Re: [iproute2] Bug#898840: Latest update breaks ip6 default gateway cli api

Re: [iproute2] Bug#898840: Latest update breaks ip6 default gateway cli api

[Luca Boccassi]

[-- Attachment #1: Type: text/plain, Size: 1439 bytes --]

On Wed, 2018-05-16 at 14:26 +0200, Hans van Kranenburg wrote:
> Package: iproute2
> Version: 4.16.0-2
> Severity: normal
> 
> Hi,
> 
> The last iproute2 update has a backwards incompatible change in
> setting
> IPv6 default routes, breaking existing configuration and scripts.
> 
> Previously, the following was possible, and now it requires an
> explicit
> -6 option to be added:
> 
> -# ip route add default via 2001:db8::1 dev eth0
> Error: inet address is expected rather than "2001:db8::1".
> 
> This works:
> -# ip -6 route add default via 2001:db8::1 dev eth0
> 
> I found out after having systems end up being unreachable after a
> reboot, because I have commands like these in network/interfaces.
> 
> I had a look at upstream changelogs, but I don't see any mention of
> this, and suspect it was not intentional. However, it's bad.

Hello Serhey and Stephen,

Hans reported a regression in v4.16.0, ip route now requires -6 to be
manually added when using v6 addresses while up to 4.15 it didn't, the
commands quoted show the problem.

Bisecting shows that the following commit from Serhey introduced the
problem:

93fa12418dc6f5943692250244be303bb162175b
utils: Always specify family and ->bytelen in get_prefix_1()

Could you please have a look when you have a moment? It's very easy to
reproduce, and it breaks existing scripts and so on.

Thanks!

-- 
Kind regards,
Luca Boccassi

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [iproute2] Bug#898840: Latest update breaks ip6 default gateway cli api

[Michal Kubecek]

On Wed, May 16, 2018 at 02:42:24PM +0100, Luca Boccassi wrote:
> Hans reported a regression in v4.16.0, ip route now requires -6 to be
> manually added when using v6 addresses while up to 4.15 it didn't, the
> commands quoted show the problem.
> 
> Bisecting shows that the following commit from Serhey introduced the
> problem:
> 
> 93fa12418dc6f5943692250244be303bb162175b
> utils: Always specify family and ->bytelen in get_prefix_1()
> 
> Could you please have a look when you have a moment? It's very easy to
> reproduce, and it breaks existing scripts and so on.

Fixed already:

------------------------------------------------------------------------------
mike@unicorn:~/work/git/iproute2> git --no-pager log --grep 93fa12418dc6
commit d42c7891d26e4d5616a55aac9fe10813767fcf9c
Author: David Ahern <dsahern@gmail.com>
Date:   Fri Apr 13 09:36:33 2018 -0700

    utils: Do not reset family for default, any, all addresses
    
    Thomas reported a change in behavior with respect to autodectecting
    address families. Specifically, 'ip ro add default via fe80::1'
    syntax was failing to treat fe80::1 as an IPv6 address as it did in
    prior releases. The root causes appears to be a change in family when
    the default keyword is parsed.
    
    'default', 'any' and 'all' are relevant outside of AF_INET. Leave the
    family arg as is for these when setting addr.
    
    Fixes: 93fa12418dc6 ("utils: Always specify family and ->bytelen in get_prefix_1()")
    Reported-by: Thomas Deutschmann <whissi@gentoo.org>
    Signed-off-by: David Ahern <dsahern@gmail.com>
    Cc: Serhey Popovych <serhe.popovych@gmail.com>
------------------------------------------------------------------------------

Michal Kubecek

Re: [iproute2] Bug#898840: Latest update breaks ip6 default gateway cli api

[Luca Boccassi]

[-- Attachment #1: Type: text/plain, Size: 789 bytes --]

On Wed, 2018-05-16 at 16:53 +0200, Michal Kubecek wrote:
> On Wed, May 16, 2018 at 02:42:24PM +0100, Luca Boccassi wrote:
> > Hans reported a regression in v4.16.0, ip route now requires -6 to
> > be
> > manually added when using v6 addresses while up to 4.15 it didn't,
> > the
> > commands quoted show the problem.
> > 
> > Bisecting shows that the following commit from Serhey introduced
> > the
> > problem:
> > 
> > 93fa12418dc6f5943692250244be303bb162175b
> > utils: Always specify family and ->bytelen in get_prefix_1()
> > 
> > Could you please have a look when you have a moment? It's very easy
> > to
> > reproduce, and it breaks existing scripts and so on.
> 
> Fixed already:

Ah fantastic, thanks, sorry I missed that.

-- 
Kind regards,
Luca Boccassi

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]