Although the actual problem is in kmod, I still think this change is
sane enough to stay in AIF, at least for now.....
-arno
On 8/14/2012 16:10, Wojciech Kusiak wrote:
On Tue, Aug 14, 2012 at 07:43:18AM +0200, Arno van Amersfoort wrote:
Thanks, I also changed it for modprobe_multi().
But you're right: kmod should really show some kind of error when it
fails to load a module, although the reasoning behind it, could be
that it somehow detects that the module is build-in and therefor
silences its error... Wouldn't hurt to report this to the maintainer
though.
And apparently, usually it does display the very same error message
as old modprobe did, so you may want to undo this -e '^$' check you added.
There's a bug in the kmod version that causes it to abort silently when
/lib/modules/$(uname -r)/modules.alias.bin doesn't exist.
Like when you're using a Xen system with monolithic kernel loaded from
outside of the VM's disk, which's how this machine is.
My fault, obviously, for not checking on a more "regular" configuration
first. I'm just too used to my VPSes, I guess...
I owe both you and Debian an apology here, for filing bug against wrong
package. Your script is a victim here. Sorry for wasting your time.
Opened a bug against kmod (Bug#684901), and well, waiting for the
consequences of my mistake.
Sincerely,
-- Wojciech
--
Arno van Amersfoort
E-mail : arn...@rocky.eld.leidenuniv.nl
Donations are welcome through Paypal!
---------------------------------------------------------------------------
Arno's (Linux IPTABLES Firewall) Homepage:
http://rocky.eld.leidenuniv.nl
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
