[SR-Users] Too many packets in rtpengine UDP receive queue

[Richard Fuchs]

On 12/22/2015 10:21 AM, Zodiac wrote:
> Hi friends:
>
> I am running rtpengine daemon on a CentOS machine functionally. SIP
> server is Kamailio. Vedio and audio calls are both OK with rtpengine.
>
> But there is always prompts like following in rtpengine’s log:
>
> Dec 22 19:57:52 localhost rtpengine[12679]:
> [oaNEqGokmqgaXzj0xWyPeJDGPX7ln0gG port 30136] Too many packets in UDP
> receive queue (more than 50), aborting loop. Dropped packets possible
> Dec 22 19:57:52 localhost rtpengine[12679]:
> [oaNEqGokmqgaXzj0xWyPeJDGPX7ln0gG port 30156] Too many packets in UDP
> receive queue (more than 50), aborting loop. Dropped packets possible
> Dec 22 19:57:52 localhost rtpengine[12679]:
> [oaNEqGokmqgaXzj0xWyPeJDGPX7ln0gG port 30156] Too many packets in UDP
> receive queue (more than 50), aborting loop. Dropped packets possible
>
>
> So I am wondering whether this prompt is normal. Otherwise, what should
> I do to prevent this to happen?

The usual cause for this is when the CPU is maxed out or when the 
machine can't keep up with the traffic in some other way (frequently 
happens in VMs). Another possibility is when signalling creates a 
forwarding loop within rtpengine, but that's usually detected and logged 
with an excplicit "loop detected" message.

Cheers