70-persistent-net.rules not created

jmscott at setex.ipcallback.com jmscott at setex.ipcallback.com
Tue Jan 12 08:23:22 PST 2010


On Tue, Jan 12, 2010 at 10:02:02AM -0500, linux fan wrote:
> On 1/12/10, jmscott at setex.ipcallback.com <jmscott at setex.ipcallback.com> wrote:
> > so, in the name of progress, i like to solict advice on what to do next,
> 
> LFS-6.4 and earlier had:
> for NIC in /sys/class/net/* ; do
>     INTERFACE=${NIC##*/} udevadm test --action=add --subsystem=net $NIC
> done
> 
> I wonder why subsystem was later removed.

not clear that adding --subsystem would force INTERFACE to be exported to
the /lib/udev/write_net_rules call.  i dont see a reference to the word
INTERFACE in a different text file underneatch /etc/udev, /lib/udev or the
udev-145 source tree, so it's not clear to me how INTERFACE ever got exported
in the first place, at least for udev-145.  udevadm seems concise 
about what is exported to the rule scripts.

right now, i don't even know how to hack a patch.
i thought perhaps an export rule could be added to the 
/etc/udev/rules.d/55-lfs.rules, but it's not clear to me how to 
write a udev rule to redefine an environment variable using 
the value of another environment variable.  i do see how
to define an environment variable as a constant string.
unfortunately, i'm uncomforatble making to many changes
to the lfs file system until i get my first kernel to boot.

this seems more like a udev-145 problem than an lfs problem.
clearly /lib/udev/write_net_rules is intended to be run from the
rules file, since /lib/udev/rules.d/75-persistent-net-generator.rules
references the script. so, i conclude, either another method exists
to export variables to scripts or this is just a bug in udev-145.

-j

> -- 
> http://linuxfromscratch.org/mailman/listinfo/lfs-support
> FAQ: http://www.linuxfromscratch.org/lfs/faq.html
> Unsubscribe: See the above information page



More information about the lfs-support mailing list