Hi,
Michael Tross wrote:
Hi,
Am Montag, den 04.08.2008, 16:20 +0200 schrieb René Rebe:
Hi,
Adam Stirk wrote:
Hi René,
I have attached my fix for the dhcp package, I don't know whether the is the
most standard/upstream conforming solution but it works. I have added a
path.path which fixes the problem with the paths sent to the dhclient-script
(these may need looking at I set them the same as the v3 package). I have also
moved the dhclient-script to the /etc folder as dhclient only looks here for
the file (doesn't use the $PATH anymore, it could be moved back to the bin
folder with some configuration changes). I have also added a test in the
rocknet script to see if the ipv6 functionality has been loaded if not it loads
the module. (Again this might be better put into the kernel instead of a module
so that it is always there as more apps are designed with ipv6 in mind)
I have also included a patch for sysvinit as setterm has moved from /usr/bin to
/bin.
Thanks for the patches.
I applied the dhcp stuff with variations: avoid the ucs in the PATH and try
sbin before bin and removed filling the site.conf which apparently it no
longer used by the new, autoconfiscated Makefiles.
Committed revision 29896.
[ ...]
this solution has disadvantages and does not work out-of-the-box, as the
default dhclient.conf is invalid, and now dhcp relies on a correct entry
for the dhclient-script, like 'script "/etc/dhclient-script";'.
The original error stems from the fact that bash does not have a default
path that includes /sbin and therefore dhclient cannot exec the
dhclient-script. The attached patch for bash fixes the default path.
I think it would be better if dhclient works without an existing
dhclient.conf, so we should revert r29896.
I would like to mess with PATH default values, I hope the packages rather
have this (e.g. no sbin) for some POSIX reason.
r29896 does not look too bad, I agree it should work without an additional
script / config, but I would rather like to have the dhcp package to be
adapted than every possible shell (pdksh, dash, etc.?) being adapted
just for the dhcp-client ...
Also DHCP 3.x just worked, we should just fix the last bits of DHCP 4.x
to just work (tm).
Yours,
--
René Rebe - ExactCODE GmbH - Europe, Germany, Berlin
Geschäftsführer: Susanne Klaus, René Rebe
Sitz: Berlin, Amtsgericht Charlottenburg HRB 105 123 B
USt-IdNr.: DE251602478
http://exactcode.de | http://t2-project.org | http://rene.rebe.name
-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
[EMAIL PROTECTED] with a subject of: unsubscribe t2
