Hello everyone,
I am a voice recording engineer and I have a few question for the community regarding RSPAN configurations and QOS requirements for a VoIP environment running on Cisco core switches.
Current Environment
My customer has two PABXs running in their organisation. One Cisco CUCM v8.6 and one Avaya/Nortel CS1000. All phones (Cisco and Avaya\Nortel) are VoIP. Currently the customer is spanning all the phones (Cisco and Avaya\Nortel) to one Voice VLAN. This VLAN is then spanned to 6 destination points, being the recording capture NICs. The customer is not running QOS on their core switches. The customer has both active and passive recorders for each type of phone in their environment.
Problems found include:
1. When running a Wireshark trace on the passive recorders, decoding all streams as RTP and analyzing the output, the processed audio quality is very poor.
2. Wireshark reports that the voice packets are not arriving in the correct order at the passive recorder NICs.
3. Some voice packets are being lost.
4. On the passive recorders, the resultant recording only has the customer side of the conversation and not the agent side.
5. In some cases there are no recorded audio for some extensions on the active recorders.
6. Duplicate packets are received on the Cisco passive recorders' NICs.
Please advise what would be the best practises from a Cisco point of view would be in order to resolve the problems described above.
Kind regards,
Craig.