Re: Dead parallel port (/dev messup)

On 2006-07-12, Dances With Crows <danSPANceswitTRAPhcrows@xxxxxxxxx> wrote:
On Wed, 12 Jul 2006 19:10:32 +0200, kaczfan staggered into the Black Sun
and said:
I have ACCIDENTALY erased all the contents of /dev directory

This is yet another reason to use udev. Device nodes are dynamically
created when udev's running, so it's harder to accidentally rm -f /dev/*

Yeah, i'd probably have to add a cron task to do it automatically :>
It's definitely better solution, but I'm short on time and don't
want to play with kernel configutation now.

I have eventually copied the whole structure from some bootable cd and
regenerated missing nodes, BUT I have a problem with the parallel port.
when I echo to /dev/parport0 or /dev/lp0, I get
bash: /dev/parport0: no such device

Do you have parport, parport_pc , and lp loaded? "modprobe lp". This
module should be modprobed automagically whenever a char device with
major 6 is accessed in any way, but if your modules.conf (or
modprobe.conf) file is mangled, that may not be happening. The module
required for "raw parport access" isn't lp, but something else.
Modprobe that one too.

I've followed your advice (the module wouldn't load automatically,
although modules.conf and modprobe.conf don't seem suspicious) and
the result is that now echoing to lp0 works fine, but parport0 still
results with 'no such device'

c 6 0 for lp0 (and, as I have just found out, /dev/par0)
c 99 0 for parport0

Is it significant that there are (at least) two files linked to the
same node?

Nah. If you have /dev/blah (b 3 1) and /dev/foo (b 3 1) , both device
nodes refer to the same physical device (usually known as /dev/hda1).
This is either a Feature or confusing, depending on your point of view.

Well, this seems reasonable to me (would be far more confusing if
it could cause the malfunction)

Thanks
.

Quantcast