NBAR2 Protocol Pack and QoS

[Kevin-John Beasley]

Hi All,

Was just reading a old NBAR2 document relating to Protocol Pack 7 when I saw this note:

Note

The RTP audio/video sub-classification parameters are generic in nature and will match only on generic RTP traffic. More specific classification such as ms-lync-audio, cisco-jabber-audio, facetime, and cisco-phone will not match as RTP, and therefore will not match the audio/video sub-classification.

Ref: https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/qos_nbar/configuration/xe-3s/qos-nbar-xe-3s-book/nbar-protocl-pack.html

Can anyone speak more to this claim and if it is still valid? I mean we are now way up on Protocol Pack 33 or 36 for Advanced but if match protocol rtp audio can't actually match stuff like cisco-phone audio and cisco-jabber-audio then it's a pretty damming statement for Cisco and further proof that the entire AVC/NBAR2 suite needs to be made a free feature of IPBase on all supported routers. I mean I personally have configured heaps of routers with match protocol rtp audio under a class-map for voice and I am sure thousands of others have done it as well. Hell I have had multiple Cisco Gold partners turn around and confirm this design and even specifically tell me that match protocol rtp audio will match Cisco IP phones and Cisco Jabber audio calls.

Am I getting worried over nothing or is this actually a huge gotcha that Cisco and partners alike have managed to overlook?

[Joseph W. Doherty]

I too recently read that and wondered on the same.  I was just on the verge of trying protocol rtp audio, for Cisco VoIP phones, but the "need" got eliminated, so, unfortunately, I didn't find out.  If protocol rtp audio didn't work, I was prepared to next try protocol rtp.

In your reference, right above that note, mention was made it works if audio is related to SIP (which would have also been true for me).

I too hope others post a response.