503 after server restart

I restarted my docker environment cause i updated my docker version.
After that I’m no longer able to connect to the console or logs. It dose not mater weather I Use rancher-compose or the WebUI.

Here the output of the rancher command:

ERRO[0002] Failed to get logs for Webserver_redis_1: websocket: bad handshake
ERRO[0002] Failed to get logs for Webserver_ShiftheroWeb_1: websocket: bad handshake
ERRO[0002] Failed to get logs for Webserver_GitlabCiRunner_1: websocket: bad handshake
ERRO[0002] Failed to get logs for Webserver_Gitlab_1: websocket: bad handshake

Any suggestions?

Greetings from Switzerland
Thomas

Can you please provide your docker version and Rancher version?

Also, could you get more details on your setup? Is it a single node or multi-node? Do you have any special configuration with it?

Finally I was able to solve the problem by launching a new instance of the rancher/server and network agent container.

A Restart of the rancher containers didn’t help

My environment is:

Single node with external DB
Rancher	v0.39.0
Cattle	v0.95.0
User Interface	v0.52.0
Rancher Compose	beta/latest
Docker version 1.7.1, build 786b29d

No special config

Do you still have the broken Rancher server container running? Wondering if we could get the logs to see what was reporting on that side after you restarted.

Are you on AWS by any chance? You could have ephemeral IPs that changed when the server was restarted.

@vincent I’m on AWS but i haven’t rebooted the box and I use a fix IP

@denise I have started a new instance of the same image to get it running. The old instance is Unfortunately already deleted. As this is my production environment I would not like to restart the current server instance to reproduce the issue.

@zauberertz No worries on restarting the broken server. I was just hoping to get some logs. Next time, could you try and grab some logs so that we could troubleshoot your issue?
http://docs.rancher.com/rancher/faqs/troubleshooting/#server-logs