On Sep 27, 2006, at 3:17 AM, Csv4Me2 wrote:
Yeah, I second that. Handling extremely groups should be possible
on machines
with a small memory footprint, say 512 Mb. That will probably mean
a major
overhaul in how pan manages loading, sorting and aggregating all those
multiparts. And perhaps Charles will make that choice, I hope he does.
How many years ago was the discussion centering on using a SQL
database to handle the metadata? :-)
On-disk storage would solve the memory footprint problem but come at
the price of much slower response time.
I think you should try 0.14.2.91 and find 2e6 to 2.5e6 messages was
about the limit with a 2GB process size limit. 0.114 is two to three
times more efficient. So at least one "major overhaul" has recently
happened. Thanks Charles!
--
David Kelly N4HHE, [EMAIL PROTECTED]
========================================================================
Whom computers would destroy, they must first drive mad.
_______________________________________________
Pan-users mailing list
Pan-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/pan-users
