alex@aipna.ed.ac.uk (Alex Zbyslaw) (11/22/88)
Machine: Vax 11/750
OS: Wisconsin 4.3BSD +NFS
We developed a hard error on an eagle disk (hp0) and tried `getting rid of
it' with bad144. We did
bad144 -f -a eagle hp0 269600
followed by
bad144 eagle hp0
and lo and behold sector 269600 had been added to the forwarding table.
Running fsck on the file system produced a hard error sn269600.
So then we tried
bad144 -a eagle hp0 269600
and
bad144 eagle hp0
and there was 269600 in the forwarding tables twice. Fsck bombed again.
So then we resorted to badsect, creating the BAD directory, and saying
badsect BAD 269600
unmounted and ran fsck just like the manual said to. Only fsck didn't
report the error that badsect(8) led us to believe it would (ie no "Unlink
file BAD/nnnn" or whatever).
Meanwhile fsck still barfed. So in the end we `mv'ed the culprit directory
to BAD and chmod 0ed both. Now no fsck problems, but undoubtedly a bad
block still on the disk.
What are we missing. I think we followed the manual pages to the letter.
(Aside: someone else in the building had exactly the same trouble under
4.2BSD on a VAX 750, so we're not the only ones missing something).
Since our disks are a bit old, and producing more ecc and retries by the
month, I expect more bad blocks to develop, so would really like to know the
best cure.
Please respond by email and I'll summarise.
Thanks in advance,
--Alex
JANET: alex@uk.ac.ed.eusip ARPA: alex%ed.eusip@nss.cs.ucl
UUCP: ...{uunet, decvax, ihnp4}!mcvax!ukc!eusip!alex
[CSNET BITNET]: alex%ed.eusip%nss.cs.ucl@[csnet-relay cunyvm]