Just so I have something to relay to others in my company... Is it fair to assume that files were deleted from James' filesystem as a result of some exception handling?
Yes Ted, long story to follows... (at least this is how it worked once, and I'm guessing is what happened)
James stores messages in the file system uses two files (one is the mime message, the other is delivery information). There were situations where James would tell the computer to delete both files, but because of some file stream not closing or something wrong, the file system would only delete one of them. This left James with 1 out of a pair of files, something usable, and something we wanted to delete in the first place.
As a result, code was added so that if there is an exception trying to load a message, have James delete the message. This seemed ok since other conceived scenarios (corrupt file, 0-byte file, something unusable) were assumed to be permanent situations.
Obviously OutOfMemory does not mean there is anything wrong with the file, so we definitely want to avoid this behavior. Also as Noel said, losing data is really pretty bad, so we should be more specific on what exceptions trigger this delete behavior.
-- Serge Knystautas Lokitech >> software . strategy . design >> http://www.lokitech.com p. 301.656.5501 e. [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
