There is a SIPsak probe that can be used, but honestly the basic ICMP probe
tends to show what's going on quite effectively. Maybe increase the size of
the probe packet to more accurately simulate RTP traffic. You can run
smokeping from a single machine somewhere within your network environment,
and just set up endpoints as needed for however long it takes to get some
useful data.

The output of smokeping is totally visual, and that's what makes it so
useful: you can send the graphic images to the customer and it becomes
immediately obvious that there are problems, regardless of how much
theoretical bandwidth they have. RTP is affected by packet loss and jitter,
and that can be a problem regardless of bandwidth.

It's also worth determining if they have any sort of QoS set up on their
router. If not, any sort of traffic can potentially cause RTP packets to be
discarded.

On Tue, Feb 10, 2015 at 2:10 PM, Chuck Mariotti <[email protected]>
wrote:

>  Is there a specific probe or setup you suggest? Do you have a central
> server gathering or just a simple install gathering the data at the
> endpoint?
>
>
>
> I guess I could see if there are any Virtual Machines that have this ready
> to go and install it on a laptop.
>
>
>
> Thanks Jim.
>
>
>
> Chuck
>
>
>
> *From:* Jim Van Meggelen [mailto:[email protected]]
> *Sent:* February-10-15 1:32 PM
> *To:* Chuck Mariotti
> *Cc:* TAUG Technical
> *Subject:* Re: [on-asterisk] Connection Performance Issues...
>
>
>
> Install smokeping and have it run for a few days. That should tell the
> tale and give you some nice graphs so you can show him the packet
> loss/jitter.
>
>
>
>
>
> On Tue, Feb 10, 2015 at 1:18 PM, Chuck Mariotti <[email protected]>
> wrote:
>
> I have a client that seems to be having internet issues. Unfortunately,
> he's paying a lot of money for a Rogers connection with a lot of speed
> (256Mbps) and when there are issues, he sends me a speedtest.net report
> indicating that speed is fine... I suspect that not all is well but not
> possible to prove since Rogers says, all is fine.
>
> Is there any other suggestions on testing the quality of the connection at
> that time? Basically they have Yealinks OpenVPNed into our datacenter.
>
> My other suspicion is maybe the OpenVPN tunnels are clashing somehow but I
> have no idea on how to track this down.
>
> Regards,
>
> Chuck
>
>
>

Reply via email to