Subj : [SemPoint] Breakthrough
To   : Martin Foster
From : August Abolins
Date : Fri Jun 07 2019 20:50:00

Hello Martin!

** 05.06.19 - 12:57, Martin Foster wrote to August Abolins:

 AA>> But remember, I was having THIS problem:

 MF>Yes.

 AA>>> 05/25/2019  07:19 PM           162,203 FIDOTEST.JDT
 AA>>> 05/26/2019  05:07 PM             1,520 FIDOTEST.JDX
 AA>>> 05/26/2019  05:13 PM     4,294,967,371 FIDOTEST.JHR
 AA>>> 05/26/2019  05:13 PM                32 FIDOTEST.JLR

 AA>> Reading existing messages and writing a new message went well.  But as
 AA>> soon as I went back the C)hange an message that I had previously posted
 AA>> (but not yet tossed or sent), the above happend.

 MF>Earlier this morning, I did a very simple test which was to delete one
 MF>single message. Doesn't get much simpler, does it?

Very simple.  :)


 MF>SemPoint immediately locked up and after watching the twirly-whirly
 MF>wheel spinning round and round for something like 20 seconds, I got
 MF>control back. The message list was still there but when I hit <Enter>
 MF>on one of the messages, the message was blank, as were all the others.
 MF>I then looked at my messagebase directory and there it was in all its
 MF>glory, the infamous 4Gb file - Pah!!

The same problem should occur if you tried to C)hange a message.


 MF>From my tests so far, I conclude that the only viable message base
 MF>format is *.MSG - Hmph!

Yes.. so we can eliminate the use of JAM if used in Win7/32, or WinXP.  :(

But then the jury is still out on Squish bases (via Fastecho).

Modifying, C)hange-ing a message in a Squish message base did not trigger  
the 4GB bloat problem.  But I still don't have Fastecho setup to fully  
test incoming and outgoing prosessing.



  ../|ug

--- OpenXP 5.0.39
 * Origin: o,,,,o§ø`ø§o,,,,o§ø`ø§o,,,,o (2:221/1.58)

.