Homepage crash loop (ENOTCONN)

[Syxpi]

Description

Hello, my homepage has decided to crash repeatedly.
for no reason.

when i look at the logs (docker logs homepage), i see “error: Error: read ENOTCONN” in a loop.

I looked with watch -n 1 docker ps, I see that as soon as I access it, the container restarts.

I'd like to know how to fix it, I tried rm the container, rm the image, repull the image, repull the container, nothing changed.

I've cleared the configuration just in case, that's not the problem.
I've added a video recording if needed.

https://youtu.be/29qU6JjC-Mg

homepage version

v0.10.9 (b5f4daa)

Installation method

Docker

Configuration

That's all default configuration.

Container Logs

[2025-01-08T00:39:02.262Z] error: Error: read ENOTCONN
at tryReadStart (node:net:710:20)
at Socket._read (node:net:725:5)
at Readable.read (node:internal/streams/readable:539:12)
at Socket.read (node:net:781:39)
at new Socket (node:net:498:12)
at Object.Socket (node:net:367:41)
at createSocket (node:internal/child_process:329:14)
at ChildProcess.spawn (node:internal/child_process:444:23)
at spawn (node:child_process:761:9)
at Object.execFile (node:child_process:351:17)

This error, in loop.

Browser Logs

No response

Troubleshooting

Problem not listed in troubleshooting, therefore useless. (and can't troubleshoot, when Homepage crash when i access it)

[shamoon]

Odd, have you tried updating your docker installation? Seems like a very low-level network error where the container can’t even open a socket

[Syxpi]

I've just updated Docker, the problem is still there.

[shamoon]

Yea sorry, I have no other bright ideas. I dont think it's a homepage issue though. I'd be googling it just like you, btw here's what gemini says:

AI Overview
The Docker error "Error: read ENOTCONN" typically indicates a connection issue between the Docker client and the Docker daemon. Here are a few troubleshooting steps you can try:

1. Check Docker Daemon Status:
   Ensure that the Docker daemon is running. You can check this by running:
   Code

   docker info
   If the daemon isn't running, start it with:
   Code

   sudo systemctl start docker

2. Restart Docker:
   Try restarting the Docker service.
   Code

   sudo systemctl restart docker
   If you are using Docker Desktop, restart the application.

3. Check Network Connectivity:
   Verify that your network connection is stable and that there are no firewall rules blocking Docker's communication.

4. Check Docker Logs:
   Examine the Docker logs for more details about the error:
   Code

   docker logs -f dockerd

5. Check for Conflicting Processes:
   Ensure that there are no other processes running on the same ports used by Docker.

6. Update Docker:
   Check if there are any updates available for Docker and update if necessary:
   Code

   sudo apt-get update && sudo apt-get install -y docker-ce docker-ce-cli containerd.io

7. Clean Up Docker Data:
   If the issue persists, you can try cleaning up Docker's data, including images, containers, and volumes. Be cautious as this will delete all your local Docker data.

8. Check for Specific Issues:
   If you are encountering the error while performing a specific operation, such as building an image or running a container, check the documentation for that operation and look for known issues.
   If none of the above solutions work, please provide the following information to help diagnose the issue further:
   The exact command you were running when you encountered the error.
   The output of docker version and docker info.
   Any relevant details about your Docker setup or the operation you were performing.

[Syxpi]

Yea sorry, I have no other bright ideas. I dont think it's a homepage issue though. I'd be googling it just like you, btw here's what gemini says:

AI Overview
The Docker error "Error: read ENOTCONN" typically indicates a connection issue between the Docker client and the Docker daemon. Here are a few troubleshooting steps you can try:

1. Check Docker Daemon Status:
   Ensure that the Docker daemon is running. You can check this by running:
   Code

   docker info
   If the daemon isn't running, start it with:
   Code

   sudo systemctl start docker

2. Restart Docker:
   Try restarting the Docker service.
   Code

   sudo systemctl restart docker
   If you are using Docker Desktop, restart the application.

3. Check Network Connectivity:
   Verify that your network connection is stable and that there are no firewall rules blocking Docker's communication.

4. Check Docker Logs:
   Examine the Docker logs for more details about the error:
   Code

   docker logs -f dockerd

5. Check for Conflicting Processes:
   Ensure that there are no other processes running on the same ports used by Docker.

6. Update Docker:
   Check if there are any updates available for Docker and update if necessary:
   Code

   sudo apt-get update && sudo apt-get install -y docker-ce docker-ce-cli containerd.io

7. Clean Up Docker Data:
   If the issue persists, you can try cleaning up Docker's data, including images, containers, and volumes. Be cautious as this will delete all your local Docker data.

8. Check for Specific Issues:
   If you are encountering the error while performing a specific operation, such as building an image or running a container, check the documentation for that operation and look for known issues.
   If none of the above solutions work, please provide the following information to help diagnose the issue further:
   The exact command you were running when you encountered the error.
   The output of docker version and docker info.
   Any relevant details about your Docker setup or the operation you were performing.

I updated docker, same thing.
If Docker didn't work, I wouldn't even have docker ps, because the daemon wouldn't have found it.
Homepage crashes when you go to it, because without going to it, it doesn't crash (utility?)
I even restarted the LXC container where it was installed (because it is in an LXC container), the problem is still present. I will see to redo a clean install of my LXC, and check if the problem is related to Homepage, or my Container which was just messed up.

[shamoon]

Ah, fair enough. Yes I would suspect the container. I also don’t really understand why one would use docker inside an LXC container, not just directly on the host. But anyway, different strokes for different folks

[Syxpi]

Update:
I deleted the image, the container, even Docker. nothing has changed on the spot.
But after rebooting the LXC, it no longer crashes.

So if you have a problem like mine, don't hesitate to follow these steps:

• Remove the problematic Container.
• Remove Docker. completely.
• Reinstall Docker
• Redo the Container.
  If it still crashes, just restart the host.

[github-actions[bot]]

This discussion has been automatically closed because it was marked as answered. See our contributing guidelines for more details.