CME on 2951 and Avaya IP Office. PRI stream. Problems with RTP.

[Oleksii Martyshchenko]

Schama is:
[ГТС] --PRI-- [Cisco2951,CME] --GRE Tunnel-- [Cisco 891] --LAN-- [Avaya IPO]
There is a SIP Trunk Between CME and Avaya

dial-peer voice 1000 voip
  description -= PSTN to Call Center | HQ CUCM1 =-
  translation-profile incoming CCToPSTN_tp
  translation-profile outgoing PSTNtoCC_tp
  destination-pattern 0*******80
  session protocol sipv2
  session target ipv4:192.168.100.54
  voice-class codec 2
  voice-class sip bind control source-interface Loopback0
  voice-class sip bind media source-interface Loopback0

Beyond Avaya are Twinkle softphones.

Call to the PSTN number received from PRI goes to Avaya IP office through the SIP Trunnk and then to the huntgroup of softphones.
Problem is that aperodically voice is missed sometimes to one side sometimes to another one.

I haven't access to Avaya, but we can ask admin for some information if needed.

Debug on Cisco permanently floods following:

deb voice rtp error
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, no gccb for callID:11828
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, no gccb for callID:11828
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, no gccb for callID:11828
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, no gccb for callID:11828
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, no gccb for callID:11828
Dec 5 14:29:33.461: voip_rtp_get_gccb:Error, invalid callID:-1

Dec 5 14:29:35.453: voip_rtp_set_non_rtp_call: ERROR - already in required mode; returning

And sometimes this (probably these are those calls without voice)

deb voice call trap
Dec 5 14:38:03.005: Potential Mute Call:
Dec 5 14:38:03.005: CallID1=12006; CallID2=12007; ConfID=5278
Dec 5 14:38:03.005: Leg 1: CallID=12006; TX packets: 97; RX packets: 0
Dec 5 14:38:03.005: Leg 2: CallID=12007; TX packets: 0; RX packets: 98

During packet analyse on a softphone I disovered, that ICMP-message from Avaya comes about udp-port unreachable (and then no voice).

Then I looked to CME and saw that such messages CME sends to Avaya

Dec 5 14:41:16.778: ICMP: dst (10.38.255.255) port unreachable sent to 192.168.100.54

I set codec 711alaw hardly to avoid transcoding

Where to dig? What to debug? And what in fact it could be?

[Oleksii Martyshchenko]

Wireshark capturing proved that unreachable ports are RTP ones. There are no data streams to that ports after icmp-notification.

246891 2013-12-16 10:41:09.842152 172.31.0.1 192.168.100.54 ICMP 98 Destination unreachable (Port unreachable)

      Internet Protocol Version 4, Src: 192.168.100.54 (192.168.100.54), Dst: 10.38.255.255 (10.38.255.255)

      User Datagram Protocol, Src Port: 16425 (16425), Dst Port: 17937 (17937)

Also remarkable is that some voice stream are RTP and some are not RTP ones. They are just udp segments without RTP payload.

RTP stream example:

1456 2013-12-16 10:35:06.655968 10.38.255.255 192.168.100.54 RTP 242 PT=ITU-T G.711 PCMA, SSRC=0x1567FFFF, Seq=1350, Time=2157427089

Generic Routing Encapsulation (IP)

Internet Protocol Version 4, Src: 10.38.255.255 (10.38.255.255), Dst: 192.168.100.54 (192.168.100.54)

User Datagram Protocol, Src Port: 17872 (17872), Dst Port: 16420 (16420)

Real-Time Transport Protocol

Stream setup by SDP (frame 306)

UDP  segment example:

1463 2013-12-16 10:35:06.666044 192.168.100.54 10.38.255.255 UDP 242 Source port: 16464 Destination port: 17868

Generic Routing Encapsulation (IP)

Internet Protocol Version 4, Src: 192.168.100.54 (192.168.100.54), Dst: 10.38.255.255 (10.38.255.255)

User Datagram Protocol, Src Port: 16464 (16464), Dst Port: 17868 (17868)

Data (172 bytes)

[mneira099]

Did you solve this?

I have a similar one-way audio issue... and debugging is showing: "voip_rtp_get_gccb:Error, no gccb for callID:97"

[Oleksii Martyshchenko]

Today I've updated IOS to c2951-universalk9-mz.SPA.154-1.T.bin

I still see

Dec 17 09:15:47.794: Potential Mute Call:

Dec 17 09:15:47.794: CallID1=2227; CallID2=2228; ConfID=957

Dec 17 09:15:47.794: Leg 1: CallID=2227; TX packets: 469; RX packets: 0

Dec 17 09:15:47.794: Leg 2: CallID=2228; TX packets: 0; RX packets: 469

in Voice call trap  debugging every minute or few.

Please, any advices.

[Oleksii Martyshchenko]

More debug information:

dpg_c2951_corert#sh voice dsp group al

DSP groups on slot 0:

dsp 1:

State: UP, firmware: 36.1.0

Max signal/voice channel: 32/32

Max credits: 480, Voice credits: 480, Video credits: 0

num_of_sig_chnls_allocated: 30

Transcoding channels allocated: 0

Group: FLEX_GROUP_VOICE, complexity: FLEX

Shared credits: 420, reserved credits: 0

Signaling channels allocated: 30

Voice channels allocated: 4

Credits used (rounded-up): 60

Voice channels:

Ch01: voice port: 0/0/0:15.28, codec: g711alaw, credits allocated: 15

Ch02: voice port: 0/0/0:15.30, codec: g711alaw, credits allocated: 15

Ch03: voice port: 0/0/0:15.31, codec: g711alaw, credits allocated: 15

Ch10: voice port: 0/0/0:15.25, codec: None, credits allocated: 15

Slot: 0

Device idx: 0

PVDM Slot: 0

Dsp Type: SP2600

DSP groups on slot 1:

This command is not applicable to slot 1

DSP groups on slot 2:

This command is not applicable to slot 2

DSP groups on slot 3:

This command is not applicable to slot 3

DSP groups on slot 4:

This command is not applicable to slot 4

0 DSP resource allocation failure

Calls wihout the codec. Maybe those are not RTP streams?

[Alen Liu]

I have this problem too.

Anyone can help?

[Amafsha1]

I'm running into the same issue.  Any fix?