Xref: utzoo comp.unix.admin:1550 comp.sys.sgi:9373 Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!decwrl!sgi!jeremy@perf2.asd.sgi.com From: jeremy@perf2.asd.sgi.com (Jeremy Higdon) Newsgroups: comp.unix.admin,comp.sys.sgi Subject: Re: How do I read a bad tape (tar)? Keywords: tar tape sgi unix Message-ID: <96501@sgi.sgi.com> Date: 10 Apr 91 03:15:49 GMT References: <1991Apr8.231456.391@gpu.utcs.utoronto.ca> <1991Apr10.004953.17211@leland.Stanford.EDU> Sender: guest@sgi.sgi.com Organization: Silicon Graphics, Inc., Mountain View, CA Lines: 29 In article <1991Apr10.004953.17211@leland.Stanford.EDU>, dhinds@elaine19.Stanford.EDU (David Hinds) writes: > In article <1991Apr9.225311.6534@lokkur.dexter.mi.us> scs@lokkur.dexter.mi.us (Steve Simmons) writes: > >olson@anchor.esd.sgi.com (Dave Olson) writes: > > > >Hmmm...some help might be possible here. First, one needs to trick > >the drive into reading past the EOT (not EOD) marker. This *can* be > >done with some drives (I've done it), as long as one is careful. An > >EOT marker is two tape marks. So *if* the conjecture about a small > >tar blotzing the head of a large one is true, one could do so by doing > > > > mt -f fsf 1 ; mt -f fsf 1 > > > >Now the tape is positioned past the EOT mark. Use GNU tar to read the > >damaged data. > > I didn't realize you could just step past the tape marks like this. You can do this with a half-inch tape, or sometimes, with an Exabyte. The original poster had a cartridge tape, so I doubt that the above will work. > What I've done before, to get past an EOD, is to rewrite another small > archive over the one causing the problem, but then eject the tape from > the drive before it writes its EOD. Then, reading the tape runs into > a media error, but hopefully tar should try to resynchronize and read > the rest of the tape. > > -David Hinds > dhinds@cb-iris.stanford.edu