Lab3.4 Calico doesn't work in worker node? Differents IP Ranges

diego.garcia.sison.next

Hello everyone.

I'm doing the lab3.4. I've installed in GCE a master node and in other Instance a controller node.
I use crio like a container engine.
When I create any pod in the Controller node the Endpoint IP range is 10.88.0.0, mean while the PODs created in master node have Endpoint with range 192.168.X.0 like it was definid by calico (I think)

[email protected]:~$ kubectl get pod -o wide
NAME                     READY   STATUS    RESTARTS   AGE    IP               NODE     NOMINATED NODE   READINESS GATES
nginx-55d5bfd679-4pvqs   1/1     Running   0          4s     10.88.0.8        worker   <none>           <none>
nginx-55d5bfd679-b7h9r   1/1     Running   0          3h8m   10.88.0.6        worker   <none>           <none>
nginx-55d5bfd679-l5c4v   1/1     Running   0          3h8m   192.168.219.68   master   <none>           <none>

I don't hace access to 10.88.0.8 or 10.88.0.6, but I can access to 192.168.219.68

[email protected]:~$ telnet 192.168.219.68 80
Trying 192.168.219.68...
Connected to 192.168.219.68.
Escape character is '^]'.

[email protected]:~$ telnet 10.88.0.8 80
Trying 10.88.0.8...

[email protected]:~$ telnet 10.88.0.6 80
Trying 10.88.0.6...

Should the PODs contained in the Controller node have the range 192.168.X.0 as well?

Thank sou much.
Regards.

chrispokorni

Hi @diego.garcia.sison.next,

User created Pods should be assigned IP addresses from the Pod IP block managed by Calico, regardless of the node where they get deployed.

You can try to run kubectl describe pod <pod-name> against one of the pods with a 192.168.x.x IP address and against one of the pods with the 10.88.y.y IP address, and look at the annotations section, it my give you clues where the IP address is coming from.

Regards,
-Chris

diego.garcia.sison.next

Hello chrispokorni!!

First of all, thank you for your help.
Today, when I've powered on the GCE Instances, the Controller node's POD has the correct IP RANGEs. I supposed that the system reboot solved it. Could be?

Name:         nginx-55d5bfd679-2rwcj
Namespace:    default
Priority:     0
Node:         worker/10.2.0.5
Start Time:   Tue, 24 Nov 2020 11:50:53 +0000
Labels:       app=nginx
              pod-template-hash=55d5bfd679
Annotations:  cni.projectcalico.org/podIP: 192.168.171.69/32
              cni.projectcalico.org/podIPs: 192.168.171.69/32
Status:       Running
IP:           192.168.171.69
IPs:
  IP:           192.168.171.69

Have a nice day