Installation: Setting up elastickube-server [ FAILED ]

[Srikanth Peetha]

Hi
I tried to install elastickube using
curl -s https://elastickube.com | bash
But the installation get struck at "elastickube-server"

---

| | | __ _ **| |() | |/ / | |** ___
| | | |/ ` / __| **| |/ _| ' / | | | ' \ / _
| |**| | (| __ \ || | (| . \ || | |) | /
|___||__,|/||**||_**,|./ _| by ElasticBox

Checking kubectl is available [ ✓ ]
Verifying Kubernetes cluster [ ✓ ]
Setting up elastickube-mongo [ ✓ ]
Setting up elastickube-mongo svc [ ✓ ]
Setting up elastickube-server [ FAILED ]

My cluster is up and I see this in dashboard
Failed to start with docker id 9502c97a429f with error: API error (500): failed to create endpoint k8s_POD.613f02c9_elastickube-server-mwsq9_kube-system_29998e10-eb7f-11e5-b935-00232491226c_4d65b79a on network bridge: Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use Error syncing pod, skipping: API error (500): failed to create endpoint k8s_POD.613f02c9_elastickube-server-mwsq9_kube-system_29998e10-eb7f-11e5-b935-00232491226c_4d65b79a on network bridge: Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use

I see similar thing /var/log/upstart/docker.log

^[[33mWARN^[[0m[6627] Failed to allocate and map port 80-80: Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use ^[[33mWARN^[[0m[6627] failed to cleanup ipc mounts: failed to umount /var/lib/docker/containers/89178c7bf043076f059e5ca393b32039c65afe468d99123cfde4c4f94033b675/shm: no such file or directory ^[[31mERRO^[[0m[6627] Handler for POST /containers/89178c7bf043076f059e5ca393b32039c65afe468d99123cfde4c4f94033b675/start returned error: failed to create endpoint k8s_POD.613f02c9_elastickube-server-mwsq9_kube-system_29998e10-eb7f-11e5-b935-00232491226c_a75f9e27 on network bridge: Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use ^[[31mERRO^[[0m[6627] Handler for GET /images/elasticbox/elastickube-api:latest/json returned error: No such image: elasticbox/elastickube-api:latest ^[[31mERRO^[[0m[6627] Handler for GET /images/elasticbox/elastickube-charts:latest/json returned error: No such image: elasticbox/elastickube-charts:latest ^[[31mERRO^[[0m[6627] Handler for GET /images/elasticbox/elastickube-nginx:latest/json returned error: No such image: elasticbox/elastickube-nginx:latest ^[[31mERRO^[[0m[6636] Handler for GET /images/elasticbox/elastickube-api:latest/json returned error: No such image: elasticbox/elastickube-api:latest ^[[31mERRO^[[0m[6636] Handler for GET /images/elasticbox/elastickube-charts:latest/json returned error: No such image: elasticbox/elastickube-charts:latest ^[[31mERRO^[[0m[6636] Handler for GET /images/elasticbox/elastickube-nginx:latest/json returned error: No such image: elasticbox/elastickube-nginx:latest ^[[33mWARN^[[0m[6637] DEPRECATED: Setting host configuration options when the container starts is deprecated and will be removed in Docker 1.12

I found that by default kubernetes-dashboard service is using port 80 as well.
I changed port used dashboard in cluster/addons/dashboard/dashboard-service.yaml but i still see the same error.

[Alberto Arias Maestro]

Hi,

Yes the port conflict is the problem. If your cluster cannot allocate unique IPs per LoadBalancer you need to change the port to avoid conflicts. I will add some parameters to the script to make this easier, but in the meantime you can download the script and modify the port on the elasticbox-server service.

curl -s elastickube.com -o install.sh

The service yaml definition is embedded into the bash script, you can just change the port with any editor.

Let me know if this solves the issue.

[Rodion Sabodash]

hi! Try run next command

sudo netstat -natp | grep 80

this fail because some process already bind to 80 port. this command shows you what process this

[Srikanth Peetha]

@albertoarias I tried to run the script changing the port but it showed up the same error.
It worked when I manually released the port.
But now I got this

Waiting for LB to be ready [ FAILED ]

So I tried running this command
kubectl --namespace=kube-system describe svc elastickube-server
and I got this

Name: elastickube-server
Namespace: kube-system
Labels: name=elastickube-server
Selector: name=elastickube-server
Type: LoadBalancer
IP: 192.168.3.155
Port: 80/TCP
NodePort: 32359/TCP
Endpoints: 172.16.9.5:80
Session Affinity: None
No events.

Currently I have only one node in my cluster. I don't know if that was the issue.

Irrespective of this I am able to launch elastickube

[Alberto Arias Maestro]

Which LoadBalancer service are you using?

I will patch the setup script as currently we are looking for the string 'LoadBalancer Ingress:' which is not part of your output.

Best

[Alberto Arias Maestro]

@srikanthpeetha I already committed a fix, please reopen it if does not fix the issue.