home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!ucbvax!lrw.com!leichter
- From: leichter@lrw.com (Jerry Leichter)
- Newsgroups: comp.os.vms
- Subject: Re: Question about RMS and MSCP-pair
- Message-ID: <9301051405.AA16112@uu3.psi.com>
- Date: 5 Jan 93 12:47:58 GMT
- Sender: daemon@ucbvax.BERKELEY.EDU
- Distribution: world
- Organization: The Internet
- Lines: 47
-
-
- [Referring to dealing with files with forced errors:]
- Use BACKUP or any other tool you have written to accept and ignore the
- forced error, copying the file. Delete the original. The block is
- happily revectored and can be reused by a WRITE operations, its just
- that the system has reason to suspect that the data in the block MIGHT
- be corrupted
-
- By the time you get a "forced error", the data is gone. The forced error
- indicator is set on the REPLACEMENT block, not the original one, and I don't
- believe the corrupted data, such as it was, is even copied into the block.
-
- and VMS taks data integrity very seriously. So seriously
- in this case that you have to be prepared to dismiss an ERROR return
- to get to it.
-
- Tool suggestions:
-
- A useful tool for someone to write (who is so inclined) would be a
- utility program to read and rewrite the blocks of a file handling and
- checking about forced errors as they happen.
-
- Just what would it do with a block that had a forced error? It has no way to
- determine what was there.
-
- A common suggestion for dealing with a file with a forced error is to back it
- up, then restore it. BACKUP will read past the block with the forced error,
- and when it restores the file, will simply substitute a block of zeroes for
- the data it couldn't read when it created the saveset. So the result of this
- suggestion is to produce a file which doesn't have a forced error - but which
- doesn't contain the right data either. This is an improvement?
-
- The tool you are suggesting would simply provide people with a simpler way to
- screw themselves. This is sweeping under the rug not dust but burning embers!
-
- Another interesting varation on that tool would be something to run on
- a privately mounted disk to read or double read and compare, overwrite
- with test patterns, and then rewrite each block on a disk.
-
- Some of this is already in ANALYZE/DISK/READ_CHECK. The rest is in the old
- BAD utility (ANALYZE/MEDIA), though that will refuse to run on DSA disks.
- However, that utility won't run on DSA disks, reporting "Unable to analyze
- <disk>, due to revector caching". You really need diagnosic-level code
- (which has its own specialized disk drivers) to do this properly on a DSA
- disk.
- -- Jerry
-
-