Greetings NVIDIA team,
I’m super happy that y’all have built this product and even include Docker in the standard image. One thing that’s come out of my playing with the Jetson is that it doesn’t have the multiport kernel module. While I’m not above building the module myself, it’d be FANTASTIC if this were to be included in standard builds. Why? Well, things like k3s (GitHub - k3s-io/k3s: Lightweight Kubernetes ) don’t function without it. I’d LOVE to use k3s or even k8s to do some things in my lab, but the lack of this module makes things challenging.
Below link have some experience for k3s.
opened 10:53PM - 20 May 19 UTC
closed 09:44PM - 05 Mar 20 UTC
**Describe the bug**
Failed to install k3s on NVIDIA Jetson TX2 and Jetson Na… no
`$ uname -a`
`Linux jetson-tx2 4.9.140-tegra rancher/k3s#1 SMP PREEMPT Wed Mar 13 00:30:11 PDT 2019 aarch64 aarch64 aarch64 GNU/Linux`
It failed on:
`kube-system pod/svclb-traefik-47p8d 0/2 ContainerCreating 0 5m1s`
**To Reproduce**
Steps to reproduce the behavior:
Installing the k3s:
`$ curl -sfL https://get.k3s.io | sh -s -`
```
vidia@jetson-tx2:~$ curl -sfL https://get.k3s.io | sh -s -
[INFO] Finding latest release
[INFO] Using v0.5.0 as release
[INFO] Downloading hash https://github.com/rancher/k3s/releases/download/v0.5.0/sha256sum-arm64.txt
[INFO] Downloading binary https://github.com/rancher/k3s/releases/download/v0.5.0/k3s-arm64
[INFO] Verifying binary download
[INFO] Installing k3s to /usr/local/bin/k3s
[INFO] Creating /usr/local/bin/kubectl symlink to k3s
[INFO] Creating /usr/local/bin/crictl symlink to k3s
[INFO] Creating killall script /usr/local/bin/k3s-killall.sh
[INFO] Creating uninstall script /usr/local/bin/k3s-uninstall.sh
[INFO] env: Creating environment file /etc/systemd/system/k3s.service.env
[INFO] systemd: Creating service file /etc/systemd/system/k3s.service
[INFO] systemd: Enabling k3s unit
Created symlink /etc/systemd/system/multi-user.target.wants/k3s.service → /etc/systemd/system/k3s.service.
[INFO] systemd: Starting k3s
```
wait 5 minutes...
`$ kubectl get all --all-namespaces`
```
nvidia@jetson-tx2:~$ kubectl get all --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system pod/coredns-695688789-c528g 1/1 Running 0 5m19s
kube-system pod/helm-install-traefik-fv9x4 0/1 Completed 0 5m20s
kube-system pod/svclb-traefik-47p8d 0/2 ContainerCreating 0 5m1s
kube-system pod/traefik-55bd9646fc-4qb5t 1/1 Running 0 5m1s
NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default service/kubernetes ClusterIP 10.43.0.1 <none> 443/TCP 5m32s
kube-system service/kube-dns ClusterIP 10.43.0.10 <none> 53/UDP,53/TCP,9153/TCP 5m32s
kube-system service/traefik LoadBalancer 10.43.243.236 <pending> 80:32018/TCP,443:30124/TCP 5m1s
NAMESPACE NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
kube-system daemonset.apps/svclb-traefik 1 1 0 1 0 <none> 5m1s
NAMESPACE NAME READY UP-TO-DATE AVAILABLE AGE
kube-system deployment.apps/coredns 1/1 1 1 5m32s
kube-system deployment.apps/traefik 1/1 1 1 5m1s
NAMESPACE NAME DESIRED CURRENT READY AGE
kube-system replicaset.apps/coredns-695688789 1 1 1 5m19s
kube-system replicaset.apps/traefik-55bd9646fc 1 1 1 5m1s
NAMESPACE NAME COMPLETIONS DURATION AGE
kube-system job.batch/helm-install-traefik 1/1 20s 5m32s
```
**Expected behavior**
`kube-system pod/svclb-traefik-47p8d 2/2 Running 0 5m1s`
**Additional context**
I tried the above steps on x86_64 machine and Raspberry Pi 3 B+ and it works fine.
Hi ShaneCCC,
Literally the post I used when requesting that the multiport module be added as part of the main build.
@ShaneCCC ,
I found [url]https://blog.hypriot.com/post/nvidia-jetson-nano-build-kernel-docker-optimized[/url] , which shows how to recompile the kernel with the requisite options. However, it would be SUPER NICE if this was included by default, and not something that users had to worry about compiling.
@aaronsachs