set up a network node with bonds and vlans and external neutron networks out there? Any suggestions? Thank you 😊 The information in this email is confidential and may be legally privileged. If you are not the intended recipient of this message, any review, disclosure, copying, distribution, retention, or any action taken or omitted to be taken in reliance on it is prohibited and may be unlawful. If you are not the intended recipient, please reply to or forward a copy of this message to the sender and delete the message, any attachments, and any copies thereof from your system.

1 0

Exploring the Intersection of AI and Open Source: Aug. 10 at 15 UTC
by Kristin Barrientos 09 Aug '23

09 Aug '23

Hi everyone, This week’s OpenInfra Live episode is brought to you by open source community leaders. Episode: If you stand at the intersection of AI and open source you’ll find a very active discussion around how each concept can apply to the other but you’ll find few definitive answers. This panel of open source community leaders will discuss the current state of AI & open source, the collaborative efforts underway to improve global understanding of this emerging use case and how open source community members can participate in solving today’s unanswered questions. Date and time: Aug 10, 2023 at 10:00 CST (15:00 UTC) You can watch us live on: YouTube: https://www.youtube.com/watch?v=LLujjvEIbPk LinkedIn: https://www.linkedin.com/events/7094342351971598337/comments/ WeChat: The recording will be posted on OpenInfra Foundation WeChat after the live stream Speakers: Mark Collier, Pamela Chestek, Stefano Maffulli, Armstrong Foundjem, Stephen O'Grady Have an idea for a future episode? Share it now at ideas.openinfra.live. Thanks, Kristin Barrientos Marketing Coordinator OpenInfra Foundation

1 0

[operators][swift][s3][ceph] interoperability s3/swift protocol using radosgw
by Benedikt Trefzer 09 Aug '23

09 Aug '23

Hi Openstack list I'm stuck with a problem: We implemented object storage using ceph radosgw, using keystone authentication and rgw multi-tenancy. So far so good. If I create a container named 'test-swift' using swift protocol, radosgw-admin shows me a bucket created as: "*\/test-swift" (* is the uuid of the opensstack project) which seems to be the correct behaviour. If I create a bucket with s3cmd named 'test-s3', radosgw-admin shows the bucket as: "test-s3" which is not the expected behaviour since this is not multi tenancy ! Can anyone point me to the right direction to solve this ? Could the use of v2 signatures for S3 be a problem ? Further information: Ceph version: Jewel Openstack keystone: train Access to rgw through Apache fcgi (sitting behind haproxy nodes) Thanks for all the hints Benedikt PS: - I know that Jewel and train are old versions, but ... - I know that S3/Swift protocols are not fully supported using rgw

1 0

Cinder Bug Report 2023-08-09 - No new Bugs This Week!
by Rajat Dhasmana 09 Aug '23

09 Aug '23

1 0

[skyline] Username or password is incorrect
by Satish Patel 09 Aug '23

09 Aug '23

Folks, Try to install skyline UI to replace horizon using doc: https://docs.openstack.org/skyline-apiserver/latest/install/docker-install-… Everything went well and I got a login page on http://x.x.x.x:9999 also it pulled Region/Domains. When I am trying to login with my account, I get an error: Username or Password is incorrect. I am using sqlite DB for skyline as per documents. No errors in logs command $ docker logs skyline When I use Chrome Developer Tools then it was indicating an error in these URLs. http://openstack.example.com:9999/api/openstack/skyline/api/v1/profile http://openstack.example.com:9999/api/openstack/skyline/api/v1/policies 401 Unauthorized ( {"detail":"no such table: revoked_token"} ) Find attached screenshot

2 2

[cinder] prevent change volume type from different AZ
by Nguyễn Hữu Khôi 09 Aug '23

09 Aug '23

Hello guys. I have 2 AZ and 1 volume backend on each AZ. My goal is when 1 volume was created from the AZ then user cannot change it to another volume type on a different AZ. Thank you. Nguyen Huu Khoi

2 2

[neutron] User survey Neutron related questions
by Slawek Kaplonski 09 Aug '23

09 Aug '23

Hi, As we talked on the Neutron team meeting yesterday [1] I prepared etherpad with Neutron related questions and possible answers from the User Survey [2]. I already added there some of my comments. Please check it and comment on it too. We can then discuss it during next week's Neutron team meeting and finally maybe send some proposals to change something there. [1] https://meetings.opendev.org/meetings/networking/2023/networking.2023-08-08… [2] https://etherpad.opendev.org/p/neutron-user-survey-questions-2023 -- Slawek Kaplonski Principal Software Engineer Red Hat

1 0

[glance][ops] Problems snapshotting a VM: image gets deleted
by Massimo Sgaravatto 09 Aug '23

