0

I have UCP cluster with 3 controllers and 3 regular nodes. I was able to successfully setup the cluster the first time. However, I am trying to do a fresh-install on one of the controllers but i am getting the following error while running the join command:

a
INFO[0001] Verifying your system is compatible with UCP
INFO[0001] Checking that required ports are available and accessible
INFO[0117] Starting local swarm containers
INFO[0120] Starting UCP Controller replica containers
ERRO[0120] Server response: {"message":"etcdserver: peerURL exists"}
ERRO[0120] Failed to start KV store.  Run "docker logs ucp-kv" for more details
FATA[0120] Failed to add member to KV store: {"message":"etcdserver: peerURL exists"}

I am running the following join command:

docker run --rm -it \
  --name ucp \
  -v /var/run/docker.sock:/var/run/docker.sock \
  docker/ucp:1.1.3 \
  join --replica --interactive
meallhour
  • 13,921
  • 21
  • 60
  • 117
  • What does `docker logs ucp-kv` output? – DevOps Dan Sep 23 '16 at 03:39
  • there are no running containers on the replica node. – meallhour Sep 23 '16 at 14:54
  • If you're going to install UCP, install the latest version (2.1.x), since 2.x brought massive improvements in stability and 2.1.x includes many self-healing features that can bring your cluster back from an inconsistent state. There is a universe of difference between 1.x and 2.x and I highly recommend install 2.x if at all possible. – L0j1k Mar 13 '17 at 03:43

0 Answers0