[ci:docs] Add a notice to remove pods/containers before starting the systemd service

[cognition9144]

The pods/containers have to be removed before starting the corresponding systemd service. Otherwise the service will fail. That hasn't been written into the doc. This PR adds that.

[rhatdan]

@vrothberg PTAL

[vrothberg]

That's not needed any more. The generated unit files take care of removing. May be you're running an older version of Podman?

[cognition9144]

That's not needed any more. The generated unit files take care of removing. May be you're running an older version of Podman?

I'm using podman version 2.2.1 which is the latest for fedora 33. Is there any way to install a newer one? Maybe you are refering to podman 3.0 which is not available to fedora 33 yet?

[vrothberg]

This one must not be replaced. I just reran the script.

[cognition9144]

I'll fix it after we figure out the complete instruction

[cognition9144]

Fixed

[vrothberg]

@xcffl, can you elaborate why the pods/containers must be removed before starting the systemd units?

I see two cases where this would be needed:

1. When there's a name conflict. This case is treated implicitly by podman. If a pod or container is created with a specific name via --name, then the generated unit files will be created with --replace.
2. When the previous containers/pods are blocking certain resources (e.g., ports). But none of the examples do that.

[cognition9144]

@xcffl, can you elaborate why the pods/containers must be removed before starting the systemd units?

Here is what I do:

❯ cat create-pod-systemd-service.sh 
#!/bin/bash

POD_NAME=${PWD##*/}

podman-compose down && podman-compose up -d
podman generate systemd --new --files --name $POD_NAME
cp *.service $HOME/.config/systemd/user
podman-compose down
systemctl --user daemon-reload
systemctl --user enable --now pod-$POD_NAME

If I remove the second podman-compose down, the problem occurs.

[vrothberg]

@xcffl can you share the error you're seeing?

[cognition9144]

Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Failed with result 'exit-code'.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Unit process 3164 (podman pause) remains running after unit stopped.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: Failed to start Podman pod-nginx.service.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Scheduled restart job, restart counter is at 5.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: Stopped Podman pod-nginx.service.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Start request repeated too quickly.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Failed with result 'exit-code'.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: pod-nginx.service: Unit process 3164 (podman pause) remains running after unit stopped.
Jan 22 01:47:33 localhost.localdomain systemd[3063]: Failed to start Podman pod-nginx.service.

And the original pod is still alive

[vrothberg]

Can you check via journalctl the error message from Podman? The upper logs don't show that, unfortunately. I suspect the Pod exposes ports (pod-nginx.service).

I propose to reword the proposed changes a bit to: "If the previously created containers or pods are using shared resources, such as ports, make sure to remove them before starting the generated systemd units."

[vrothberg]

Error: error adding pod to state: name "test" is in use: pod already exists

Thanks! This should not happen as the unit files should be generated with --replace. Can you share the generated unit files?

[cognition9144]

# pod-nginx.service
# autogenerated by Podman 2.2.1

[Unit]
Description=Podman pod-nginx.service
Documentation=man:podman-generate-systemd(1)
Wants=network.target
After=network-online.target
Requires=container-nginx_swag_1.service
Before=container-nginx_swag_1.service

[Service]
Environment=PODMAN_SYSTEMD_UNIT=%n
Restart=on-failure
ExecStartPre=/bin/rm -f %t/pod-nginx.pid %t/pod-nginx.pod-id
ExecStartPre=/usr/bin/podman pod create --infra-conmon-pidfile %t/pod-nginx.pid --pod-id-file %t/pod-nginx.pod-id --name=nginx --share net
ExecStart=/usr/bin/podman pod start --pod-id-file %t/pod-nginx.pod-id
ExecStop=/usr/bin/podman pod stop --ignore --pod-id-file %t/pod-nginx.pod-id -t 10
ExecStopPost=/usr/bin/podman pod rm --ignore -f --pod-id-file %t/pod-nginx.pod-id
PIDFile=%t/pod-nginx.pid
KillMode=none
Type=forking

[Install]
WantedBy=multi-user.target default.target

❯ cat ~/.config/systemd/user/container-nginx_swag_1.service 
# container-nginx_swag_1.service
# autogenerated by Podman 2.2.1

[Unit]
Description=Podman container-nginx_swag_1.service
Documentation=man:podman-generate-systemd(1)
Wants=network.target
After=network-online.target

[Service]
Environment=PODMAN_SYSTEMD_UNIT=%n
Restart=on-failure
ExecStartPre=/bin/rm -f %t/container-nginx_swag_1.pid %t/container-nginx_swag_1.ctr-id
ExecStart=/usr/bin/podman run --conmon-pidfile %t/container-nginx_swag_1.pid --cidfile %t/container-nginx_swag_1.ctr-id --cgroups=no-conmon --pod-id-file %t/pod-nginx.pod-id --replace --name=nginx_swag_1 -d --label io.podman.compose.config-hash=123 --label io.podman.compose.project=nginx --label io.podman.compose.version=0.0.1 --label com.docker.compose.container-number=1 --label com.docker.compose.service=swag --network host -e PUID=1000 ** ghcr.io/linuxserver/swag
ExecStop=/usr/bin/podman stop --ignore --cidfile %t/container-nginx_swag_1.ctr-id -t 10
ExecStopPost=/usr/bin/podman rm --ignore -f --cidfile %t/container-nginx_swag_1.ctr-id
PIDFile=%t/container-nginx_swag_1.pid
KillMode=none
Type=forking

[Install]
WantedBy=multi-user.target default.target

[zhangguanzhang]

need rebase to one commit

[vrothberg]

LGTM, thanks!

[cognition9144]

need rebase to one commit

Sure. Done.

[vrothberg]

@zhangguanzhang @rhatdan PTAL

[rhatdan]

/lgtm

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

Approval requirements bypassed by manually added approval.

This pull-request has been approved by: xcffl

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment