Docker Swarm Issue on Azure

General
1

Hello, I seem to have an issue with Docker Swarms on Azure. For reference I followed the steps in this Docker video hosted on YouTube: https://www.youtube.com/watch?v=LlpyiGAVBVg

My issue is that when I shutdown/restart the ‘Manager’ and ‘Worker’ nodes in Azure the Swarm fails to come back up as I would expect and is shown with a purple status in the console with the following status message: “Docker cloud has stopped receiving heartbeats from this Swarm.” Interestingly when I try to get the Docker version information it fails to Connect for the server information - see below:

Client:
Version: 17.06.0-ce
API version: 1.30
Go version: go1.8.3
Git commit: 02c1d87
Built: Fri Jun 23 21:30:30 2017
OS/Arch: windows/amd64

error during connect: Get http://%2F%2F.%2Fpipe%2Fdocker_cloud_9ec2589e-29e4-4466-83f6-a9489c94bf27/v1.30/version: open //./pipe/docker_cloud_9ec2589e-29e4-4466-83f6-a9489c94bf27: The system cannot find the file specified

If I build a completely new swarm which I don’t restart then I have no problem getting the server info. What’s stopping/breaking the communication here, and why does restarting the manager and worker node VMs in Azure break a working swarm???

Perhaps I’ve overlooked something as I am completely new to Docker.

2

Any word on this? I’m having this exact same issue. After restarting the two VMSS’s nothing works. I can’t even SSH into the manager any more

3

I didn’t actually get to the bottom of this Andrew.

What I was able to do however was eventually get SSH access back into a manager node, and then ran the “bring your own swarm” command. This brought them back into view and operational again. If you do manage to get a solution to this then please post back in this thread as I’d be keen to understand any ways to get around this - doesn’t make sense that a reboot of the VMSS should lead to this kind of issue.