How to write a kubernetes pod configuration to start two containers

Question

I would like to create a kubernetes pod that contains 2 containers, both with different images, so I can start both containers together.

Currently I have tried the following configuration:

{
  "id": "podId",
  "desiredState": {
    "manifest": {
      "version": "v1beta1",
      "id": "podId",
      "containers": [{
        "name": "type1",
        "image": "local/image"
        },
        {
        "name": "type2",
        "image": "local/secondary"
        }]
    }
  },
  "labels": {
    "name": "imageTest"
  }
}

However when I execute kubecfg -c app.json create /pods I get the following error:

F0909 08:40:13.028433 01141 kubecfg.go:283] Got request error: request [&http.Request{Method:"POST", URL:(*url.URL)(0xc20800ee00), Proto:"HTTP/1.1", ProtoMajor:1, ProtoMinor:1, Header:http.Header{}, B
ody:ioutil.nopCloser{Reader:(*bytes.Buffer)(0xc20800ed20)}, ContentLength:396, TransferEncoding:[]string(nil), Close:false, Host:"127.0.0.1:8080", Form:url.Values(nil), PostForm:url.Values(nil), Multi
partForm:(*multipart.Form)(nil), Trailer:http.Header(nil), RemoteAddr:"", RequestURI:"", TLS:(*tls.ConnectionState)(nil)}] failed (500) 500 Internal Server Error: {"kind":"Status","creationTimestamp":
null,"apiVersion":"v1beta1","status":"failure","message":"failed to find fit for api.Pod{JSONBase:api.JSONBase{Kind:\"\", ID:\"SSH podId\", CreationTimestamp:util.Time{Time:time.Time{sec:63545848813, nsec
:0x14114e1, loc:(*time.Location)(0xb9a720)}}, SelfLink:\"\", ResourceVersion:0x0, APIVersion:\"\"}, Labels:map[string]string{\"name\":\"imageTest\"}, DesiredState:api.PodState{Manifest:api.ContainerMa
nifest{Version:\"v1beta1\", ID:\"podId\", Volumes:[]api.Volume(nil), Containers:[]api.Container{api.Container{Name:\"type1\", Image:\"local/image\", Command:[]string(nil), WorkingDir:\"\", Ports:[]ap
i.Port(nil), Env:[]api.EnvVar(nil), Memory:0, CPU:0, VolumeMounts:[]api.VolumeMount(nil), LivenessProbe:(*api.LivenessProbe)(nil)}, api.Container{Name:\"type2\", Image:\"local/secondary\", Command:[]string(n
il), WorkingDir:\"\", Ports:[]api.Port(nil), Env:[]api.EnvVar(nil), Memory:0, CPU:0, VolumeMounts:[]api.VolumeMount(nil), LivenessProbe:(*api.LivenessProbe)(nil)}}}, Status:\"\", Host:\"\", HostIP:\"\
", PodIP:\"\", Info:api.PodInfo(nil), RestartPolicy:api.RestartPolicy{Type:\"RestartAlways\"}}, CurrentState:api.PodState{Manifest:api.ContainerManifest{Version:\"\", ID:\"\", Volumes:[]api.Volume(nil
), Containers:[]api.Container(nil)}, Status:\"\", Host:\"\", HostIP:\"\", PodIP:\"\", Info:api.PodInfo(nil), RestartPolicy:api.RestartPolicy{Type:\"\"}}}","code":500}

How can I modify the configuration accordingly?

Running kubernetes on a vagrant vm (yungsang/coreos).

Cowman

Were you able to solve this? It's now a bit out of date - v1 may address your issue. – Lindon

I no longer use this so i do not know. Yes its quite old now so it may be fixed – Cowman

Answer 1

The error in question here is "failed to find fit". This generally happens when you have a port conflict (try and use the same hostPort too many times or perhaps you don't have any worker nodes/minions.

I'd suggest you either use the Vagrant file that is in the Kubernetes git repo (see http://kubernetes.io) as we have been trying to make sure that stays working as Kubernetes is under very active development. If you want to make it work with the CoreOS single machine set up, I suggest you hop on IRC (#google-containers on freenode) and try and get in touch with Kelsey Hightower.

Kronos

the Vagrantfile didn't work as it was searching for files under a linux distribution and I am on windows – Cowman

Answer 2

Your pod spec file looks like invalid. According to http://kubernetes.io/v1.0/docs/user-guide/walkthrough/README.html#multiple-containers, a valid multiple containers pod spec should like this

apiVersion: v1
kind: Pod
metadata:
  name: www
spec:
  containers:
  - name: nginx
    image: nginx
    volumeMounts:
    - mountPath: /srv/www
      name: www-data
      readOnly: true
  - name: git-monitor
    image: kubernetes/git-monitor
    env:
    - name: GIT_REPO
      value: http://github.com/some/repo.git
    volumeMounts:
    - mountPath: /data
      name: www-data
  volumes:
  - name: www-data
    emptyDir: {}

Latest doc at http://kubernetes.io/docs/user-guide/walkthrough/#multiple-containers

Scarlet

Answer 3

apiVersion: v1
kind: Pod
metadata:
 name: test
spec:
 containers:
 - name: wp
   image: wordpress
   resources: 
    requests:
     memory: "64Mi"
     cpu: "250m" 
    limits:
     memory: "128Mi"
     cpu: "500m"
 - name: ng
   image: nginx
   imagePullPolicy: IfNotPresent

Pullen

Where to put this? This is a code only answer, so any explanation would be fine. – Triphammer

create a yaml file say first.yaml and put this code in it and then run the command kubectl create -f first.yaml – Pullen