> From: John_Wang AT enron DOT net [mailto:John_Wang AT enron DOT net]
> Sent: Wednesday 08 November, 2000 22:00

> OTM cache size problems would result in code 11's not code 1's.

 depends. most often, yes, but sometimes when OTM cache is
full, code 1 could arise.
 a good practise for OTM is to change the default config from
"fail backup if OTM fails" to "continue without OTM", so if the
OTM cache is full, the backup will still continue instead of
failing.

> I believe that code 1's indicate problems with individual files
> within the filesystem being backed up such as long filenames or
> read errors.

 yes, sometimes. but most often, it's because of exclusive lock
on file by an application: Microsoft Office is for me the most
common cause.

> I suppose you could turn on debugging to see which file is causing
> the problems (anybody with any hints on an easy way to turn all
> those pesky newlines to carriage return line feeds in the bpkar
> logfile?)

 hum, before activating logs, just check the "problem with backup"
for this client with the report application. it will tell you
what file(s) couldn't backuped with the Win32 API error code.

> or you could just scan your disks (would an NT person care
> to comment on how, can't seem to find scandisk on my laptop).

 there is no scandisk on WinNT nor Win2k. use chkdsk or right
click on the disk in the explorer, choose tools then check
disk.

> I believe that code 1's don't invalidate the backup run like
> code 11's do, so you've still got backups minus the problem files.

 exactly. code 1 means "a backup has been made but some file couldn't
be backuped". all codes higher than 1 means "there has been a critical
problem and no backup was made".

 fx

#include <std_disclaimer.h>

--
     fx AT veritas DOT com       | Three things are certain:
François-Xavier Peretmere | Death, taxes and lost data
 http://www.veritas.com/  | Guess which has occurred.