Re: Bug#404760: closed: fixed in inetutils 2:1.8-1
On Mon, 06 Dec 2010 at 05:05:54 +0100, Guillem Jover wrote:
> I guess the inetd se_v4mapped logical inversion fix and the “ping -w”
> support, both from upstream 1.8, would be important to have.
My backport of making tcp/udp be v4-only already included the inversion fix
as part of the conflict resolution, in fact. I've added the ping -w patch and
put an updated proto-NMU here:
http://git.debian.org/?p=users/smcv/qa/inetutils.git;a=shortlog;h=refs/heads/squeeze
I've done some basic testing, but I don't really know what to look for in this
package, so I'm not really comfortable with NMUing this without some review,
and to be honest I'd prefer a maintainer upload. (I also don't have a kFreeBSD
machine around to test that aspect of it.)
While smoke-testing it I did notice http://bugs.debian.org/559744 (the escape
character in telnet doesn't work), which seems pretty glaring, and has been
open for a year. I'm somewhat surprised anyone uses this variant of telnet
with that bug present (it'd certainly drive me mad), particularly with
netkit-telnet and telnet-ssl both available on all release architectures too;
does the inetutils version have any killer advantages?
A lot of the binary packages in inetutils don't seem to have any compelling
advantages over their higher-package-priority counterparts, in fact; if some
of them don't work very well, might it be worth dropping some binary packages?
The only thing in inetutils that seems to be particularly important is the
ping implementation, for kFreeBSD's benefit (because iputils is Linux-only).
> But then
> at that point the 1.6 Debian release would be the same as the one in
> unstable except for the indentation changes, the header cleanups, the
> unconditionalization of free() calls and the argp switch
... none of which really sound like changes to make during a freeze.
Regards,
Simon
Reply to: