There is an "undo"... :) john ===================================================================== john meister sr CAD sys admin, Intermec ( http://www.wagoneers.com/ john@wagoneerscom ) ===================================================================== # -----Original Message----- # Sent: Thursday, July 15, 1999 11:39 AM # Subject: [Fwd: Unix Undelete command] # # Subject: # Re: Unix Undelete command # Date: Mon, 5 Jul 1999 12:03:45 -0400 # From: Barrett Richardson (email removed to prevent spam) # Newsgroups: comp.unix.admin # # Well, If you are desparate and have a larger extra disk you can throw # on the the system, you can dd the raw partition to files on a file # system on the spare disk. It more manageable if it is split in to # uniform size chunks (< 10 mb). # # You sequentially run the dd till you run out of chunks. # # dd if=/dev/rsd01g of=/spare/part1 bs=8192 count=1024 # dd if=/dev/rsd01g of=/spare/part2 bs=8192 count=1024 skip=1024 # dd if=/dev/rsd01g of=/spare/part3 bs=8192 count-1024 skip=2048 # # You *MUST* do this before much disk writing happens on this file system # because the disk blocks that your data resides on are fair game for # reuse when blocks get allocated to newly created and growing files. Once # you have all the data in chunks you can use the strings command to look # for familiar patterns of data (i.e. 'strings part1 | grep # some_pattern') # if the lost file is a text file. Once it is ascertained wich chunk # the lost file is in, you can use the old familiar hammer (vi) and # save the lost portion to a file. # # If you cant scrounge enough disk space you can do it kind of # piece meal. # Save a chunk to a different file system, scan it for your lost data, # toss it if it doesn't have it, repeat for next chunk. # # - # # Barrett # -- # ======================================= # | john meister | # | sr CAD sys admin | # | intermec technologies | # | everett, wa | # | http://www.wagoneers.com | # | john@wagoneerscom | # ======================================= #