<p>Docker has transformed the way we build, ship, and run applications. But like any powerful tool, it’s easy to misuse. Many developers (even seasoned ones) stumble into mistakes that lead to bloated images, security risks, or unpredictable deployments.</p><p>In this blog, we’ll explore <strong>20 common Docker mistakes to avoid</strong>, along with <strong>practical fixes and best practices</strong>.</p><h2>1. Running Containers as Root</h2><p><strong>Mistake:</strong> By default, containers run as <code>root</code>, which poses security risks if compromised.</p><p><strong>Fix:</strong> Add a non-root user in your Dockerfile:</p><pre><code class="language-dockerfile">RUN useradd -m appuser
USER appuser</code></pre><hr><h2>2. Using the <code>latest</code> Tag Everywhere</h2><p><strong>Mistake:</strong> Relying on <code>:latest</code> makes builds unpredictable since the underlying image can change anytime.</p><p><strong>Fix:</strong> Pin image versions (e.g., <code>python:3.11-slim</code>) and update intentionally.</p><hr><h2>3. Ignoring Image Size</h2><p><strong>Mistake:</strong> Large images slow down builds, deployments, and waste bandwidth.</p><p><strong>Fix:</strong></p><ul><li><p>Use slim or alpine images.</p></li><li><p>Multi-stage builds to discard build tools.</p></li><li><p>Clean caches with <code>apt-get clean &amp;&amp; rm -rf /var/lib/apt/lists/*</code>.</p></li></ul><h2>4. Not Using Multi-Stage Builds</h2><p><strong>Mistake:</strong> Keeping compilers, build dependencies, and temporary files in production images.</p><p><strong>Fix:</strong></p><pre><code class="language-dockerfile">FROM golang:1.21 AS builder
WORKDIR /app
COPY . .
RUN go build -o main

FROM debian:bookworm-slim
COPY --from=builder /app/main /usr/local/bin/main
CMD ["main"]</code></pre><h2>5. Hardcoding Secrets in Dockerfile</h2><p><strong>Mistake:</strong> Adding API keys, passwords, or tokens in Dockerfiles or ENV variables.</p><p><strong>Fix:</strong> Use secret managers (AWS Secrets Manager, Vault, Kubernetes secrets) and pass them at runtime.</p><h2>6. Not Using <code>.dockerignore</code></h2><p><strong>Mistake:</strong> Sending unnecessary files (node_modules, logs, .git) to the build context.</p><p><strong>Fix:</strong> Create a <code>.dockerignore</code>:</p><pre><code>.git
node_modules
*.log</code></pre><h2>7. Copying Everything in Dockerfile</h2><p><strong>Mistake:</strong> Using <code>COPY . .</code> blindly includes extra files.</p><p><strong>Fix:</strong> Be explicit:</p><pre><code class="language-dockerfile">COPY package.json yarn.lock ./
RUN yarn install
COPY src/ ./src</code></pre><h2>8. Not Layering Commands Efficiently</h2><p><strong>Mistake:</strong> Each Dockerfile instruction creates a new layer, bloating the image.</p><p><strong>Fix:</strong> Combine related commands:</p><pre><code class="language-dockerfile">RUN apt-get update &amp;&amp; apt-get install -y \
    curl wget git \
 &amp;&amp; rm -rf /var/lib/apt/lists/*</code></pre><h2>9. Overusing <code>ADD</code> Instead of <code>COPY</code></h2><p><strong>Mistake:</strong> <code>ADD</code> has hidden behaviors (e.g., auto-extraction) that aren’t always needed.</p><p><strong>Fix:</strong> Use <code>COPY</code> unless you specifically need <code>ADD</code>.</p><h2>10. Not Setting Health Checks</h2><p><strong>Mistake:</strong> Containers can be running but unhealthy, and Docker won’t know.</p><p><strong>Fix:</strong></p><pre><code class="language-dockerfile">HEALTHCHECK CMD curl -f http://localhost:8080/ || exit 1</code></pre><h2>11. Ignoring Logs</h2><p><strong>Mistake:</strong> Writing logs to files inside containers makes them inaccessible.</p><p><strong>Fix:</strong> Write logs to <code>stdout</code>/<code>stderr</code> so Docker logging drivers can capture them.</p><h2>12. Forgetting About Resource Limits</h2><p><strong>Mistake:</strong> Containers without limits can consume all CPU/memory.</p><p><strong>Fix:</strong> Use:</p><pre><code class="language-bash">docker run --memory=512m --cpus=1 myapp</code></pre><h2>13. Running Too Many Processes in One Container</h2><p><strong>Mistake:</strong> Treating containers like VMs and running multiple unrelated processes inside.</p><p><strong>Fix:</strong> Follow <strong>“one process per container”</strong>. Use Docker Compose or Kubernetes to orchestrate multiple services.</p><h2>14. Using Containers for Persistent Storage</h2><p><strong>Mistake:</strong> Storing database data directly inside containers.</p><p><strong>Fix:</strong> Use <strong>volumes</strong> or external storage for persistence.</p><pre><code class="language-bash">docker run -v db_data:/var/lib/postgresql/data postgres</code></pre><h2>15. Neglecting Security Updates</h2><p><strong>Mistake:</strong> Using old base images full of vulnerabilities.</p><p><strong>Fix:</strong></p><ul><li><p>Regularly rebuild images.</p></li><li><p>Use <code>docker scan</code> or tools like Trivy to check vulnerabilities.</p></li></ul><h2>16. Building Images Without Caching in Mind</h2><p><strong>Mistake:</strong> Putting frequently changing layers (like source code) before dependencies, invalidating cache.</p><p><strong>Fix:</strong> Order Dockerfile steps properly: install dependencies first, then copy source code.</p><h2>17. Ignoring Networking Basics</h2><p><strong>Mistake:</strong> Hardcoding IPs or using <code>localhost</code> when linking containers.</p><p><strong>Fix:</strong> Use Docker networks:</p><pre><code class="language-bash">docker network create appnet
docker run --network appnet myapp</code></pre><h2>18. Not Cleaning Up Dangling Images/Volumes</h2><p><strong>Mistake:</strong> Unused images, volumes, and containers pile up, wasting disk.</p><p><strong>Fix:</strong></p><pre><code class="language-bash">docker system prune -af
docker volume prune</code></pre><h2>19. Forgetting About Multi-Architecture Builds</h2><p><strong>Mistake:</strong> Building only for <code>amd64</code>, which breaks on ARM (Raspberry Pi, Apple M1/M2).</p><p><strong>Fix:</strong> Use <code>docker buildx</code>:</p><pre><code class="language-bash">docker buildx build --platform linux/amd64,linux/arm64 -t myapp .</code></pre><h2>20. Skipping Documentation &amp; Version Control for Dockerfiles</h2><p><strong>Mistake:</strong> No comments, no versioning, no clarity.</p><p><strong>Fix:</strong> Document decisions in the Dockerfile and keep it in source control alongside your app.</p><h1>Final Thoughts</h1><p>Docker is simple to get started with but tricky to master. These <strong>20 mistakes</strong> are among the most common pitfalls developers face. By adopting these best practices—secure, lean, and well-documented Dockerfiles—you’ll avoid headaches in production and ensure smoother deployments.</p><p>What Docker mistakes have you made before? Drop them in the comments—I’d love to hear your stories!</p>

Docker

20 Common Docker Mistakes to Avoid and How to Fix