Hi Stuart, (1) Absolutely - journaling filesystems depend upon the maintenance of the order of device write operations. Where this is not the case, file system recovery based upon the journal has the potential to be very dangerous. (2) The "PS. Kernel 2.6" comment was made by Majek in his original e-mail. The implications of ordered device writes vs journaling filesystems apply across the board including 2.6. The potential problems are inherent to the the very flexibility of a loop device - you can potentially construct looped filesystems n-levels deep. BTW, I cannot claim the arguments as "mine" or take credit :-) Happy to accept the blame for all errors. Comments/clarifications welcome. D. -- Daniel Harvey <daniel@xxxxxxxxxxxxxxx> On Wed, 2004-10-20 at 00:53, IT3 Stuart B. Tener, USNR wrote: > Mr. Daniel Harvey: > > Referencing your comments of 18 OCT 2004, I comment as follows: > > Based on your argument, is it presumable that using a journaling filesystem > on a file backed loop device (even if you are not using loop-aes, or any > encryption mechanism) can be quite a dangerous, correct? > > By the way at the bottom of your comments, you say "PS. Kernel 2.6", does > that mean that the issue you speak of does not apply in Kernel 2.6 as it has > been remediated in that kernel version? > > > Very Respectfully, > > IT3 Stuart Blake Tener, USNR > Beverly Hills, California > > NR SPAWAR 0886 > NAF WASHINGTON, ANDREWS AFB, MD > > Email: teners@xxxxxxxxxxx > NIPR : stuart.tener@xxxxxxxx > CNRF : tenersb@xxxxxxxxxxxxxxxx > ARMY : stuart.tener@xxxxxxxxxxx > > Amateur/HAM Radio Callsign: N3GWG > > west coast: (310)-358-0202 P.O. Box 16043, Beverly Hills, CA 90209-2043 > east coast: (215)-338-6005 P.O. Box 45859, Philadelphia, PA 19149-5859 > Telecopier: (419)-715-6073 fax to email gateway via www.efax.com > > JOIN THE US NAVY RESERVE, SERVE YOUR COUNTRY, AND BENEFIT FROM IT ALL. > > Tuesday, October 19, 2004 12:46 PM > > Subject: Re: journaling file system > From: Daniel Harvey <daniel@xxxxxxxxxxxxxxx> > Date: Mon, 18 Oct 2004 08:22:57 +0800 > > >>From Jari's README for loop-aes: > > > 2.2. Use of journaling file systems on loop device > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Don't use a journaling file system on top of file backed loop device. Device > backed loop device can be used with journaling file systems as device backed > loops guarantee that writes reach disk platters in order required by > journaling file system (write caching must be disabled on the disk drive, of > course). With file backed loop devices, correct write ordering may extend > only to page cache (which resides in RAM) of underlying file system. VM can > write such pages to disk in any order it wishes, and thus break write order > expectation of journaling file system. > > Daniel. > > On Mon, 2004-10-18 at 03:56, majek04 wrote: > > Hi. > > > > Is there any possibility of running journaling file system on the top > > of crypted partition or file? It doesn't have to be loop-aes... > > > > Yours sincerely > > majek04 > > > > PS. kernel 2.6 > > > > > > - > > Linux-crypto: cryptography in and on the Linux system > > Archive: http://mail.nl.linux.org/linux-crypto/ - Linux-crypto: cryptography in and on the Linux system Archive: http://mail.nl.linux.org/linux-crypto/
