todd@ivucsb.UUCP (Todd Day <todd>) (08/25/88)
I recently enquired about removing a bad block that sometimes shows up
in my unix.log. Well, I suspect that my bad block has something to do
with the uucp stuff, as the error times always seem to coincide with
some item in /usr/spool/uucp/LOGFILE. If I get rid of this bad block
it could be catastrophic to my mailing system.
I was wondering if there is some way to go out and actually read the block
before I delete it, so I can see what it is and prepare for its removal.
Actually, what I was really thinking of doing was renaming whatever file
it happened to be in to file.old and copy it to itself.
Another thing I was wondering is what causes soft errors. I mean, how
does the computer recover from something like a bad hard disk read?
Thanks in advance.
--
/| Todd Day (805)968-9352 |\ "I asked for a car, I got a computer. How's
+-+ | The Audio Club at UCSB | +-+ that for being born under a bad sign?"
+-+ | 926 B Camino Del Sur | +-+ comdesign!ivucsb!todd@pyramid.com
\| Isla Vista, CA 93117 |/ {sun}!pyramid!comdesign!ivucsb!toddrichard@uhccux.uhcc.hawaii.edu (Richard Foulk) (08/26/88)
} I was wondering if there is some way to go out and actually read the block } before I delete it, so I can see what it is and prepare for its removal. } Actually, what I was really thinking of doing was renaming whatever file } it happened to be in to file.old and copy it to itself. If you do find the file, perhaps by noticing that when you cat it to /dev/null that you get another error reported in your log file, you should be able to mv it to another name (same filesystem). Then it would be best restore the file from your backups. } Another thing I was wondering is what causes soft errors. I mean, how } does the computer recover from something like a bad hard disk read? Either the disk controller catches the error and corrects it through various means, or the driver does some retries until it succeeds without an error report from the controller. Richard