hwrep

HWRep gathers hardware information from the linux operating system.

Items of interest

Hostname

System:

• CPU cores
• CPU sockets
• number of threads
• CPU model info

Memory

• Amount of RAM
• Amount of SWAP
• NUMA layout

Storage

• drives
• capacity
• mount

Manufacturer info

• Bios/Bios ID
• Serial number
• ...

HWRep can display this information to the screen in plain text or a json format.

Why JSON?

With contemporary tools and languages we can easily parse the json object to extract details. example using jq

[vpalat@linux01 ~]$ ./hwrep -j  |jq '.cpu_info'
{
  "model": "Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz",
  "physical_cores": 10,
  "execution_units": 40,
  "threads_per_core": 2,
  "sockets": 2
}
[vpalat@linux01 ~]$ ./hwrep -j  |jq '.cpu_info.model'
"Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz"
[vpalat@linux01 ~]$ ./hwrep -j  |jq '.cpu_info.model' -r
Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz
[vpalat@linux01 ~]$ ./hwrep -j  |jq '.cpu_info.physical_cores' -r
10

Building

hwrep can be built as a static binary.

 rustup target add x86_64-unknown-linux-musl
 cargo build --target x86_64-unknown-linux-musl

Note you will also need a linker on MacOS.

brew install FiloSottile/musl-cross/musl-cross

Details on how to build for linux on a mac: https://timryan.org/2018/07/27/cross-compiling-linux-binaries-from-macos.html

Running hwrep

For the report:

./hwrep

In JSON:

./hwrep -j

Motivation

1. Inventory some older machines and really understand what we have.
2. Learn a little rust to build a CLI and web client.
3. Understand how linux represents the hardware.