Disk Spacee Analysis

When disk usage approaches critical levels, begin by examining storage utilization with the df -h command:

[root@docker ~]# df -h
Filesystem      Size  Used Avail Use% Mounted on
devtmpfs        937M     0  937M   0% /dev
tmpfs           949M     0  949M   0% /dev/shm
tmpfs           949M  520K  948M   1% /run
tmpfs           949M     0  949M   0% /sys/fs/cgroup
/dev/vda1       50G    48G   2G   98% /
/dev/vdb1       985G   565G  420G 57% /
overlay         50G    48G   2G   98% /var/lib/docker/overlay2/merged
tmpfs           190M     0  190M   0% /run/user/0

The overlay filesystem represents Docker's storage driver, with actual data residing on /dev/vda1. Primary causes for excessive disk consumption include:

• Accumulated unused images and containers
• Unmanaged container log files

Cleaning Unused Docker Resources

Assess storage allocation across Docker components:

docker system df -v

Remove all unused images, containers, volumes, and networks:

docker system prune -a

Manaigng Container Logs

Container logs accumulate in /var/lib/docker/containers/{containerId}/. Identify large log files:

cd /var/lib/docker/containers/
du -h --max-depth=1

Sample output showing a large log file:

356K    ./container_short_id1
542M    ./container_short_id2
20G     ./container_short_id3

Navigate to the problematic container directory and check log size:

cd container_short_id3
du -sh *-json.log

Implementing Log Rotation

Configure Docker daemon log management in /etc/docker/daemon.json:

{
  "log-driver": "json-file",
  "log-opts": {
    "max-size": "100m",
    "max-file": "3"
  }
}

This configuraton limits individual log files to 100MB while retaining up to 3 rotated files. Apply changes:

systemctl restart docker

Immediate Log Cleanup

For immediate space recovery, truncate existing log files:

cat /dev/null > container_id-json.log