Plausible crashing on startup

[yaakov-h]

Hi,

I was previously running Plausible CE adapter from the docker-compose file to a Kubernetes deployment and had that working just by running the same container image with the right volume mounts and environment variables set.

I'm now trying to do the same thing but as a .container podman/systemd/Quadlet definition, as follows:

[Unit]
Description=Plausible

[Container]
ContainerName=plausible
Image=ghcr.io/plausible/community-edition:v2.1.4
Label=io.containers.autoupdate=local
Network=slirp4netns:allow_host_loopback=true
Volume=/opt/plausible:/var/lib/plausible
Ulimit=nofile=65535:65535
PublishPort=8016:8000

Exec=sh -c "/entrypoint.sh db createdb && /entrypoint.sh db migrate && /entrypoint.sh run"

Environment=TMPDIR=/var/lib/plausible/tmp
Environment=BASE_URL=https://my.site.example.com
Environment=SECRET_KEY_BASE=<random stuff>

Environment=CLICKHOUSE_DATABASE_URL=http://host.containers.internal:8123/plausible_events_db
Environment=DATABASE_URL=postgres://plausible:<password>@host.containers.internal:5432/plausible

Environment=HTTP_PORT=8000

[Service]
Restart=always

[Install]
WantedBy=default.target

I've managed to get through db creation and migrations but ./entrypoint.sh run is failing and crashing without outputting any logs.

When it runs I just get this:

Loading plausible..
Starting dependencies..
Starting repos..
Plausible.Repo database already exists
Plausible.IngestRepo database already exists
Creation of Db successful!
Loading plausible..
Starting dependencies..
Starting repos..
Runtime terminating during boot (terminating)

Crash dump is being written to: erl_crash.dump...

Any ideas what I may be missing here?

[ruslandoga]

👋 @yaakov-h

Could it be a permission issue on /opt/plauisble?

Similar to plausible/community-edition#206

[yaakov-h]

Thanks, that pointed me in the right direction. It was a permissions issue, but chown 999 didn't work as podman has a user and group mapping system when using rootless containers.

I had to:

1. chmod 777 to get Plausible to start
2. Check to see what uid and gid Plausible was actually running as (uid was 100998 for example). A quick way to do this is ps -o user $(pidof beam.smp)
3. chmod 755 to get permissions back to normal
4. chown to the actual host uid and gid