[asterisk-bugs] [JIRA] (ASTERISK-25578) [patch] SIP/SDP: No rtpmap for static RTP payload IDs

Joshua Colp (JIRA) noreply at issues.asterisk.org
Wed Jun 8 05:15:56 CDT 2016


     [ https://issues.asterisk.org/jira/browse/ASTERISK-25578?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Joshua Colp updated ASTERISK-25578:
-----------------------------------

    Assignee: Alexander Traud

> [patch] SIP/SDP: No rtpmap for static RTP payload IDs
> -----------------------------------------------------
>
>                 Key: ASTERISK-25578
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-25578
>             Project: Asterisk
>          Issue Type: Improvement
>      Security Level: None
>          Components: Channels/chan_sip/NewFeature
>            Reporter: Alexander Traud
>            Assignee: Alexander Traud
>            Severity: Minor
>         Attachments: no_rtpmap_for_static.patch
>
>
> If a lot of audio and video codecs are enabled (allowed), an outgoing call might hit the MTU bearer of 1300 bytes for SIP over UDP. In [SDP|https://tools.ietf.org/html/rfc4566#section-5.14], specifying the rtpmap for a [static RTP payload ID|http://www.iana.org/assignments/rtp-parameters] is a SHOULD but optional. This saves around 20 bytes per format. This affects G.722, G.711, G.729, and GSM for example. Consequently, more than 100 bytes can be saved even on typical deployments = around 10%.
> FreeSWITCH does this on default but reports some broken clients implementations. Therefore, FreeSWITCH included the variable [verbose_sdp|https://wiki.freeswitch.org/wiki/Variable_verbose_sdp] to disable this feature. Consequently, although in different specifications, I coupled this new feature with [SIP Compact Form|https://tools.ietf.org/html/rfc3261#section-7.3.3]. Compact Form is disabled on default. When someone enables those compact headers, he is about reducing the message size and does benefit from this change here as well.
> This reduces the complexity of the attached patch to two Boolean statements. I would be more than happy, if you accept this new feature for inclusion. If the Asterisk team is OK with this coupling, please, say so. If not, and because I am not an integrator, please, create a name for this parameter/switch.



--
This message was sent by Atlassian JIRA
(v6.2#6252)



More information about the asterisk-bugs mailing list