Developers of Sympa

[Olivier Salaun - CRU <address@concealed>]

Sorry for responding late....

Your sympa.pl probably spends lots of time coping with the big msg spool.
In its main loop sympa.pl first checks all messages in spool to process 
the ones with higher priority. You could probably short-circuit this 
part of the code and process messages according to their arrival order 
instead. It starts with this :
   ## Scans files in queue
   ## Search file with highest priority
   foreach my $t_filename (sort @qfile) {
   [...]

Sergiy Zhuk wrote:

> [...]
>
> Sympa seems to be rather slow, it can only process about 2-3 msgs/sec tops
> on my dual-CPU machine.
> Same lists expanded thru aliases work much faster.
> I understand sympa is doing various checks, but probably something can be
> done to speed-up the processing.
> Lots of sites can't afford several hours delay because of one list receiving
> a large number of messages.
> Any idea what part of code needs to be optimized ?
> I'm still on 3.x though, so it may have changed in 4.x...
>  
--
Olivier Salaun
Comite Reseau des Universites