Fixes crash when logrotate fails

[chaptergy]

Fixes #1250
Somehow the owner of the logs is sometimes not root, which causes logrotate to fail. This makes sure that a failing of logrotate does not crash npm.

[jc21]

This is an automated message from CI:

Docker Image for build 1 is available on DockerHub as jc21/nginx-proxy-manager:github-pr-1255

Note: ensure you backup your NPM instance before testing this PR image! Especially if this PR contains database changes.

[jc21]

So this just logs the problem but doesn't really fix it. In addition to this change you can add a new file

docker/rootfs/etc/cont-init.d/01-perms

and chmod +x it before committing it. In this file, you can perform operations that are executing at container start:

#!/usr/bin/with-contenv bash

mkdir -p /data/logs
chown -R root:root /data/logs

[chaptergy]

Well It fixes the npm crashing problem not the logrotate problem, yeah.

But with the chown at container start is a great idea, I wouldn'tve thought of that

[chaptergy]

Works as expected. Let's hope this won't break accessing the logs for anyone 🤞

[jc21]

This is an automated message from CI:

Docker Image for build 2 is available on DockerHub as jc21/nginx-proxy-manager:github-pr-1255

Note: ensure you backup your NPM instance before testing this PR image! Especially if this PR contains database changes.