postman#@andrew.cmu.edu.UUCP (01/31/87)
ReSent-To:nntp-xmit#@andrew.cmu.edu Return-path: <dk1z#@andrew.cmu.edu> To: outnews#ext.nn.comp.unix.questions@andrew.cmu.edu (Outbound News) I have a 6250 bpi tar tape containing about 50MB of files. (About 5 years worth of college stuff.) Unfortunately, it seems that someone trying to read it in from me munged the header as every time I read it, I get an EOF immediately and no amount of effort can get past it. Also, the tape now seems to think it's a 1600 bpi tape and not a 6250 bpi tape. I've tried using dd to get past it, and I've tried using combinations of fseek, lseek, read, et all. No luck. Only thing I've not tried is physically cutting the header off and seeing what happens. As that is rather final, I'd like to know if there are any other things I could try. Are there any services out there that recover munged tapes? -David Kovar
ken@rochester.UUCP (02/01/87)
I would have mailed but I can't figure out your address. Try using mt to skip one file, then dd. It looks like somebody overwrote the first part of it. Some tape controllers have automatic density sensing so if you do any reads on it at all it will think a 1600 bpi tape was mounted. Ken
ggs@ulysses.UUCP (02/03/87)
In article <24358@rochester.ARPA>, ken@rochester.ARPA (SKY) writes: > I would have mailed but I can't figure out your address. > > Try using mt to skip one file, then dd. It looks like somebody > overwrote the first part of it. Some tape controllers have automatic > density sensing so if you do any reads on it at all it will think a > 1600 bpi tape was mounted. > > Ken If the 6250 BPI tape really had some crud written at 1600, you will probably have to write a new tape mark at 6250. The tape drives I have seen not only sense density, but refuse to change density from that of the first record. Since the density is encoded in a "density id burst" that preceeds the first record, a file skip won't help. Try : > /dev/high-density-auto-rewind-name (assuming [bk]sh syntax). The next trick will be to convince tar to read the munged file. I don't have any good solution to that one. If all else fails, dump the raw tar file onto a disk then edit out the tar headers. -- Griff Smith AT&T (Bell Laboratories), Murray Hill Phone: 1-201-582-7736 UUCP: {allegra|ihnp4}!ulysses!ggs Internet: ggs@ulysses.uucp
ken@rochester.ARPA (SKY) (02/04/87)
|(assuming [bk]sh syntax). The next trick will be to convince tar to |read the munged file. I don't have any good solution to that one. |If all else fails, dump the raw tar file onto a disk then edit out |the tar headers. That is the easier part. You just have snip off stuff to skip to a valid file header, at a 256 byte boundary. I had to do this once. You will lose the first one or few files, of course. Ken