Bug with nse-composition example.

[isZams]

Expected Behavior

Hi everyone. I'm trying to use the nse-composition, modified to handle kernel interfaces in the composition. However, using a cluster with a single node, everything is ok. Using a cluster with 2 worker nodes and 1 master, the pod in the middle of the chain (composed by 3 elements) cannot start. Here there are:

• folder example-composition.zip to deploy the K8s resources
• logs file logs.txt about the error of the pod in the middle of the chain. The logs underline how is missing the config.yaml that is used by the PBR example. However i didn't configure this example to handle also this part and in the other case this error does not appear.

Thanks for your help in advance!

EDIT

I was checking where the various pods were located in the nodes. Could there be a problem with pods, even of the same namespace, being located on different nodes?

suricata             nse-kernel-86db86bcd5-zbdhr                 2/2     Running            3 (60m ago)      17h   10.244.1.4   nsm-worker2         <none>           <none>
suricata             nse-passthrough-4-5868687d95-2zmnj          2/3     CrashLoopBackOff   25 (2m44s ago)   17h   10.244.2.4   nsm-worker          <none>           <none> 

[isZams]

I deployed the pods within the same node but the problem is still there

[anselmobattisti]

@isZams how did you solve this problem?

[isZams]

@anselmobattisti Actually i didn't solve the problem. I tried to deploy the resources in a cluster with different nodes. With 3 nodes should work. May the new versions solve the problem?