[isalist] Re: Question regarding video conferencing
- From: Rob Moore <RMoore@xxxxxxxx>
- To: "isalist@xxxxxxxxxxxxx" <isalist@xxxxxxxxxxxxx>
- Date: Thu, 19 Jan 2012 08:45:24 -0500
OK, I looked in the application log from our last testing period. There were,
indeed, a few "exceeded connection limit" warnings at that time. I don't know
for sure who they belonged to, though. I'll have to set up another test and
watch for that.
Meanwhile, is there a good way to reconfigure the connection limits? I suppose
I could just turn off Flood Mitigation all together, but I'm not keen on that
(except maybe for testing purposes). Do I need to add IP Exceptions for the
Vidyo portal? (Remember, it's outside the network.) For the whole internal
subnet? Both?
Thanks,
Rob
From: isalist-bounce@xxxxxxxxxxxxx [mailto:isalist-bounce@xxxxxxxxxxxxx] On
Behalf Of Rob Moore
Sent: Thursday, January 19, 2012 8:06 AM
To: isalist@xxxxxxxxxxxxx
Subject: [isalist] Re: Question regarding video conferencing
Where would I find the connection alert limits? In the event logs? Or somewhere
in the TMG console?
I got the ports and protocols from someone at Vidyo.
Rob
From: isalist-bounce@xxxxxxxxxxxxx<mailto:isalist-bounce@xxxxxxxxxxxxx>
[mailto:isalist-bounce@xxxxxxxxxxxxx]<mailto:[mailto:isalist-bounce@xxxxxxxxxxxxx]>
On Behalf Of Jim Harrison
Sent: Wednesday, January 18, 2012 8:22 PM
To: isalist@xxxxxxxxxxxxx<mailto:isalist@xxxxxxxxxxxxx>
Subject: [isalist] Re: Question regarding video conferencing
Odds are good that you're hitting TMG connection limits with these clients.
The more they interact, the more traffic and connections they create.
Do you see any connection limit alerts from that time frame?
Where did you get those ports/protocols?
I tried searching the Vidyo site and while it's very pretty and all
SilverLight-y, it's a major PITA to find anything more than marketing spew.
It's also entirely possible that you simply can't use this application across a
truly stateful firewall such as TMG.
From: isalist-bounce@xxxxxxxxxxxxx<mailto:isalist-bounce@xxxxxxxxxxxxx>
[mailto:isalist-bounce@xxxxxxxxxxxxx]<mailto:[mailto:isalist-bounce@xxxxxxxxxxxxx]>
On Behalf Of Rob Moore
Sent: Wednesday, January 18, 2012 9:21 AM
To: isalist@xxxxxxxxxxxxx<mailto:isalist@xxxxxxxxxxxxx>
Subject: [isalist] Question regarding video conferencing
Hello-
We have our own video conferencing solution, from a company called Vidyo. The
simplest solution was to put the hardware outside our TMG firewall. (We have a
group that manages the Vidyo portal for us and they say it's fine to have it
out there.) I have allowed a variety of traffic to it, including: HTTP 10000,
17990, 17992, 3306, 50000, and UDP 50000 - 65535.
We are having some rather quirky issues with it and have done a lot of testing
to try isolate exactly where the problems lie. What happens is, we get people
to connect to our portal. Everything works normally for everyone. Then, maybe
10 minutes into the conference, bad things start happening. Audio starts
breaking up. Video gets jerky. The number of people you can see in the meeting
drops to only two or three instead of staying steady at six or eight. (This is
a feature of the Vidyo software. If you have poor bandwidth, Vidyo only shows
you the person speaking and maybe the last speaker or two. But we don't have
poor bandwidth. We have a 100Gb connection to the Internet.) The weird thing is
that these bad things don't happen to everyone. They don't happen to people
connecting to the portal from outside the network. And they only happen to some
of the people connected inside the network. We've had as many as eight
simultaneous connections, and some will be working fine showing eight
participants, while others will be having trouble and only showing two or
three, with audio and/or video problems. I've looked to see if it could be a
switch problem, but some of us without problems are connected to the same
switches as those with problems.
We've also tried connecting to a remote Vidyo portal. The remote users had no
issues, but we who were connecting through our TMG server did have problems.
It seems like the most likely culprit is TMG, although the data don't fully
support this conclusion. It seems to me if TMG were at fault, all users would
experience the same problems at the same time. But that's not the case.
Since the problems don't start until something like 10 minutes into the
conference, it's as if something were throttling the traffic at least for some
of the computers. As far as I know, nothing like that is going on. In the past
I've looked to see if I can see any problems by monitoring traffic on TMG, but
never found anything. It's been a while since I've done that, though. So it
might be worth looking again.
Any ideas about what might be going on, or suggestions to pursue?
Thanks,
Rob
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Rob Moore
Network Manager
215-241-7870
Helpdesk: 800-500-AFSC
Other related posts:
