On (11 Nov 02) Michael Dillon wrote to Frank Vest...

Hello Michael,

FV> Welcome aboard. Congratulations on being the one to work on the Amail
FV> code (I assume this is correct). Thanks for working on the code for
FV> the Spitfire Sysops and others (Synchronet would be nice, hint :)

 MD> You are correct in that assumption.  I've been working on the hang
 MD> problem and it appears to directly (as previously thought) refer to
 MD> long messages. I am uncertain at this time why exactly it's having the
 MD> hanging issue but I do know it's locking itself into an infinite loop
 MD> reading the same message multiple times.  I will fix this.  Ignoring
 MD> long messages will be the first fix until I can get it to read long
 MD> messages and toss them safely.  This is the current goal.

Larry should be sending you a "lock-up.zip" file that contains a
packet that locks Amail. 

I'm not a programmer of any sort but I will offer this.... I used to
run a program to split long messages in packets and fix some grunged
messages before running Amail. This helped a lot with the tossing.
Maybe some sort of routine in Amail could be made to do the same
thing.

The "packet sort" program that I used also reorganized the packets
in a bundle into one big packet that had the echos in order by
subject. This increased the speed of tossing and took little time to
do compared to the speed increase. Of course, that was in the days of
386 and lower CPUs with large Fidonet packets where we needed all the
speed help we could get. :-))


Regards,

Frank

http://pages.sbcglobal.net/flv
http://biseonline.com/r19

--- PPoint 3.01