On Tue, Sep 28, 2021 at 09:41:22AM +0200, tomas@tuxteam.de wrote:
> On Mon, Sep 27, 2021 at 04:47:07PM -0400, Henning Follmann wrote:
>
> [...]
>
> > And N-M is not "buggy". [...]
>
> Uh-huh.
>
What a great argument!
But I play along.
Are there bugs filed against N-M? Yes there are!
Are there reasons why people have issues with N-M? You bet!
Up until Jessie N-M had a lot of issues and missing features, which led to
....
But since buster, if you trust the work and knowledge of the
package maintainers N-M results in reliable network configurations
when we are talking about a default desktop setup. Most issues here
are mainly related to propietary drivers from various chip
manufacturers and not an N-M issue.
But please stress your comment "when we are talking about a default desktop setup". In the non-desktop server environment, N-M is disabled. Indeed it is a type of risk. Last times I've been on AWS and Google cloud, their VM instances also remove N-M. Same reason.
In these environments wifi does not exist, so network connections should be known ahead of time, and a fixed set.
My comment to the OP was basically on the nebulous source (most VPN Providers)
and the generalized categorization (N-M is buggy), which I disagree with.
That is not the view that I have heard expressed.
N-M is a 1-size-fits-all approach to network setup that only fits 1 use-case. But we all get to have it :-)
Some developers would call it "over engineered". That I agree with.