[asterisk-bugs] [JIRA] (ASTERISK-29953) Asterisk sends 488 Unacceptable when the Bind port of RTCP fails.

[Piyush Badkul (JIRA)]

     [ https://issues.asterisk.org/jira/browse/ASTERISK-29953?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Piyush Badkul updated ASTERISK-29953:
-------------------------------------

    Issue Consultant:   (was: Unassigned)

> Asterisk sends 488 Unacceptable when the Bind port of RTCP fails.
> -----------------------------------------------------------------
>
>                 Key: ASTERISK-29953
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-29953
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Channels/chan_sip/CodecHandling, Channels/chan_sip/SRTP
>    Affects Versions: 19.0.0, 19.1.0, 19.2.1
>         Environment: This problem will come for every environment and for every release number.
>            Reporter: Piyush Badkul
>            Assignee: Piyush Badkul
>
> Whenever a call establishes through asterisk Softswitch, asterisk goes out to find an available RTP port from a predefined port range given in rtp.conf. Once it finds the available port, then it attempts to bind it and if the binding of that RTP port fails, then it jumps to the next available port in the list. After binding to the RTP port correctly, just sets the RTCP port to next higher odd port number. It only attempts to bind the RTCP port once, and if it fails the asterisk doesn't care and move on to handle other things required for the setting up of a call. 
> This problem only arises when the asterisk behaves like a media server for the encrypted call and several other applications run on the blade using the same port range. This problem is random as specified from the above description when the binding of RTCP is failed. Since RTCP is mandatory for the encrypted call.   
> However, it should not matter that multiple applications are running on the same blade, customer should not face failure when initiating a call. I believe that algorithm for the same can be modified and this problem can be resolved.
> More details at- https://app.hackernoon.com/mobile/VltlIsRerlWvDLXnr6ND
> If leads at asterisk believe that this is really a problem and should be addressed, I will be happy to explain the problem in more detail and I will also like to submit the patch for the following. If any better way than what I am proposing exists, I will be happy to understand it and implement it for asterisk.
> In case it should be fixed, please allow me to provide a patch and assign this issue to me. 



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