ctop and lazydocker: Docker Monitoring Tools

The Problem with docker stats

You run docker stats and get a wall of numbers. CPU, memory, network I/O, block I/O. Raw and unwieldy.

You kill the command and try something else. Maybe docker ps -a | grep something.

Minutes pass. You still don’t know what’s actually consuming CPU.

Two tools fix this: ctop and lazydocker. Both give you interactive dashboards for Docker inspection.

ctop: Top for Containers

ctop is like the top command but for containers. It shows CPU, memory, and I/O in a sortable table.

Installation

# macOS
brew install ctop

# Linux
sudo apt-get install ctop

# Or Docker (meta, I know)
docker run --rm -it --name=ctop \
  --volume /var/run/docker.sock:/var/run/docker.sock \
  quay.io/vektorlab/ctop:latest

Usage

ctop

You get:

CONTAINER                  CPU    MEM    NET RX/TX    BLOCK RX/TX    STATE
postgres:14                8.5%   512MB  120KB/32KB   0B/4.2MB       Running
redis:7                    0.2%   48MB   8.5KB/6.2KB  0B/0B          Running
my-api:latest              42%    892MB  2.3MB/5.4MB  0B/2.1MB       Running
nginx-proxy                1.2%   32MB   1.2MB/1.5MB  0B/0B          Running

It’s real-time and updates every second. Press p to sort by CPU. Press m to sort by memory. Press q to quit.

Practical Scenario

Your Docker host is sluggish. Run ctop, sort by CPU, and immediately see which container is the culprit. No guesswork. No running three commands.

Keyboard Shortcuts

p → Sort by CPU
m → Sort by memory
n → Sort by name
o → View container details
h → Help
q → Quit

Press o on any container to see logs and inspect options.

lazydocker: The Git Diff of Docker

lazydocker is a terminal UI for Docker management. Lighter than Portainer, more interactive than CLI.

Installation

# macOS
brew install lazydocker

# Linux (using Go)
go install github.com/jesseduffield/lazydocker@latest

# Or Docker (to keep your system clean)
docker run --rm -it \
  -v /var/run/docker.sock:/var/run/docker.sock \
  -v /yourpath/config:/home/user/.config/lazydocker \
  lazyteam/lazydocker

What You Get

lazydocker

Interactive UI:

Containers    Images    Volumes
─────────────────────────────────
[postgres:14              ] Running
[redis:7                  ] Running
[my-api:latest            ] Running
[nginx-proxy              ] Running

DETAIL VIEW:
Logs                  Stats
─────────────────────────────────
2026-04-02 10:34:22 INFO...     CPU: 8.5%
2026-04-02 10:34:21 INFO...     Mem: 512MB
2026-04-02 10:34:20 INFO...     PID: 1234

You can:

• View logs for any container
• See resource usage
• Restart containers
• Stop/start containers
• Remove containers
• Exec into containers
• Inspect configuration

Practical Scenario

Your API is OOMing (out of memory). Open lazydocker, select the container, view stats while you watch logs. See the exact moment memory spikes. Check the logs for the cause.

All without running five separate commands.

Keyboard Shortcuts

Arrows → Navigate
Enter  → Select/Expand
d      → Delete container
s      → Stop container
r      → Restart container
e      → Exec into container
l      → View logs
c      → Container stats
? → Help

Comparing Them

ctop is for quick inspection. “What’s using CPU right now?”

lazydocker is for troubleshooting. “Let me see logs and stats while I restart this.”

Real-World Example: Debugging a Memory Leak

Your API container is restarting every hour. Memory climbs until the kernel OOMKills it.

lazydocker

1. Select the api container
2. Press l → view logs (see the OOMKill events)
3. Press c → view stats (watch memory climb over time)
4. Press e → exec into container (run ps aux, top, check for leaks)
5. Press r → restart container

All without ssh’ing into the host, without running docker logs, without juggling six terminal windows.

Installation Recommendations

For servers: Install both. ctop for quick checks, lazydocker for deep investigation.

For local dev: lazydocker is fine. Lighter than Docker Desktop’s UI, faster than Portainer.

For minimal systems: ctop only. Super lightweight, instant startup.

Advanced: Monitoring Multiple Hosts

Both tools work on a single Docker daemon. For multi-host Docker monitoring, they’re not the solution. That’s where you’d use Portainer or Prometheus exporters.

But for a single host or a dev machine? These tools are unmatched.

The Payoff

You stop guessing about container resource usage. You see it in real-time.

Install ctop. It takes 30 seconds. You’ll use it constantly.

Then try lazydocker for deeper troubleshooting. It’s worth the extra second of startup time.

Your docker debugging life just got a lot easier.