Using Bash to monitor running Docker containers

Using Bash to Monitor Running Docker Containers: A Comprehensive Guide

Docker has undoubtedly changed the development landscape by encapsulating applications in containers, leading to simpler deployments and scalability. However, managing and monitoring these containers to ensure they run smoothly can be a challenge. Linux Bash, with its powerful command-line utilities, serves as a crucial tool in the monitoring and management of Docker containers. This guide will walk you through different Bash commands and scripts you can use to monitor your running Docker containers effectively.

Understanding Docker Commands

Before diving into Bash scripts, it's essential to understand basic Docker commands that provide insights into container states.

docker ps: This command lists all currently running containers.
docker stats: Displays a live stream of container(s) resource usage statistics.
docker logs [container_id]: Fetches the logs of a container. Useful for debugging.
docker top [container_id]: Shows running processes in a specific container.

Bash Scripts for Monitoring

Basic Monitoring Script

Let's start by creating a simple script that checks if all necessary Docker containers are up and running:

#!/bin/bash

# Define your containers
containers=("db" "web" "nginx")

# Check each container
for container in "${containers[@]}"; do
  if ! docker ps | grep -q $container; then
    echo "Container $container is not running!"
  else
    echo "Container $container is running."
  fi
done

This script will iterate over a predefined list of container names and check if they appear in the list of running containers.

Monitoring CPU and Memory Usage

The following Bash script uses docker stats command to check the CPU and memory usage of each running container. It will alert if the usage exceeds a given threshold:

#!/bin/bash

# Thresholds
max_cpu=80.0
max_mem=500000000 # 500MB

# Fetch stats from all running containers
docker stats --no-stream --format "table {{.Name}}\t{{.CPUPerc}}\t{{.MemUsage}}" | while read name cpu mem; do
  # Skip header or unwanted lines
  if [[ "$name" == "NAME" || -z "$name" ]]; then
    continue
  fi

  # Remove percentage sign and compare CPU
  cpu=${cpu%.*}
  if (( $(echo "$cpu > $max_cpu" | bc -l) )); then
    echo "High CPU usage on $name: $cpu%"
  fi

  # Extract current memory usage in bytes
  current_mem=$(echo $mem | awk '{print $1*1000000}')
  if (( $(echo "$current_mem > $max_mem" | bc -l) )); then
    echo "High memory usage on $name: $(echo $mem | awk '{print $1, $2}')"
  endif
done

This script uses docker stats --no-stream to fetch the resource usage data one time instead of continuously streaming. It processes the data using awk and Bash built-in arithmetic.

Creating a Log Monitoring Script

Monitoring logs can help in understanding what actions took place in a container. Below is a simple script that tails the latest logs:

#!/bin/bash

# Check if the container name is provided
if [ -z "$1" ]; then
  echo "Please specify a container name."
  exit 1
fi

container_name=$1

# Tail logs
docker logs --tail 50 -f $container_name

This script accepts a container name as an argument and displays the last 50 lines of the container's logs, continuing to follow new logs.

Combining Everything into a Dashboard

For a more interactive monitoring session, you can combine these scripts in a simple text-based dashboard using watch command:

watch -n 5 './monitoring_script.sh'

This command will rerun your monitoring script every 5 seconds, providing you a quasi-real-time dashboard in your terminal.

Conclusion

Monitoring Docker containers doesn't require complex software; often, straightforward Bash scripts can suffice for many use cases. These scripts provide a foundation upon which more complex monitoring solutions can be built. Whether you're managing a few containers on a local machine or numerous containers across multiple servers, mastering the use of Bash in conjunction with Docker commands can significantly streamline your monitoring process.