Crash report: exceptions.OSError

[Ben Escoto]

--==_Exmh_-2130006796P
Content-Type: text/plain; charset=us-ascii

>>>>> "LB" == Leo Breebaart <leo@lspace.org>
>>>>> wrote the following on Fri, 15 Mar 2002 16:56:30 +0100

  >> Are you sure it really crashed? On my system I was able to
  >> recreate part of the error: [...]

  LB> You are right. I've consulted the log file, and there was indeed
  LB> a 'Skipping file' there that I had missed. The presence of the
  LB> exception stack trace fooled me into thinking the program had
  LB> actually crashed when it hadn't -- my apologies.

  LB> Does 0.6.0 now handle this case differently, or is a trace still
  LB> generated?

Yeah, if the error message happens at the end it is misleading.  Maybe
later some of this will be improved.  But the idea was that if
something goes wrong, a lot of information should be printed, so you
can post, say, to this list and get a reasonable answer.  Before I was
getting complaints where I didn't have a clue what the problem was.

    Also, if you saw a stack trace in the log file, it means
rdiff-backup hadn't crashed when that trace was written.  Otherwise,
who would be writing it to the log?  Of course, it could crash
directly afterwards...


--
Ben Escoto

--==_Exmh_-2130006796P
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Exmh version 2.5 01/15/2001

iD8DBQE8k5Y++owuOvknOnURAuNdAJ4uRAR6FCGv4kKVCkuEbetxz/C4AgCdHwFZ
jmOJ5yTNb1T2WqMf7O8QTFk=
=5DGQ
-----END PGP SIGNATURE-----

--==_Exmh_-2130006796P--