podman pod rm -f -> state improper

[ikke-t]

Is this a BUG REPORT or FEATURE REQUEST? (leave only one on its own line)

/kind bug

Description

Podman sometimes complains about improper state while removing pod.

Steps to reproduce the issue:

1. start pod with play kube

2. remove pod

podman pod rm -f awx

Describe the results you received:

[root@ikke-fedora ~]# podman ps -a
CONTAINER ID  IMAGE                                 COMMAND               CREATED        STATUS            PORTS                   NAMES
64acc9be331b  docker.io/ansible/awx_rabbitmq:3.7.4  docker-entrypoint...  9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  rabbitmq
2985effc04ed  docker.io/ansible/awx_task:latest     /tini -- /bin/sh ...  9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  awxtask
ff85a508a08b  docker.io/ansible/awx_web:latest      /tini -- /bin/sh ...  9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  awxweb
9f49483fb6ae  docker.io/library/memcached:alpine    docker-entrypoint...  9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  memcached
50afac523c1b  docker.io/library/postgres:9.6        docker-entrypoint...  9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  postgres
1bb1bc4e8532  k8s.gcr.io/pause:3.1                                        9 minutes ago  Up 9 minutes ago  0.0.0.0:8052->8052/tcp  bc97cebf2cfb-infra
[root@ikke-fedora ~]# podman pod rm -f awx
Error: cannot remove storage for container 1bb1bc4e85324e22e2b2bfac4aa3bcb7cb3b7f913c96643b80c829c2d399bafc as it is running or paused: container state improper

yet after while it works:

[root@ikke-fedora ~]# podman pod rm -f awx
bc97cebf2cfb69e0b980f12664953f89880ab5728688d5f912661bac5a056151

Describe the results you expected:

I'd expect it to work immediately at once. Perhaps some incorrect order or race doing it? Trying remove init before others? I thought that was fixed in earlier issue while ago.

Additional information you deem important (e.g. issue happens only occasionally):

Output of podman version:

Name        : podman
Epoch       : 2
Version     : 1.2.0
Release     : 24.dev.git0458daf.fc31

Output of podman info --debug:

(paste your output here)

Additional environment details (AWS, VirtualBox, physical, etc.):

[mheon]

Theory: might be the logic in updateContainerStatus() (oci.go line 487) - it looks strange on initial glance. Will investigate on Monday.

[imain]

I have a similar issue in that I have a pod with no containers in it (not sure how it happened). Now I can list the pod but I cannot delete it.

125  ~/dev-scripts  master $ sudo podman pod ls
POD ID         NAME         STATUS    CREATED       # OF CONTAINERS   INFRA ID
75844985f5e8   ironic-pod   Created   2 hours ago   1                 1cec1f248ac5
 0  ~/dev-scripts  master $ sudo podman pod rm ironic-pod
Error: no such file or directory
 125  ~/dev-scripts  master $

Actually I ran out of disk space on / at one point so this may be the issue.

[mheon]

Are you sure the error is fatal? That looks like a non-fatal cleanup error. The pod's still there after rm runs?

[baude]

@mheon wdyt the status is here?

[mheon]

Cleanup errors are fixed... They might have taken this with them. Will poke around.

[mheon]

Try a podman ps -a --sync - that should clear up status issues