On Wed, Nov 22, 2017 at 09:25:18PM -0500, Taiidan@??? wrote:
> What is so much better about the ifconfig replacement ip? Why should I learn
> how to use yet another tool that has no tangible benefit but is being
> foisted on me?

Because ifconfig is broken. It works only in simplest cases, and it can't
even detect when it is wrong.

>From the technical side, net-tools use a different set of APIs than iproute
(BSD ioctls vs netlink), the former are not extensible. Well, one could
make a complete new set of ioctls, but that's quite pointless.

Likewise, the user-facing format is different, as net-tools can't express
the data:

netstat:
tcp6       0      0 2001:470:64f4::6:22     2001:470:64f4:0:edd:efff:fefc:2ad7:64167 ESTABLISHED 27287/sshd: kilobyt

ss:
tcp   ESTAB      0      0        2001:470:64f4::6:22                 2001:470:64f4:0:edd:efff:fefc:2ad7:64167               users:(("sshd",pid=27296,fd=3),("sshd",pid=27287,fd=3))

The former is incomplete: shows only one process that uses this socket,
truncates fields, and fails to escape user-controlled strings it prints.


Here's a car analogy: imagine a car that can go only over single-lane roads,
and will crash if you try to drive over a multi-lane one. Such a car was
adequate in 1920, or in a good part of today's Poland, but in a civilized
country you won't make it father than a couple blocks.


> Would I be correct in guessing it is made by red-hat?

No.


Meow!
--
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢰⠒⠀⣿⡁ Imagine there are bandits in your house, your kid is bleeding out,
⢿⡄⠘⠷⠚⠋⠀ the house is on fire, and seven big-ass trumpets are playing in the
⠈⠳⣄⠀⠀⠀⠀ sky. Your cat demands food. The priority should be obvious...