You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 2 Current »

 

In one downstream forking scenario where RTP streams are generated by multiple early dialogs and the

Unable to show "metadata-from": No such page "_space_variables"
 needs to cut through only one stream, the Downstream Forking employs “Early Dialog Media Selection Profile” to control
Unable to show "metadata-from": No such page "_space_variables"
 behavior. Support is based on the PEM value of the 18x or UPDATE or the first/last provisional response.

The following options are available:

  • Receipt of RTP (Deferred)—The first/last RTP stream is cut-thru. A flag controls whether Local Ringback Tone generated by the
    Unable to show "metadata-from": No such page "_space_variables"
     counts as RTP stream for an early dialog. RTP Server Profile is also distinguishes between allowed/disallowed streams. Disallowed streams will not qualify for the decision process.
  • Receipt of Provisional Response—RTP stream for the first/last sender of a provisional response is cut-thru. RTP Server Profile distinguishes between allowed/disallowed streams. Disallowed streams will not qualify for the decision process.
  • Use P-Early-Media Header—P-Early-Media header value is used for selection the RTP streams according to the following rules:
  • The last sent P-Early-Media value for each early dialog is used for comparison purposes.
  • The precedence order of the values will, from highest to lowest, “sendrecv”, “sendonly”, “recvonly” and “inactive”.
  • If two early dialogs have the same P-Early-Media value, media for the early dialog which last sent a reply with P-Early-Media header will cut through.
  • A P-Early-Media value update on the early dialog with the currently active RTP stream will not trigger a new comparison but will impact the media cut-through behavior.

 

When using P-Early-Media for cut-through, do not enable P-Early-Media header transparency.

Frequent changes to RTP cut-through is not desirable, thus multiple early-dialogs capability is best suited in environments where it is known that all early dialogs use the same codec or the number of early dialogs is limited.

 


 

  • No labels