[asterisk-users] Sonicwall potentially causing long ping times to SIP phones

[Bill Michaelson]

Kristian Kielhofner wrote:

> On 10/23/08, Bruce Komito <brucek at bagel.com> wrote:
>   
>> > We've had LOTS of problems with Sonicwalls doing bad things to SIP and RTP
>> >  connections.  I've seen the delay thing, as well as the Sonicwall throwing
>> >  away entries from the ARP table because of inactivity.  I've also seen
>> >  sporadic, intermittent problems with transfer from one phone to another.
>> >  I have no doubt that a new, properly configured Sonicwall can be made to
>> >  function properly in a VoIP environment, but we are not Sonicwall experts,
>> >  nor are many of the purported experts.  In every case where we've had
>> >  problems with VoIP behind a Sonicwall, the problems ALL disappear when we
>> >  put the phones on a LAN segment that does not pass through the Sonicwall.
>> >  So, now that's our going in position.  If it works, great, but if it
>> >  doesn't, our solution is to take the Sonicwall out of the picture.
>> >
>> >  My $.02 .
>> >
>> >  Bruce Komito
>> >  WPTI Telecom
>> >  (775) 236-5815
>> >
>>     
> I wouldn't single out SonicWalls when it comes to breaking SIP 
> traffic. Most of the "anything but simple PAT" devices I've seen that 
> implement any SIP specific fixups usually end up breaking something 
> along the line. Unless the product is from a company where SIP is 
> their core competency (like Ingate, or /maybe/ Cisco) it's best to 
> stay away and/or disable the SIP specific fixups wherever possible. 
> I'm looking forward to the day when SIP-TLS is the norm and these 
> devices have no idea what kind of traffic is flowing through them!
> -
I sympathize, especially since a client of mine is facing the same 
situation. A potential update to their configuration involves exactly 
what you (Kristian) suggest: layering TLS in-between. I've run SIP/RTP 
and IAX over openVPN without issue routinely. What worries me is that 
the problem is not related to SIP awareness, and that some erratic 
performance by the Sonicwall that is benign in most circumstances 
manifests as a quality issue when carrying media streams. Seems 
unlikely, but does anybody have any clarity on this?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-users/attachments/20081024/7aba25a4/attachment.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3234 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.digium.com/pipermail/asterisk-users/attachments/20081024/7aba25a4/attachment.bin