# RKE Cluster setup failed

**URL:** <https://forums.suse.com/t/rke-cluster-setup-failed/17089>\
**Category:** SUSE Rancher Prime\
**Created:** [April 7, 2020, 7:15pm UTC](https://forums.suse.com/t/rke-cluster-setup-failed/17089 "2020-04-07T19:15:20Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![obohovyk](https://avatars.discourse-cdn.com/v4/letter/o/58f4c7/32.png) [@obohovyk](https://forums.suse.com/u/obohovyk)\
**Post date:** [April 7, 2020, 7:15pm UTC](https://forums.suse.com/t/rke-cluster-setup-failed/17089/1 "2020-04-07T19:15:20Z")

</div>

Hi everyone! Unfortunately I’ve tried to setup my cluster on DigitalOcean first time using Rancher rke (just wanted to try) and dissapointed. I spend almost 2 hours with no result. Please help me to realize what’s a problem with rke.  
So I have rke **v1.1.0** (but I tried to use older versions - same result).

Here’s example of my cluster.yml:

> nodes:
> 
> - address: 10.133.1.xx  
> port: “22”  
> internal\_address: 10.133.1.xx  
> role:
> - controlplane  
> hostname\_override: k8s-master  
> user: rke  
> docker\_socket: /var/run/docker.sock  
> ssh\_key: “”  
> ssh\_key\_path: ~/.ssh/id\_rsa  
> ssh\_cert: “”  
> ssh\_cert\_path: “”  
> labels: {}  
> taints: 
> 
> - address: 10.133.47.xx  
> port: “22”  
> internal\_address: 10.133.47.xx  
> role:
> - worker
> - etcd  
> hostname\_override: k8s-worker1  
> user: rke  
> docker\_socket: /var/run/docker.sock  
> ssh\_key: “”  
> ssh\_key\_path: ~/.ssh/id\_rsa  
> ssh\_cert: “”  
> ssh\_cert\_path: “”  
> labels: {}  
> taints:   
> services:  
> etcd:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> external\_urls:   
> ca\_cert: “”  
> cert: “”  
> key: “”  
> path: “”  
> uid: 0  
> gid: 0  
> snapshot: null  
> retention: “”  
> creation: “”  
> backup\_config: null  
> kube-api:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> service\_cluster\_ip\_range: 10.43.0.0/16  
> service\_node\_port\_range: “”  
> pod\_security\_policy: false  
> always\_pull\_images: false  
> secrets\_encryption\_config: null  
> audit\_log: null  
> admission\_configuration: null  
> event\_rate\_limit: null  
> kube-controller:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> cluster\_cidr: 10.42.0.0/16  
> service\_cluster\_ip\_range: 10.43.0.0/16  
> scheduler:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> kubelet:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> cluster\_domain: cluster.local  
> infra\_container\_image: “”  
> cluster\_dns\_server: 10.43.0.10  
> fail\_swap\_on: false  
> generate\_serving\_certificate: false  
> kubeproxy:  
> image: “”  
> extra\_args: {}  
> extra\_binds:   
> extra\_env:   
> …

In general nothing special, but all the time I get an error on master from worker:

`FATA[0000] [Failed to start [rke-etcd-port-listener] container on host [10.133.47.xx]: Error response from daemon: driver failed programming external connectivity on endpoint rke-etcd-port-listener (7afc0c060f6ec0b6fd000d99b1538a77e46c6a12497c326553ea8b30af7b1be8): Bind for 0.0.0.0:2380 failed: port is already allocated]`

This container created from image - rancher/rke-tools:v0.1.56  
`893b62146ebe rancher/rke-tools:v0.1.56 "nc -kl -p 1337 -e e…" 8 minutes ago Created rke-etcd-port-listener`

So at the end I have on worker:

- `893b62146ebe rancher/rke-tools:v0.1.56 "nc -kl -p 1337 -e e…" 11 minutes ago Created rke-etcd-port-listener`
- `6ebc23ccb813 rancher/rke-tools:v0.1.56 "/bin/bash" 11 minutes ago Exited (0) 11 minutes ago cluster-state-deployer`

I’ve researched solutions like [this](https://github.com/rancher/rke/issues/1481) for example and restarted Docker with iptables clean up rules and chains. Same result.

When I try to start `rke-etcd-port-listener` by hands I get this error:  
`Error response from daemon: driver failed programming external connectivity on endpoint rke-etcd-port-listener (b30d465d7f467969e61542948d612b5199eb65310477d56574db730b77709591): Bind for 0.0.0.0:2380 failed: port is already allocated Error: failed to start containers: 893b62146ebe`

Looks like `rke up` simply doesn’t work from scratch. Don’t have any other options to try…

---

<div class="post-metadata">

**Author:** ![catherineluse](https://avatars.discourse-cdn.com/v4/letter/c/cc9497/32.png) [@catherineluse](https://forums.suse.com/u/catherineluse)\
**Post date:** [April 8, 2020, 1:13am UTC](https://forums.suse.com/t/rke-cluster-setup-failed/17089/2 "2020-04-08T01:13:04Z")

</div>

I think there might be a formatting or indentation issue with your cluster.yml. I noticed that our RKE example cluster.yml file doesn’t have quotation marks around the port, but your cluster.yml does have the quotation marks. [https://rancher.com/docs/rke/latest/en/example-yamls/](https://rancher.com/docs/rke/latest/en/example-yamls/)

---

<div class="post-metadata">

**Author:** ![obohovyk](https://avatars.discourse-cdn.com/v4/letter/o/58f4c7/32.png) [@obohovyk](https://forums.suse.com/u/obohovyk)\
**Post date:** [April 9, 2020, 12:25pm UTC](https://forums.suse.com/t/rke-cluster-setup-failed/17089/3 "2020-04-09T12:25:13Z")

</div>

> [@catherineluse](#):
>
> nk there might be

Hm… Interesting. Thank you for an answer, I will check this