Link to home
Start Free TrialLog in
Avatar of Greg Mason
Greg Mason

asked on

VoIp/VPN...

I have a bit of a challenge with a site-to-site VOIP situation.
We are using a Avaya system - PBX hosted on-site - in a 5 location business. They use a SIP trunk provider outbound and that's not a problem. But they also use the system as a sort of "intercom" to communicate between the sites. To make it work, we have setup VPN over the public internet in a "star" pattern, with one of the sites acting as the "hub" - the others as the "spokes". Traffic flows between the sites through the hub, or from the sites TO the hub, depending on who is being called.

Call quality is the problem. Choppy, dropouts or bad voice quality happen but NOT consistently. Just occasionally enough to be a pain. The business uses the "intercom" feature quite frequently, it's becoming a problem.

We use SonicWALL TZ300's at the spokes and a TZ400 at the "hub". We have QoS and Bandwidth management configured and that has helped. We have spoken to Sonic About it and they have put their 2 cents in.

Any suggestions would be appreciated.
Avatar of arnold
arnold
Flag of United States of America image

Everything flows through the hub, does their wan upload capable of supporting.

Pick a pair pf spokes and setup a VPN between them with qos and see whether a direct path between spokes improves the intercom functionality.


Qos has two one is to prioritize voice over data, the other additional qos deals with prioritizing voice traffic within the VPN.

To that, the hub wan bandwidth has to reserve let's say 30% to handle the sip trunk from the provider. 30% for VPN traffic
Then see if that improves

Similar qos policies reserving bandwidth for VPN .

What is the queuing setting, FIFO?
Avatar of Greg Mason
Greg Mason

ASKER

Thanks. We originally had the sites set to route direct point-to-point. In an effort to trouble shoot we decided to simplify things and build the star topo. We are using 802.1p QOS and have it set to "phone". Bandwidth management is enabled on the VPN. It has improved things a bit.

I am unaware of any "FIFO" settings in the sonic O/S having to do with queuing.

Hope this helps.
interface queuing on the sonic, deals with behavior if saturated, drop packetsor queue?

the reserved bandwidth might not be enough.

does sip, h.232, voip fall into the "phone" within vpn.  deals with whether there is competing data flowing through the vpn ....
I'll check with Sonic support but the traffic requirements for the "intercom" system between sites are not that dramatic - maybe 4 calls simultaneous, max - typically 1 or 2 at a time. The VPN circuits also handle Outlook traffic back/forth to the business exchange service and VP update traffic.
Depending which your intercom uses, it can be between 170kb to 199kb per call bandwidth, when it has to go through the hub, it consumes double the requirement on the hub per call, in a hub/mesh if path through any, but with multiple VPNs you have that overhead consuming the bandwidth at .....

Identifying issues by looking/monitoring bandwidth utilization at each spoke/hub..
This question needs an answer!
Become an EE member today
7 DAY FREE TRIAL
Members can start a 7-Day Free trial then enjoy unlimited access to the platform.
View membership options
or
Learn why we charge membership fees
We get it - no one likes a content blocker. Take one extra minute and find out why we block content.