Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.networking > #1939 > unrolled thread

ifconfig(8) vs. iproute

Back to article view | Back to comp.os.linux.networking

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  ifconfig(8) vs. iproute Ivan Shmakov <oneingray@gmail.com> - 2013-02-09 17:18 +0000
    Re: ifconfig(8) vs. iproute Jorgen Grahn <grahn+nntp@snipabacken.se> - 2013-02-09 18:36 +0000
    Re: ifconfig(8) vs. iproute J G Miller <miller@yoyo.ORG> - 2013-02-09 18:55 +0000

#1939 — ifconfig(8) vs. iproute

>>>>> J G Miller <miller@yoyo.ORG> writes:
>>>>> Pascal Hambourg explained:

	[Cross-posting to news:comp.os.linux.networking, and setting
	Followup-To: there.]

 >> All these *except eth0:1* are indeed examples of virtual interfaces.

 > Is *alias* the correct name for eth0:{0,1,2,3,...} eth1:{0,1,2,3...}
 > etc?

 > From the manual page for ifconfig

	... Isn't ifconfig considered obsolete these days, in favor of
	the almighty ip(8)?  And, FWIW, ip(8) doesn't seem to contain
	the word "alias."  (And neither does its output; see below.)

	AIUI, initially, Linux (the kernel) allowed just one IPv4
	address per interface.  Thus, the notion of "aliases" -- several
	"logical" network interface per single "physical" one -- was
	introduced.  With the kernel's IP implementation extended to
	handle the case in an (arguably) better and more logical
	fashion, the very notion of an "interface alias" became
	obsolete.  (And, as was already noted in this thread, these
	weren't "first-class" interfaces to begin with; for instance, I
	doubt that one'd use such an "alias" in the route(8) command.)

$ ip addr list 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000
    link/ether f0:12:34:56:78:9a brd ff:ff:ff:ff:ff:ff
    inet 192.0.2.34/24 brd 192.0.2.255 scope global eth5
    inet6 2001:db8:1337::cafe/64 scope global 
       valid_lft forever preferred_lft forever
    inet6 2001:db8:2001::da:42/64 scope global 
       valid_lft forever preferred_lft forever
...
$ 

[...]

-- 
FSF associate member #7257

[toc] | [next] | [standalone]

#1940

On Sat, 2013-02-09, Ivan Shmakov wrote:
>>>>>> J G Miller <miller@yoyo.ORG> writes:
>>>>>> Pascal Hambourg explained:
>
> 	[Cross-posting to news:comp.os.linux.networking, and setting
> 	Followup-To: there.]
>
> >> All these *except eth0:1* are indeed examples of virtual interfaces.
>
> > Is *alias* the correct name for eth0:{0,1,2,3,...} eth1:{0,1,2,3...}
> > etc?
>
> > From the manual page for ifconfig
>
> 	... Isn't ifconfig considered obsolete these days, in favor of
> 	the almighty ip(8)?

Some people claim it is.  What it boils down to is whether the people
who maintain ifconfig are dedicated to improve and extend it over
time, or if they just keep it running.

ip(8) /is/ IME more aligned with today's Linux networking support than
ifconfig, route and the rest of net-tools (the name of the Debian
package).  A bit unfortunate -- almost everyone I know use the latter
tools by habit, and so they miss out on features.

/Jorgen

-- 
  // Jorgen Grahn <grahn@  Oo  o.   .     .
\X/     snipabacken.se>   O  o   .

[toc] | [prev] | [next] | [standalone]

#1941

On Saturday, February 9th, 2013, at 17:18:40h +0000,
Ivan Shmakov asked:

> 	... Isn't ifconfig considered obsolete these days

Probably, but Debian/Ubuntu/Mint appear to be still using it.

       ./if-pre-up.d/wireless-tools:      ifconfig "$IFACE" up

[toc] | [prev] | [standalone]

Back to top | Article view | comp.os.linux.networking

csiph-web