Running instance not visible in project context

[Remo Maurer]

I'm running an Openstack cluster (ZED) with 3 controllers and 30 worker nodes.

The cluster has been created on Ocata. It has been upgraded to each newer version available.

I have some old instances. They are running and doing their job.

Doing: openstack server list --all-projects, the instance is listed

+----------------------------+----------------------------+-------------------+----------------------------+----------------------------+-------------------------------+
| ID | Name | Status | Networks | Image | Flavor |
+----------------------------+----------------------------+-------------------+----------------------------+----------------------------+-------------------------------+
| df9871d8-efff-486d-aa9f- | turtle | ACTIVE | internal=10.0.3.82, | N/A (booted from volume) | p1.project_turtle.1 |
| 67681be25513 | | | xxx.xx.xxx.xxx | | |

Doing: openstack server list --project 5c41bffc00d54776a01c8c5173e9764c --print-empty, the instance is not listed

+----+------+--------+----------+-------+--------+
| ID | Name | Status | Networks | Image | Flavor |
+----+------+--------+----------+-------+--------+
+----+------+--------+----------+-------+--------+

Doing: openstack server show df9871d8-efff-486d-aa9f-67681be25513, you see, that the instance is assigned to the project 5c41bffc00d54776a01c8c5173e9764c.

+------------------------------------+------------------------------------+
| Field | Value |
+------------------------------------+------------------------------------+
| OS-DCF:diskConfig | AUTO |
| OS-EXT-AZ:availability_zone | nova |
| OS-EXT-SRV-ATTR:host | apu-worker09.internal |
| OS-EXT-SRV-ATTR:hostname | turtle |
| OS-EXT-SRV- | apu-worker09.internal |
| ATTR:hypervisor_hostname | |
| OS-EXT-SRV-ATTR:instance_name | instance-000019e9 |
| OS-EXT-SRV-ATTR:kernel_id | |
| OS-EXT-SRV-ATTR:launch_index | 0 |
| OS-EXT-SRV-ATTR:ramdisk_id | |
| OS-EXT-SRV-ATTR:reservation_id | r-z6u1w4kw |
| OS-EXT-SRV-ATTR:root_device_name | /dev/vda |
| OS-EXT-SRV-ATTR:user_data | None |
| OS-EXT-STS:power_state | Running |
| OS-EXT-STS:task_state | None |
| OS-EXT-STS:vm_state | active |
| OS-SRV-USG:launched_at | 2020-05-18T11:32:31.000000 |
| OS-SRV-USG:terminated_at | None |
| accessIPv4 | |
| accessIPv6 | |
| addresses | internal=10.0.3.82, xxx.xx.xxx.xxx |
| config_drive | |
| created | 2020-05-18T11:32:13Z |
| description | None |
| flavor | disk='200', ephemeral='0', , origi |
| | nal_name='p1.project_turtle.1', |
| | ram='16384', swap='0', vcpus='16' |
| hostId | 94b099c90d9b40e93ab62568a78b6e9098 |
| | ca7f78a1833ace100d127c |
| host_status | UP |
| id | df9871d8-efff-486d-aa9f- |
| | 67681be25513 |
| image | N/A (booted from volume) |
| key_name | rema |
| locked | False |
| locked_reason | None |
| name | turtle |
| progress | 0 |
| project_id | 053f39774bd741e7b84ee6367ff437f0 |
| properties | |
| security_groups | name='Allow_HTTP_HTTPS' |
| | name='Allow_8080_8181' |
| | name='Allow_ICMP' |
| | name='Allow_FTP' |
| | name='Allow_SSH' |
| | name='default' |
| server_groups | [] |
| status | ACTIVE |
| tags | |
| trusted_image_certificates | None |
| updated | 2023-02-02T10:08:22Z |
| user_id | ca1bd6d4e1b7417c905630cba84c17cf |
| volumes_attached | delete_on_termination='False', |
| | id='a7374290-35aa-4fdb- |
| | bc29-c05e612c9ca0' |
+------------------------------------+------------------------------------+

Environment info:
Hostnodes: Ubuntu Jammy
Openstack Version: ZED (deployed with Kolla-Ansible)
Cluster history: Deployed on Ocata, upgraded to every version until ZED
CLI-Client: python-openstackclient 6.0.0

The same issue on the GUI. The instance is listed under Admin, Compute, Instances. It is not listed on Project, Compute, Instances.

Everything was working as expected on older version Yoga.

Does anybody have an idea, what could be wrong here?

BR Remo

[Sylvain Bauza (sylvain-bauza)]

Maybe I misunderstood your point, but I can see in the nova show command that the instance is tied to the project 053f39774bd741e7b84ee6367ff437f0 :

 | project_id | 053f39774bd741e7b84ee6367ff437f0 |

Since you're asking for another project ID with openstack server list --project 5c41bffc00d54776a01c8c5173e9764c --print-empty this sounds to me normal that the instance isn't showing up.

Could you please clarify which instance and which project are related ?

[Sylvain Bauza (sylvain-bauza)]

Note : please set the bug report status back to New when you reply or we wouldn't be able to see your reply.

[Remo Maurer (murm)]

I'm running the command as admin. That's why I don't understand, that the instance is shown when I use 'openstack server list --all-projects', when I run (as an admin) 'openstack server list --project <the project where the instance belongs to>, it is not listed.
This effect is only observed on instances that have been created on a pre Victoria Openstack release. The cluster I'm running has been created on Ocata and has been upgraded to each newer release since then.
I'm not sure, which logfile could help to get details about the issue. I'm attaching them when you can give me a hint where to look for this issue.

[Remo Maurer (murm)]

Was messing up some things. No Bug. Sorry

BR Remo