[Freeswitch-users] Problem with rtp streams and stair-stepped (or ramped) increased lag

[William King]

I've seen a similar issue on a small network behind a pfsense router.
Network diagram is:
Hard Phones(3+ Yealink and Polycom mix) -> switch -> PFsense -> Comcast
-> FS conferencing server.

In my case I was seeing ~10 seconds of RTP delay(tested from doing a/b
response testing), but instant sip handling. The audio was in perfect
sync coming from the conference server to the 3 endpoints.

Nathan, what version of FS are you running?

I have not yet ruled out that the issue isn't a local network problem.

William King
Senior Engineer
Quentus Technologies, INC
1037 NE 65th St Suite 273
Seattle, WA 98115
Main:   (877) 211-9337
Office: (206) 388-4772
Cell:   (253) 686-5518
william.king at quentustech.com

On 06/24/2013 04:40 PM, Nathan Neulinger wrote:
> In one particular stream, which is  call mgr <-> FS, the CCM to freeswitch side of the stream shows no issues, but 
> freeswitch to CCM direction shows a number of drifts, some just one ptime worth, others larger.
> 
> This instance is running as a VM, but allocated significantly more reserved cpu (and very high shares) on a lightly 
> loaded host. Concurrent calls on the FS instance is in the single digits at most at this point. Typical end to end 
> network induced latencies less than 3 ms at most.
> 
> Any suggestions on what to look at or steps to take here? Certainly not happening on every call, but I do believe I've 
> seen it on both cisco and polycom phone connections. Usually, there is just a slow stairstep over the period of the call.
> 
> -- Nathan
> 
> On 06/24/2013 06:24 PM, Nathan Neulinger wrote:
>> Having some occasional user reports of problem calls where they are fine, and then something happens on the call and lag
>> increases significantly, sometimes up to 2+ seconds.
>>
>> I've got severe examples on both cisco and polycom phones. What I'm seeing looks like the call is going perfectly for a
>> period of time (sometimes up to a few minutes), and then there is a period of time with higher latency packets, and the
>> overall delay in the graph increases steadily at that point, and never recovers.
>>
>> It sounded like rtp-autoflush-during-bridge might be something relevant, but that's set to the default. Also saw
>> "rtp-rewrite-timestamps", but unclear if that is necessary here.
>>
>> I've got pcap captures of the rtp streams if that'll help any.
>>
>> -- Nathan
>>
>> ------------------------------------------------------------
>> Nathan Neulinger                       nneul at mst.edu
>> Missouri S&T Information Technology    (573) 612-1412
>> System Administrator - Architect
>>
>> _________________________________________________________________________
>> Professional FreeSWITCH Consulting Services:
>> consulting at freeswitch.org
>> http://www.freeswitchsolutions.com
>>
>> 
>> 
>>
>> Official FreeSWITCH Sites
>> http://www.freeswitch.org
>> http://wiki.freeswitch.org
>> http://www.cluecon.com
>>
>> FreeSWITCH-users mailing list
>> FreeSWITCH-users at lists.freeswitch.org
>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
>> http://www.freeswitch.org
>>
>