Hi David, It obviously took me way too long to look into this and respond. I am sorry about this.
If you are still around, it would be great if you could read on and give
me your feedback.
Firstly, version 2.2.0-881 was not an actual Debian package, i.e.
something that you installed from a Debian repository, or was it?
Secondly, did you by any chance try to do this with a standard Debian
kernel?
From your message, I am not 100% sure whether networking fails during
restore. From your mondoarchive parameters I understand that you back up
to optical media and the presumably restore from it, so you would not
need the nwetwork working during the restore run. I therefore beleive
that the issue is with networking not working in the restored Debian
installation because of the change in z25_persistent-net.rules. Is this
correct?
I have asked Bruno (Hi Bruno!) and here is what he pointed out regarding
the matter:
$ pbg z25_persistent-net
./mindi/mindi: rm
-f ./etc/udev/rules.d/z25_persistent-net.rules
IIRC this is a patch you provided ;-)
svn diff -r 1789:1790
Index: mindi/mindi
===================================================================
--- mindi/mindi (révision 1789)
+++ mindi/mindi (révision 1790)
@@ -2821,6 +2821,8 @@
echo "udev device manager found" > tmp/USE-UDEV
LogIt "udev device manager found"
cp --parents -Rdf /etc/udev . 2> /dev/null
+ # This avoids NIC remapping if on another machine at
restore time on Debian at least
+ rm -f ./etc/udev/rules.d/z25_persistent-net.rules
cp --parents -Rdf /lib/udev /lib64/udev . 2> /dev/null
if [ -x /sbin/udevd ]; then
lis2=`grep -Ev '^#' $MINDI_CONF/udev.files`
I am not actually sure that I provided this patch, although it rings a
very faint bell - anyway.
z25_persistent-net.rules stores the result of udev running
persistent-net-generator.rules so that it is preserved during reboots.
The idea behind deleting this would be that when restoring, especially
on totally different or at least somewhat changed hardware (e.g. new
mainboard and/or different network hardware when restoring, udev should
newly determine what the network interfaces are. This is generally a
good and correct approach I believe. So, we should really keep deleting
z25_persistent-net.rules as per the above patch I think.
However, we delete in the restore environment only and do not change the
udev config of the restored system. At least I think this is what we do
- Bruno, do you think this is right or am I overlooking something here?
With all the above said, would you be able to try and reproduce the
issue with the latest version in Debian unstable using a standard
kernel? This might give me enough information to successfully reproduce
the problem which I have not been able to (things work fine for me in
this department)
Cheers,
Andree
On Sat, 2007-05-12 at 07:23 -0500, mark david mcCreary wrote:
> Package: mondo
> Version: 2.2.0-881
>
> After backup up a Debian Etch system with a custom kernel, using
>
> mondoarchive -OVr -d /dev/hdb -9 -s 4200m -F
>
> and then restoring to a different, but similar machine, using
> the nuke option of mondorescue, networking fails.
>
> This file
>
> /etc/udev/rules.d/z25_persistent-net.rules
>
> has switched the eth0 to eth1.
>
> Additional files that may be broken include
>
> /etc/udev/rules.d/25-iftab.rules
> /etc/iftab
>
> although these last two files are not on my system.
>
>
--
Andree Leidenfrost
@ Debian Developer
Sydney - Australia
signature.asc
Description: This is a digitally signed message part
