0

I'm facing an issue where my Docker container sometimes hangs and becomes completely unresponsive. When this happens:

I can't stop the container using docker stop <container_id>.
I can't retrieve logs using docker logs <container_id>.

The only way I've been able to resolve this so far is by restarting the entire Docker service:

sudo systemctl restart docker

While this works, it's not ideal because it affects other running containers as well.

My questions:

Is there a way to forcefully stop or remove a hanged container without restarting the Docker service? How can I diagnose what is causing the container to hang? Is it an issue with the application, API, resource limits (CPU, memory), or something else? Can changing the update policy (e.g., --update-order with start-first or stop-first) prevent such hangs, or help in any way? What tools or techniques can I use to monitor and debug containers to identify the root cause? Any help or guidance on how to stop hanged containers and trace the cause would be greatly appreciated!

Improve this question

1 Answer 1

Reset to default
1

docker kill <container_id>: This sends a SIGKILL signal to the container, which can terminate it forcefully even if it's unresponsive. However, it might not always be the best approach if you need to preserve data or state.

docker rm -f <container_id>: This removes the container without stopping it first. Use this option with caution as it can lead to data loss if the container is still active.

Improve this answer
5
  • But, We cant add this method in restarts in swarm or docker compose. Thanks for answering
    – playwithcode
    Commented Oct 4 at 13:32
  • Could you share more context then? It seems either compose configuration issue not a single container
    – Chachatonel Hashimotto
    Commented Oct 4 at 17:57
  • "I encountered an issue with Docker Swarm where the container got stuck. Initially, I thought it was an API issue (using Django), but it turned out to be a connection issue when the application tries to connect to an RDS m5 instance. There seems to be a max connection limit. Additionally, I'm using the Loki plugin to send container logs to Loki, and if the connection breaks, it fails."
    – playwithcode
    Commented Oct 5 at 12:45
  • Can you update your question with the error and with all configuration files? Intuitively, I'd say it seems VPC network configuration for the RDS since you mentioned the application connection issue but I will know precisely once I have all picture.
    – Chachatonel Hashimotto
    Commented Oct 6 at 9:47
  • I found the issue. because for Longer conenctions with RDS and EC2 ( Django ). Because the those hanging connections. Our API is getting hanged. Then Applying Pulsar or Kafka or SQS to prevent these issue. Not solvable in Docker Swarm itself
    – playwithcode
    Commented Nov 11 at 6:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.