Ensure k8s 1.8 works with Contiv

[blaksmit]

Description

Currently we support k8s 1.4 and 1.6. We need to ensure k8s 1.8 is working correctly with Contiv 1.1.5+.

Acceptance Criteria.

1. Ensure k8s 1.7 is working correctly with Contiv 1.1.5+
2. Update required install files

[CNTV-100] created by kahou.lei

[blaksmit]

Hopefully this work can be done in https://github.com/metacloud/kubespray, I am working on rebasing to latest from upstream and should be done by the end of this week.

by vijkatam

[blaksmit]

Solar ticket -> SOLAR-2458 Open

by vijkatam

[blaksmit]

Sorry for the confusion, I am not doing the rebase anymore.

by vijkatam

[blaksmit]

K8s 1.8.2 is working with Contiv 1.1.7 in vagrant environment. Two pods on default network are able to ping.

[vagrant@kubeadm-master ~]$ rpm -qa | grep kub
kubernetes-cni-0.5.1-1.x86_64
kubeadm-1.8.2-0.x86_64
kubelet-1.8.2-0.x86_64
kubectl-1.8.2-0.x86_64
[vagrant@kubeadm-master ~]$ netctl net ls
Tenant   Network      Nw Type  Encap type  Packet tag  SubnetGateway    IPv6Subnet  IPv6Gateway  Cfgd Tag 
------   -------      -------  ----------  ----------  -------       ------     ----------  -----------  ---------
default  contivh1     infra    vxlan       0   132.1.1.0/24  132.1.1.1   
default  default-net  data     vxlan       0   20.1.1.0/24   20.1.1.1    

[vagrant@kubeadm-master ~]$ netctl group create default-net default
Creating EndpointGroup default:default
[vagrant@kubeadm-master ~]$ netctl group ls
Tenant   Group    Network      IP Pool  CfgdTag  Policies  Network profile
------   -----    -------      -------  -------  --------  ---------------
default  default  default-net  
[vagrant@kubeadm-master ~]$ cat bb_pod.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: bb1
  labels:
    app: test
    io.contiv.tenant: default
    io.contiv.network: default-net
    io.contiv.net-group: default
spec:
  containers:
  - name: busybox
    image: busybox:latest
    command:
      - tail
      - -f
      - /dev/null
  restartPolicy: Never    
[vagrant@kubeadm-master ~]$ kubectl create -f bb_pod.yaml
pod "bb1" created
[vagrant@kubeadm-master ~]$ kubectl create -f bb_pod.yaml
pod "bb2" created
[vagrant@kubeadm-master ~]$ kubectl get pods -o wide
NAME      READY     STATUS    RESTARTS   AGE       IP NODE
bb1       1/1       Running   0  28s       20.1.1.3   kubeadm-worker0
bb2       1/1       Running   0  3s20.1.1.4   kubeadm-worker0
[vagrant@kubeadm-master ~]$ kubectl exec -it bb1 sh
/ # ifconfig
eth0      Link encap:Ethernet  HWaddr 02:02:14:01:01:03  
  inet addr:20.1.1.3  Bcast:0.0.0.0  Mask:255.255.255.0
  inet6 addr: fe80::2:14ff:fe01:103/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:1450  Metric:1
  RX packets:8 errors:0 dropped:0 overruns:0 frame:0
  TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:0 
  RX bytes:648 (648.0 B)  TX bytes:648 (648.0 B)

loLink encap:Local Loopback  
  inet addr:127.0.0.1  Mask:255.0.0.0
  inet6 addr: ::1/128 Scope:Host
  UP LOOPBACK RUNNING  MTU:65536  Metric:1
  RX packets:0 errors:0 dropped:0 overruns:0 frame:0
  TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:1 
  RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

/ # ping 20.1.1.4
PING 20.1.1.4 (20.1.1.4): 56 data bytes
64 bytes from 20.1.1.4: seq=0 ttl=64 time=39.250 ms
64 bytes from 20.1.1.4: seq=1 ttl=64 time=0.062 ms
^C
--- 20.1.1.4 ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.062/19.656/39.250 ms
/ # 

by amccormi

[blaksmit]

K8s 1.8.2 is working with Contiv 1.1.7 in proxmox environment. Two pods on default network are able to ping.

[root@mcp1 tmp]# netctl tenant create Tester
Creating tenant: Tester
[root@mcp1 tmp]# netctl net create --subnet=1.1.1.0/24 -g 1.1.1.1 -t Tester TestNet
Creating network Tester:TestNet
root@mcp1 tmp]# netctl group create -t Tester TestNet TestEPG
Creating EndpointGroup Tester:TestEPG
[root@mcp1 tmp]# cat k8s_pod.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: bb1
  labels:
    app: test
    io.contiv.tenant: Tester
    io.contiv.network: TestNet 
    io.contiv.net-group: TestEPG 
spec:
  containers:
  - name: busybox
    image: busybox:latest
    command:
      - tail
      - -f
      - /dev/null
  restartPolicy: Never
[root@mcp1 tmp]# kubectl create -f k8s_pod.yaml
pod "bb2" created
[root@mcp1 tmp]# kubectl create -f k8s_pod.yaml
pod "bb1" created
[root@mcp1 tmp]# kubectl get pods  -o wide
NAME   READY     STATUS    RESTARTS   AGE       IP    NODE
bb1    1/1       Running   0  6s1.1.1.3       mhv2
bb2    1/1       Running   0  24s       1.1.1.2       mhv1
[root@mcp1 tmp]# kubectl exec -it bb2 sh
/ # ip a
1: lo:  mtu 65536 qdisc noqueue qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
34: eth0@if33:  mtu 1450 qdisc noqueue 
    link/ether 02:02:01:01:01:02 brd ff:ff:ff:ff:ff:ff
    inet 1.1.1.2/24 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::2:1ff:fe01:102/64 scope link 
       valid_lft forever preferred_lft forever
/ # ping 1.1.1.3
PING 1.1.1.3 (1.1.1.3): 56 data bytes
64 bytes from 1.1.1.3: seq=0 ttl=64 time=1.350 ms
64 bytes from 1.1.1.3: seq=1 ttl=64 time=0.226 ms

by amccormi

[blaksmit]

PR for the vagrant work: https://github.com/contiv/install/pull/311

by amccormi