Shouldn't this "well known" bug in dm-crypt be fixed in both sarge and etch? Isn't fs corruption release critical? For others, see http://lists.debian.org/debian-kernel/2006/12/msg00099.html which points to http://www.spinics.net/lists/dm-crypt/msg00481.html Someone should ask if it should be queued for 2.6.16 too. Thanks, Drew Daniels Resume: http://www.boxheap.net/ddaniels/resume.html