Docker Community Forums

Share and learn in the Docker community.

Error when configuring Docker daemon for UCP multi-host networking

(Ken42) #1

Possibly related to this issue though I am not sure.

I went through the multi-host UCP networking steps from here:

When I restart the Docker daemon towards the end of the instructions I get the following error:

level=error msg="discovery error: Unexpected watch error"
level=warning msg="Registering as (IP.address) in discovery failed: client: etcd cluster is unavailable or misconfigured"
level=error msg=“discovery error: client: etcd cluster is unavailable or misconfigured”

I’m doing this on the UCP master node. (There are no follower nodes at this point, just trying to get it to work on 1 node)

I can see that the “ucp-kv” container is running fine and doesn’t seem to show any errors in the logs.

Please let me know any further info I can provide


Software versions:
Redhat 7.2
Docker Engine 1.10
Docker-compose 1.6.0
UCP 0.8.0
Swarm 1.1.0-rc2 (installed by the UCP install)

All on EC2 instances.

(Dcowden) #2

I just had exactly the same issue., but on
UCP 1.0.1
Redhat 7
Docker engine 1.10
compose 1.6

except i’m using VMWare instances instead of EC2 nodes.

(Vivek Saraswat) #3

Hi dcowden,

Are you using the beta docs or the current GA instructions for UCP networking?

(Dcowden) #4

I am following the ones you linked to, whichever those are.

(Vivek Saraswat) #5

Ok got it, so it should be the updated docs then.

In order to solve this problem, let’s walk through the setup steps. Can you show me the terminal input/output from when you set up the networks and the logging output from the ucp-controller and ucp-kv containers when the error occurred?