gru panics when clearing cmos

[rustydb]

SUMMARY

Attempting to clear CMOS results in a traceback.

Target was an Nvidia Grace+Hopper (CPU+GPU) server.

ncn-m001:/mnt/developer/rusty # ./gru bios set --clear-cmos -j x9000c1s1b0
Asynchronously updating [    1] hosts ...
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0xb0 pc=0x90a339]

goroutine 18 [running]:
github.com/stmcginnis/gofish/redfish.(*Bios).ResetBios(...)
	/Users/rusty/gitstuffs/cray-shasta/gru/vendor/github.com/stmcginnis/gofish/redfish/bios.go:172
github.com/Cray-HPE/gru/pkg/cmd/cli/bios.resetBios({0x7ffdd4615a33?, 0x0?})
	/Users/rusty/gitstuffs/cray-shasta/gru/pkg/cmd/cli/bios/set.go:179 +0x39
github.com/Cray-HPE/gru/internal/set.AsyncCall.func1({0x7ffdd4615a33, 0xb})
	/Users/rusty/gitstuffs/cray-shasta/gru/internal/set/set.go:117 +0x67
created by github.com/Cray-HPE/gru/internal/set.AsyncCall in goroutine 1
	/Users/rusty/gitstuffs/cray-shasta/gru/internal/set/set.go:114 +0xd8
ncn-m001:/mnt/developer/rusty # ./gru --version
gru version 0.0.6
ncn-m001:/mnt/developer/rusty # ./gru show system -j x9000c1s1b0
Asynchronously querying [    1] hosts ...
{
  "x9000c1s1b0": {
    "biosVersion": "",
    "firmwareVersion": "",
    "processorModel": "TBD",
    "manufacturer": "HPE",
    "model": "HPE Cray EX254n",
    "serialNumber": "FP23080032"
  }
}
ncn-m001:/mnt/developer/rusty # ./gru show proc -j x9000c1s1b0
Asynchronously querying [    1] hosts ...
{
  "x9000c1s1b0": [
    {
      "architecture": "",
      "totalCores": 0,
      "model": "TBD",
      "socket": "",
      "threads": 0,
      "vendorID": ""
    },
    {
      "architecture": "ARM",
      "totalCores": 72,
      "model": "Grace A01",
      "socket": "CPU3",
      "threads": 72,
      "vendorID": "NVIDIA"
    },
    {
      "architecture": "ARM",
      "totalCores": 72,
      "model": "Grace A01",
      "socket": "CPU1",
      "threads": 72,
      "vendorID": "NVIDIA"
    },
    {
      "architecture": "",
      "totalCores": 0,
      "model": "TBD",
      "socket": "",
      "threads": 0,
      "vendorID": ""
    },
    {
      "architecture": "ARM",
      "totalCores": 72,
      "model": "Grace A01",
      "socket": "CPU0",
      "threads": 72,
      "vendorID": "NVIDIA"
    },
    {
      "architecture": "",
      "totalCores": 0,
      "model": "TBD",
      "socket": "",
      "threads": 0,
      "vendorID": ""
    },
    {
      "architecture": "",
      "totalCores": 0,
      "model": "TBD",
      "socket": "",
      "threads": 0,
      "vendorID": ""
    },
    {
      "architecture": "ARM",
      "totalCores": 72,
      "model": "Grace A01",
      "socket": "CPU2",
      "threads": 72,
      "vendorID": "NVIDIA"
    }
  ]
}

ISSUE TYPE

• Bug Report

STEPS TO REPRODUCE

1. Invoke gru bios set --clear-cmos <target>

EXPECTED RESULTS

A success or error message should return after a CMOS clear attempt.

ACTUAL RESULTS

Gru panics.