[rdiff-backup-users] Missing snapshot.gz (Poss...path+file length >256 chars)

[listserv . traffic]

I posted this last week, but got no feedback - so I tried poking
around a bit, and trying some stuff from the FAQ and the mailing list.

However...
*** I THINK I NOW KNOW WHAT IS WRONG ***
The whole path of the file and it's name exceeds 256 characters.
(I figured this out by trying to create the file and wasn't able to
get the whole file name in...)

So is there a fix where the path+file name length is >256 chars.
Remember the source and dest are NTFS system drives, both local.

I assume I can try to shorten the destination path...but any other
thoughts?

***

The error I'm having follows.
(This is a repost of the orig. post for history sake...and clarity.)
---
Exception '[Errno 2] No such file or directory: 
'E:/rdiff-backup-rep2/data/rdiff-backup-data/increments/Engineering/illustrations/pdf/manuals/New/229_1021
 (Utility Center Delivery assembly)/il1588 General Dimensions - Universal 
Utility Center Delivery 
system.pdf.2009-06-11T20;05800;05801-07;05800.snapshot.gz'' raised of class 
'<type 'exceptions.IOError'>':
  File "rdiff_backup\robust.pyc", line 32, in check_common_error
  File "rdiff_backup\increment.pyc", line 44, in Increment
  File "rdiff_backup\increment.pyc", line 73, in makesnapshot
  File "rdiff_backup\rpath.pyc", line 243, in copy_with_attribs
  File "rdiff_backup\rpath.pyc", line 105, in copy
  File "rdiff_backup\rpath.pyc", line 133, in copy_reg_file
  File "rdiff_backup\rpath.pyc", line 1194, in write_from_fileobj
  File "rdiff_backup\rpath.pyc", line 1173, in open
  File "gzip.pyc", line 79, in __init__

  The error is above.

  If it matters, here's my env.
  Windows 2000 Server
  Windows binary
  Source and dest are independent local HDD's

It appears on closer inspection that the dest above doesn't contain a
increment that rdiff expects to find - and that missing increment
causes the backup to crash.

Upon looking at the E: destination above, I do see that no increment
exists.

So, how would one go about fixing this? I'd like to keep my
repository and simply acknowledge that we have a missing diff - but
to simply ignore it's loss and continue as best possible.

Any thoughts on how to accomplish this, since I'm at a loss to know
what to do?

-Greg