logrotate: Preventing Log Files from Filling Disks

Application log files grow indefinitely, eventually filling the disk and causing service failures.

Configure logrotate with appropriate rotation schedule, compression, and post-rotate scripts.

# Test logrotate config without running
logrotate --debug /etc/logrotate.d/myapp

# Force rotation (ignore schedule)
logrotate --force /etc/logrotate.d/myapp

# Example config: /etc/logrotate.d/myapp
# /var/log/myapp/*.log {
#   daily
#   rotate 14
#   compress
#   delaycompress
#   missingok
#   notifempty
#   create 0640 www-data adm
#   sharedscripts
#   postrotate
#     kill -USR1 $(cat /var/run/myapp.pid) 2>/dev/null || true
#   endscript
# }

# Common options explained:
# daily/weekly/monthly = rotation frequency
# rotate N = keep N old files
# compress = gzip old files
# delaycompress = compress the previous rotation (not current)
# missingok = don't error if log file is missing
# notifempty = skip rotation if file is empty
# postrotate = run script after rotation

Most apps keep file handles open to their log files. After logrotate renames the file, the app still writes to the old inode. The postrotate script sends a signal (SIGHUP or USR1) to make the app re-open its log file under the new name.