Magnetic tape archiving (was Disk archival techniques)

[Stan Sieler]

Re:
> The big thing missing from the '.tap' format is it assumes all the blocks are
> good. There is no way to describe a partially read block, or to encode any of
> the CRC or LRC check data that was in the tape block on the physical medium.

Ties in with problems I posted about last year in .tap :)

I have my own tape archiver (written on an HP 3000 under MPE/iX).

For each record read I record:

   # bytes

   any error information (if any)

   Setmark (if any) (for DDS)

   EOF (if any)

   EOT (if any)

(And, I optionally compress the tape data, on a per-record basis,
to minimize .tap file size.)

At the start of the .tap file is a header, which records
the maximum read size I was issuing ... that's *important*
beause with many tape devices, if you have a record of N
bytes and ask to read K (where K < N) bytes, you'll silently
lose data.

Stan
-- 
Stan Sieler
sieler at allegro.com
www.allegro.com/sieler/wanted/index.html