Add a warning about unique machine-ids

[pvdputte]

No machines, especially controllers, should ever share machine-ids.

Cfr. #1547 (comment)

"The root cause seems to be that the controllers have matching machine IDs which k0s adds as --server-id."

Signed-off-by: pvdputte [email protected]

Description

I've been bitten by intermittent timeouts and general instability in my k0s testing with a HA control plane. It took me a long time to troubleshoot it. Eventually I found out that both my vagrant boxes and VMware templates were not resetting machine-id properly. A warning about the importance of having unique machine-ids when setting up HA could be helpful to others.

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

How Has This Been Tested?

• Manual test
• Auto test added

Checklist:

• My code follows the style guidelines of this project
• My commit messages are signed-off
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules
• I have checked my code and corrected any misspellings

[trawler]

The /etc/machine-id file used by kubernetes is generated by the OS kernel when a machine is booted. An identical machine-id would not just cause a problem with Konnektivity but also with services such as DHCP.
In any case, the problem with issue #1547 was caused due to a cloned VM, that was not cleaned properly: #1547 (comment)
Cloning a VM without cleaning up machine-id is a bad practice and has nothing to do with k0s.

[pvdputte]

Well, if it wasn't for these Konnectivity issues we were having, we wouldn't have noticed our machine-id issue for a very long time to come.

In fact the problem has been lurking in our images and templates for years. No problems with DHCP in Vagrant/VirtualBox and also no issues with hundreds of instances in the datacenter (static IPs).

If I had been less persistent about troubleshooting while evaluating k0s, I could have just told myself I should use an alternative because it's no good. It was by pure chance I eventually discovered #1547.

I'm adamant about running k0s in production eventually so this was more about adding some potentially helpful 'troubleshooting tips' to the docs. Indeed it has nothing to do with k0s itself, I guess others will just find their way to closed github issues as well then.

[pvdputte]

Cfr. k0sproject/k0sctl#435