kubernetes: Connection reset by peer error while running SCTP client/server in Kubernetes.

What happened: I am running a SCTP based client outside kubernetes cluster and SCTP based server inside kubernetes cluster. The client is getting connected to server but while send/recv it is giving “connnection reset by peer error”.

What you expected to happen: The client and server should be able to send and recv messages.

How to reproduce it (as minimally and precisely as possible):

1. Run a SCTP based client outside kubernetes cluster.
2. Run a SCTP based server inside kubernetes cluster.
3. expose the server to client using NodePort service using SCTP protocol.

Anything else we need to know?:

1. my client is running on the master node with ip : 10.129.27.127
2. server is running on a pod with ip: 10.244.1.98
3. server is exposed using Nodeport service where port = 38412, target port = 38412, nodeport = 38412

The client will connect to the server at port 38412. The slave-node is running at ip = 10.129.26.68 where the server gets deployed.

Please find the attached wireshark result:

Below is the description of NodePort service:

Below is the description of server :

Environment:

• Kubernetes version (use kubectl version):

Client Version: version.Info{Major:“1”, Minor:“15”, GitVersion:“v1.15.2”, GitCommit:“f6278300bebbb750328ac16ee6dd3aa7d3549568”, GitTreeState:“clean”, BuildDate:“2019-08-05T09:23:26Z”, GoVersion:“go1.12.5”, Compiler:“gc”, Platform:“linux/amd64”}

Server Version: version.Info{Major:“1”, Minor:“15”, GitVersion:“v1.15.3”, GitCommit:“2d3c76f9091b6bec110a5e63777c332469e0cba2”, GitTreeState:“clean”, BuildDate:“2019-08-19T11:05:50Z”, GoVersion:“go1.12.9”, Compiler:“gc”, Platform:“linux/amd64”}

• Cloud provider or hardware configuration:

I have installed kubeadm on my own system.

• OS (e.g: cat /etc/os-release):

VERSION=“18.04.2 LTS (Bionic Beaver)”

• Kernel (e.g. uname -a):

Linux master-node 5.0.0-29-generic #31~18.04.1-Ubuntu SMP Thu Sep 12 18:29:21 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux

• Install tools:
• Network plugin and version (if this is a network-related bug):
• Others:

The kubernetes nodes are virtual machines.