LXC readings are wrong

[SamTV12345]

First of all thanks for the great crate. It works awesome on my dev machine.

Describe the bug
I am trying to use this crate in a Proxmox LXC. Unfortunately it doesn't report information correctly. I assigned 10 Gigabyte of SSD storage to the LXC but it reports 70 Gigabyte SSD storage. Also the memory is used from the host system not the LXC memory.

To Reproduce
Start an LXC container in Proxmox. Download docker-compose and docker.

Use this docker-compose:

version: '3'
services:
  podgrabv2:
    image: samuel19982/podgrabv2:latest
    ports:
      - "80:8000"
    volumes:
      - podgrab-podcasts:/app/podcasts
      - podgrab-db:/app/podcast.db
    environment:
      - POLLING_INTERVAL=60
      - SERVER_URL=http://<url to the server>

SERVER_URL is the ip of your machine. Go to http://SERVER_URL/ui. Select Info in the sidebar.

[GuillaumeGomez]

As this is a very specific issue, I don't plan to work on it myself. However, if you want to send a fix for it, it'd be very welcome!

[ThinkRedstone]

I ran into a similar issue with refresh_memory- the problem, at least with memory stats, is that refresh_memory uses read_u64 to read cgroups files; and read_u64 uses u64::from_str to parse the contents of the file. If we compare to the kernel source code, we can see that the kernel does not output a pure number like u64::from_str expects, but rather ends the number with a newline, causing read_u64 to incorrectly assume there's a problem with the cgroup file.