Jan Johansson wrote:
And it is why I would never use software raid on a
production server!
Gee, then I wonder why I have had the exact same thing happen with hardware
RAID? When I upgraded a server from a 2.0 -> 2.2 kernel the cards got
initialized in the opposite order, and hence the volumes got interchanged.
Guess why there is the possibility to mount by label. :-)
And also by "UUID" at least for XFS.
I'm using labels as much as possibel nowadays.
Even automounting via autofs supports labels/UUIDs.
But as this doesn't(*) harmonise very well with encryption i guess this
is OT here. :-)
*:
If you seperate the losetup & mounting tasks then this should work.
But if you have more than one encrypted partition and use different keys
for each partition. And in case the keys are stored seperated from the
partition, you can't be sure you have the right combination until you
try to mount it or test otherwise that the combination was correct.
(When you use lables you can e.g. try to read the label)
So you need a bit of logic, same key, keys in front of the partition or
other method to bind the key to the device-content and not the
device-name for the losetup-part.
Bis denn
--
Real Programmers consider "what you see is what you get" to be just as
bad a concept in Text Editors as it is in women. No, the Real Programmer
wants a "you asked for it, you got it" text editor -- complicated,
cryptic, powerful, unforgiving, dangerous.
-
Linux-crypto: cryptography in and on the Linux system
Archive: http://mail.nl.linux.org/linux-crypto/
