nocloud deployment panics controller-runtime due to inability to parse network-config

[SeanWallace]

Bug Report

When attempting to deploy a nocloud cluster (in my case on a Harvester cluster) the network platform config controller will continually panic attempting to read the network-config from a cidata partition. This is despite the partition containing the correct file.

Description

I have not had the opportunity to recompile Talos to enable more debugging but by all appearances #9351 has not only added the requirement for cidata/network-config but it also doesn't seem to be able to find it even when it is there...at least in my case. I have mounted my cloud-init partition onto a different linux host and verified the necessary files are there and legal (Ubuntu is happy to parse and use it for example):

root@test:~# mount /dev/vdb /mnt
mount: /mnt: WARNING: source write-protected, mounted read-only.
root@test:~# cd /mnt
root@test:/mnt# ls
meta-data  network-config  user-data
root@test:/mnt# cat network-config
network:
  version: 2
  ethernets:
    eth0:
      dhcp4: false

In truth I actually have no interest in using the network-config part of cloud-init, I only want to use the user-data that contains my network config in my machine template.

I should also note this doesn't seem to prevent the cluster from becoming usable, after bootstrapping everything works perfectly fine, but the logs are spammed with this failure.

Logs

 user: warning: [2024-10-27T02:11:58.936377322Z]: [talos] waiting for devices to be ready...
 user: warning: [2024-10-27T02:11:58.979707322Z]: [talos] found config disk (cidata) at /dev/vdb
 kern:   debug: [2024-10-27T02:11:58.981563322Z]: ISO 9660 Extensions: Microsoft Joliet Level 3
 kern:   debug: [2024-10-27T02:11:58.982009322Z]: ISO 9660 Extensions: RRIP_1991A
 user: warning: [2024-10-27T02:11:58.982071322Z]: [talos] fetching meta config from: cidata/meta-data
 user: warning: [2024-10-27T02:11:58.983794322Z]: [talos] fetching network config from: cidata/network-config
 user: warning: [2024-10-27T02:11:58.985538322Z]: [talos] failed to read network-config
 user: warning: [2024-10-27T02:11:58.986687322Z]: [talos] fetching machine config from: cidata/user-data
 kern:    info: [2024-10-27T02:11:58.989573322Z]: init[2139]: segfault at 0 ip 0000000000f36d8a sp 000000c000b67c18 error 4 in init[400000+2837000] likely on CPU 3 (core 3, socket 0)
 kern:    info: [2024-10-27T02:11:58.992675322Z]: Code: 0f 10 44 24 70 41 0f 11 40 10 48 89 f0 48 8b 8c 24 d0 00 00 00 48 8b 54 24 40 48 8b 9c 24 80 00 00 00 4c 8b 84 24 b8 00 00 00 <4d> 8b 08 49 83 f9 01 0f 84 a2 00 00 00 49 83 f9 02 75 42 4c 89 c3
 user: warning: [2024-10-27T02:11:58.996799322Z]: [talos] platform panicked {"component": "controller-runtime", "controller": "network.PlatformConfigController", "stack": "github.com/siderolabs/talos/internal/app/machined/pkg/controllers/network.(*PlatformConfigController).runWithPanicHandler.func1\n\t/src/internal/app/machined/pkg/controllers/network/platform_config.go:564\nruntime.gopanic\n\t/toolchain/go/src/runtime/panic.go:770\nruntime.panicmem\n\t/toolchain/go/src/runtime/panic.go:261\nruntime.sigpanic\n\t/toolchain/go/src/runtime/signal_unix.go:881\ngithub.com/siderolabs/talos/internal/app/machined/pkg/runtime/v1alpha1/platform/nocloud(*Nocloud).ParseMetadata\n\t/src/internal/app/machined/pkg/runtime/v1alpha1/platform/nocloud/nocloud.go:54\ngithub.com/siderolabs/talos/internal/app/machined/pkg/runtime/v1alpha1/platform/nocloud(*Nocloud).NetworkConfiguration\n\t/src/internal/app/machined/pkg/runtime/v1alpha1/platform/nocloud/nocloud.go:136\ngithub.com/siderolabs/talos/internal/app/machined/pkg/contr...
 user: warning: [2024-10-27T02:11:59.012959322Z]: [talos] restarting platform network config {"component": "controller-runtime", "controller": "network.PlatformConfigController", "interval": "1m0.8004681s", "error": "panic: runtime error: invalid memory address or nil pointer dereference"}
 user: warning: [2024-10-27T02:11:59.912971322Z]: [talos] controller failed {"component": "controller-runtime", "controller": "k8s.KubeletStaticPodController", "error": "error refreshing pod status: error fetching pod status: an error on the server (\"Authorization error (user=apiserver-kubelet-client, verb=get, resource=nodes, subresource=proxy)\") has prevented the request from succeeding"}

Environment

• Talos version: v1.8.1
• Kubernetes version: v1.30.5
• Platform: Harvester