Add some more "fine print" to the installation requirements

[afbjorklund]

Somebody recently tried to install minikube on x86 Alpine Linux:

#10805

This failed, for two different reasons:

• the regular minikube binary requires GNU C Library
• we don't have any 32-bit binaries for x86 available

So we probably need to make it clear that those are required.

I don't think we want to do static builds, and k8s doesn't do i686

[afbjorklund]

Similarly, we should document that armv7 is needed for ARM.
(there are no binaries available anymore, for armv6 and older)

RPi	CPU	RAM
0	ARMv6	512M
1	ARMv6	512M
2	ARMv7	1G
3	ARMv8	1G
4	ARMv8	1-8G

And probably also recommend running arm64, if at all possible ?
https://www.raspberrypi.org/blog/latest-raspberry-pi-os-update-may-2020/

Along with yesterday’s launch of the new 8GB Raspberry Pi 4, we launched a beta 64-bit ARM version of Debian with the Raspberry Pi Desktop, so you could use all those extra gigabytes. We also updated the 32-bit version of Raspberry Pi OS (the new name for Raspbian)

https://www.raspberrypi.org/forums/viewtopic.php?f=117&t=275370

We still recommend the 32 bit operating system for all Pis at this time, although have decided it is now time to begin the move toward a 64bit OS. For the moment this is a 'beta' program, the OS is in heavy flux and its functionality is likely to change significantly over the next few months.

See https://www.raspberrypi.org/software/operating-systems/

The recommendation is https://ubuntu.com/download/raspberry-pi

[afbjorklund]

Even the most innocent statements are still ambiguous:

https://minikube.sigs.k8s.io/docs/start/

What you’ll need

• 2 CPUs or more
• 2GB of free memory
• 20GB of free disk space

Would that be "cores", or "vCPU" (threads) ? And is it "megabytes" or "mebibytes" ?

The traditional interpretation has used 2 vCPU, 2048 MiB mem and 20,000 MiB disk.

🔥  Creating virtualbox VM (CPUs=2, Memory=2048MB, Disk=20000MB) ...

That normally works out to around 600 MiB memory and 15,000 MiB disk available.

                         _             _            
            _         _ ( )           ( )           
  ___ ___  (_)  ___  (_)| |/')  _   _ | |_      __  
/' _ ` _ `\| |/' _ `\| || , <  ( ) ( )| '_`\  /'__`\
| ( ) ( ) || || ( ) || || |\`\ | (_) || |_) )(  ___/
(_) (_) (_)(_)(_) (_)(_)(_) (_)`\___/'(_,__/'`\____)

$ free -m
              total        used        free      shared  buff/cache   available
Mem:           1985         613          79         611        1292         609
Swap:             0           0           0
$ df -m / /dev/sda1
Filesystem     1M-blocks  Used Available Use% Mounted on
tmpfs               1788   595      1193  34% /
/dev/sda1          17368  1324     15029   9% /data

Most people would like to double these, to 4 vCPU and 4 GiB mem and 32-40 GiB disk.

Probably also should mention that these numbers are per node, so if you want multi-node...

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle rotten

[sharifelgamal]

I'm still in favor of doing this, so long as our documentation doesn't get too cluttered.