[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#247054: Crypto-root patch updated to initrd-tools 0.1.70

"Wesley W. Terpstra" <terpstra@gkec.tu-darmstadt.de> - Tue, Nov 30, 2004:

> The debian patch made swsusp a module which could be loaded after swap was
> setup via whatever means were necessary. To the best of my knowledge, the
> swsusp restart present in the stock kernel can not be delayed like this.
> If I am wrong, I would be happy to be so. I used to use swsusp with crypto,
> and when the module got pulled I lost my ability to have quick power on/off.

 I frankly don't understand how that worked, but since the debian patch
 is gone, I suggest we discard this discussion.

> The devices listed were /dev/mapper/muffin-* which are LVM block devices.
> The 'dmname' parameter just suggests the new encrypted devices' name.

 [ This discussion might be a little off-topic, but I suggest we finish
 it nevertheless. ]

 Still, the fstab is for me a "device <-> mountpoint" table, and dm
 allows infinite cascading of dm, independently from any mount point,
 that's why I don't understand how you achieve a complete disparition of
 crypttab-like files.  I do understand how that works with LVM -- I
 believe this really ugly though -- but I'm not aware of a similar
 capability in dm: can it detect that /dev/hda1 and /deb/hdb1 are both
 part of a RAID 1 array, and that this array is a crypted device etc.?


Loïc Minier <lool@dooz.org>

Reply to: