netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Volkan Yazıcı" <volkan.yazici@gmail.com>
To: netdev@vger.kernel.org
Subject: Deleting an alias causes rest to get deleted
Date: Mon, 02 Jul 2012 22:54:28 +0300	[thread overview]
Message-ID: <4FF1FC74.8080401@gmail.com> (raw)

Hi!

I observe an IP aliasing anomaly that occurs when I try to delete an IP 
alias from an interface. That is, when I delete the first address in a 
set of IP aliased addresses assigned according to a particular subnet, 
rest of the aliases get deleted as well. Check out the below snippet.

    $ *for I in `seq 1 6`; do sudo ip addr add 192.168.2.$I/29 dev eth0;
    done*
    $ 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
    state UP qlen 1000
         link/ether 00:24:54:b9:1c:f8 brd ff:ff:ff:ff:ff:ff
    *inet 192.168.1.200/24 brd 192.168.1.255 scope global eth0**
         inet 192.168.2.1/29 scope global eth0
         inet 192.168.2.2/29 scope global secondary eth0
         inet 192.168.2.3/29 scope global secondary eth0
         inet 192.168.2.4/29 scope global secondary eth0
         inet 192.168.2.5/29 scope global secondary eth0
         inet 192.168.2.6/29 scope global secondary eth0*
         inet6 fe80::224:54ff:feb9:1cf8/64 scope link
            valid_lft forever preferred_lft forever
    3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
         link/ether e8:39:df:6a:21:2a brd ff:ff:ff:ff:ff:ff
    $ *sudo ip addr del 192.168.2.1/29 dev eth0*
    $ 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
    state UP qlen 1000
         link/ether 00:24:54:b9:1c:f8 brd ff:ff:ff:ff:ff:ff
    *inet 192.168.1.200/24 brd 192.168.1.255 scope global eth0*
         inet6 fe80::224:54ff:feb9:1cf8/64 scope link
            valid_lft forever preferred_lft forever
    3: wlan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
         link/ether e8:39:df:6a:21:2a brd ff:ff:ff:ff:ff:ff

Per see, deleting 192.168.2.1/29 causes the rest of the aliased 
interfaces get deleted as well. This is something that is slightly 
documented in the ifconfig manual: /for every scope (i.e. same net with 
address/netmask combination) all aliases are deleted, if you delete the 
first (primary)/. So what is the right way of just deleting the first 
(primary) alias without affecting the rest? If this is a scoping issue, 
is it possible to assign each alias as primary within its own dedicated 
scope?

As a side note, when I first asked this question to Stephen Hemminger 
(he forwarded me to this mailing list) he also told me that "/In Linux 
the interface aliases are really a legacy from the BSD style addressing, 
and don't act the same. It is not common practice to use them./" Is that 
really the case? Because, as you know, IP aliasing is the heart of a 
majority of the high-availability and clustering solutions in Linux. Is 
IP aliasing a really deprecated technology in Linux? Should we avoid 
using it? If so, what do you recommend as an alternative?


Best.

             reply	other threads:[~2012-07-02 20:03 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-02 19:54 Volkan Yazıcı [this message]
2012-07-03  1:12 ` Deleting an alias causes rest to get deleted Ben Hutchings
2012-07-03  2:44   ` Stephen Hemminger
2012-07-03  6:34 ` Julian Anastasov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4FF1FC74.8080401@gmail.com \
    --to=volkan.yazici@gmail.com \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).