Sir, that was 100% my case, solved instantly thanks to you. Many thanks indeed.

Great article. It helped me solve and understand my 100% cpu problem.

Thanks! I had a system complaining that it was out of space (from creating so many zero byte files?) and the logrotate config for a different app was the culprit.

Better use this one-liner to safely delete the files which are outdated anyways:

rm $(find /var/log/nginx/ -type f | egrep -v "(error|access).log(.[0-9]+)?(.gz)?$")

This will prevent rm from deleting files such as "access.log", "access.log.1" and "access.log.2.gz"