09 Aug '23

I have problems snapshotting a particular VM. This is what happens: [root@cld-ctrl-01 ~]# nova image-create --show --poll 8cfd9323-f9d8-41f8-b607-41597d9c79e9 ns3-new-snap-2023-08-09-2 Server snapshotting... 0% completeERROR (NotFound): No image found with ID 133d4978-1de3-4cef-8804-8b0f07b12e4e<br /><br /> (HTTP 404) [root@cld-ctrl-01 ~]# According to the logs [*], it looks like the snapshot is for some reason deleted, but I don't understand what the problem is Any hints ? Thanks, Massimo [*] 2023-08-09 08:35:05.683 5507 INFO eventlet.wsgi.server [req-b0908999-5e95-4520-ae5a-614c503e91e9 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 192.168.60.105\ ,192.168.60.232 - - [09/Aug/2023 08:35:05] "POST /v2/images/133d4978-1de3-4cef-8804-8b0f07b12e4e/members HTTP/1.1" 200 424 0.037093 2023-08-09 08:35:05.689 5506 INFO eventlet.wsgi.server [-] 192.168.60.234 - - [09/Aug/2023 08:35:05] "OPTIONS /versions HTTP/1.0" 200 1781 0.001015 2023-08-09 08:35:05.690 5507 INFO eventlet.wsgi.server [req-b0908999-5e95-4520-ae5a-614c503e91e9 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 192.168.60.105\ ,192.168.60.232 - - [09/Aug/2023 08:35:05] "GET /v2/schemas/member HTTP/1.1" 200 872 0.003611 2023-08-09 08:35:05.690 5501 INFO eventlet.wsgi.server [-] 192.168.60.234 - - [09/Aug/2023 08:35:05] "OPTIONS /versions HTTP/1.0" 200 1781 0.000932 2023-08-09 08:35:05.770 5500 INFO eventlet.wsgi.server [req-4146429a-27a6-4ac9-a3f7-4534c32c6dd0 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 193.206.210.24\ 0,192.168.60.232 - - [09/Aug/2023 08:35:05] "GET /v2/images/133d4978-1de3-4cef-8804-8b0f07b12e4e HTTP/1.1" 200 1456 0.012723 2023-08-09 08:35:05.812 5500 INFO eventlet.wsgi.server [-] 192.168.60.234 - - [09/Aug/2023 08:35:05] "OPTIONS /versions HTTP/1.0" 200 1781 0.000906 2023-08-09 08:35:05.812 5506 INFO eventlet.wsgi.server [-] 192.168.60.234 - - [09/Aug/2023 08:35:05] "OPTIONS /versions HTTP/1.0" 200 1781 0.001003 2023-08-09 08:35:05.862 5503 INFO eventlet.wsgi.server [req-b0908999-5e95-4520-ae5a-614c503e91e9 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 192.168.60.159\ ,192.168.60.232 - - [09/Aug/2023 08:35:05] "GET /v2/images/133d4978-1de3-4cef-8804-8b0f07b12e4e HTTP/1.1" 200 1456 0.012832 2023-08-09 08:35:05.868 5506 INFO eventlet.wsgi.server [req-b0908999-5e95-4520-ae5a-614c503e91e9 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 192.168.60.159\ ,192.168.60.232 - - [09/Aug/2023 08:35:05] "GET /v2/schemas/image HTTP/1.1" 200 6278 0.002641 2023-08-09 08:35:06.014 5500 INFO eventlet.wsgi.server [req-b0908999-5e95-4520-ae5a-614c503e91e9 a1f915a7a36c471d87d6702255016df4 beaeede3841b47efb6b665a1a667e5b1 - default default] 192.168.60.159\ ,192.168.60.232 - - [09/Aug/2023 08:35:06] "DELETE /v2/images/133d4978-1de3-4cef-8804-8b0f07b12e4e HTTP/1.1" 204 208 0.140910

2 2

[openstack][neutron] set a fixed ip for router gateway
by Nguyễn Hữu Khôi 09 Aug '23

09 Aug '23

Hello guys. I am curious why we can not set a fixed ip for router gateway. [image: image.png] Although I see these codes were merged many years ago. https://review.opendev.org/c/openstack/neutron/+/83664 Nguyen Huu Khoi

2 2

Why service use UTC in db and log use local timezone
by kaqiu pi 08 Aug '23

08 Aug '23

Hello, I have a openstack cluster. And the local time of server is different from UTC. I saw that the log date of openstack service is local time. But when I use the command line to query the openstack service status, time zone is UTC. Such as `opentack compute service list`, "Update At" is UTC. In the email archive[1], I saw that there was a discussion earlier: > i would guess it a deliberate design descisn to only store data information in utc format and leave it to the clinets to convert to local timezones if desired But at that time it was just a guess, I would like to ask if this is the meaning of the original design, why the service use UTC instead of local time? I would appreciate any kind of guidance or help. Best wishes. [1] https://lists.openstack.org/pipermail/openstack-discuss/2021-June/022857.ht…

2 1