Here is why I ask:

[EMAIL PROTECTED] openser]# netstat -su
Udp:
    159293 packets received
    2016 packets to unknown port received.
    153899 packet receive errors
    185342 packets sent

On my proxies, as traffic levels increase, I start getting udp errors, which
I think is pretty much a udp buffer overflow. Once it overflows, that proxy
is toast until traffic levels die down. I have a pretty basic config, but I
can't seem to handle more than 4 or 5 new calls per second. I would expect
to be able to do a lot more than that. So I was trying to make sure that no
unnecessary udp traffic was going in or out.

The server has two Dual Core Opteron processors, 4 Gb of RAM, running CentOS
5 x86_64.

Any suggestions are appreciated... and I'll happily pay for consulting if
someone wants to look into it.

Michael


-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Michael Young
Sent: Friday, May 16, 2008 9:20 PM
To: [email protected]
Subject: [OpenSER-Users] Openser 1.3.2 rev_dns=no

In my openser.cfg (running 1.3.2), I have:

dns=no
rev_dns=no

but a packet capture shows OpenSER doing a reverse DNS lookup on my
provider's proxy IP address with every invite. Are those lines no longer
supported? I tried starting OpenSER with -R, but same behavior.

Michael Young


No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.524 / Virus Database: 269.23.16/1446 - Release Date: 5/16/2008
7:42 AM
 


_______________________________________________
Users mailing list
[email protected]
http://lists.openser.org/cgi-bin/mailman/listinfo/users

No virus found in this incoming message.
Checked by AVG. 
Version: 7.5.524 / Virus Database: 269.23.16/1446 - Release Date: 5/16/2008
7:42 AM
 

No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.524 / Virus Database: 269.23.20/1453 - Release Date: 5/18/2008
9:31 AM
 


_______________________________________________
Users mailing list
[email protected]
http://lists.openser.org/cgi-bin/mailman/listinfo/users

Reply via email to