When the container is recycled, the logs are lost. The workaround is to use another logging tool, but that is unnecessary complexity, especially for small deployments.
Also, in general, the logs are not in a friendly format - it's much simpler, especially in small deployments, to simply have the log in a file that can be easily manipulated. Not everyone uses large complex logging apps and servers, etc.
SOLUTION
Many apps have an option to log to file as well. It would be nice if the log could be sent to a file instead of (or in addition to) stdout. If enabled, it could be written to /var/lib/rustdesk/app.log by default.
Personally, I would mount it as /var/log/rustdesk.log:/var/log/rustdesk.log so that I could have it on the host, use logrotate on it, add it to backups etc.
ALTERNATIVES
An unnecessarily complicated environment that includes log aggregation and monitoring, and who knows what else. For small servers that is overkill and hard to manage.
BACKGROUND
Rustdesk seems to log to
stdout
only, and thus to docker's built-in "json driver". The logs can be access like so:$ docker logs -t $CONTAINER_NAME
sudo tail -f /var/lib/docker/containers/$CONTAINER_ID/$CONTAINER_ID-json.log
PROBLEM
When the container is recycled, the logs are lost. The workaround is to use another logging tool, but that is unnecessary complexity, especially for small deployments.
Also, in general, the logs are not in a friendly format - it's much simpler, especially in small deployments, to simply have the log in a file that can be easily manipulated. Not everyone uses large complex logging apps and servers, etc.
SOLUTION
Many apps have an option to log to file as well. It would be nice if the log could be sent to a file instead of (or in addition to) stdout. If enabled, it could be written to
/var/lib/rustdesk/app.log
by default.Personally, I would mount it as
/var/log/rustdesk.log:/var/log/rustdesk.log
so that I could have it on the host, use logrotate on it, add it to backups etc.ALTERNATIVES
An unnecessarily complicated environment that includes log aggregation and monitoring, and who knows what else. For small servers that is overkill and hard to manage.
Additional context
Thank you for considering it!