[asterisk-bugs] [JIRA] (ASTERISK-27251) chan_sip doesn't honour rtptimeout setting
Ian Gilmour (JIRA)
noreply at issues.asterisk.org
Tue Sep 5 11:17:07 CDT 2017
[ https://issues.asterisk.org/jira/browse/ASTERISK-27251?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ian Gilmour updated ASTERISK-27251:
-----------------------------------
Description:
I have Asterisk running in the cloud with sip.conf configured with:
{noformat}
rtptimeout=60
rtpholdtimeout=300
rtpkeepalive=20
{noformat}
I have a confbridge managed conference room, configured to play MOH if there is only 1 participant.
I have a SIP client (Jitsi) configured to use TLS+SRTP. It registers with Asterisk on a chan_sip managed IP address + port.
The SIP client is behind a NAT.
If the SIP client enters the conference room MOH plays as expected. If I then disconnect the SIP client from the network (so it doesn't deregister, or issue a SIP BYE) I see Asterisk sending media (MOH) to the SIP client's orginal IP address and port indefinitely (>24hours+) and the CLI shows the user as present in the conf call. Media output only terminates if I kick the user manually out of the conf call via the CLI.
Once the SIP client is disconnected from the network wireshark shows ICMP Destination unreachable messages being returned in response to each Asterisk (MOH) outgoing media packet.
n.b. I see similar behaviour if I run both Asterisk and client locally. i.e. no NAT.
The bug seems to be related to ASTERISK-26523.
Reverting the ASTERISK-26523 change so that it only updates the lastrtprx if the frame isn't AST_FRAME_NULL gives me a chan_sip that honours the sip.conf rtptimeout.
was:
I have Asterisk running in the cloud with sip.conf configured with:
{noformat}
rtptimeout=60
rtpholdtimeout=300
rtpkeepalive=20
{noformat}
I have a confbridge managed conference room, configured to play MOH if there is only 1 participant.
I have a SIP client (Jitsi) configured to use TLS+SRTP. It registers with Asterisk on a chan_sip managed IP address + port.
The SIP client is behind a NAT.
If the SIP client enters the conference room MOH plays as expected. If I then disconnect the SIP client from the network (so it doesn't deregister, or issue a SIP BYE) I see Asterisk sending media (MOH) to the SIP client's orginal IP address and port indefinitely (>24hours+) and the CLI shows the user as present in the conf call. Media output only terminates if I kick the user manually out of the conf call via the CLI.
Once the SIP client is disconnected from the network wireshark shows ICMP Destination unreachable messages being returned in response to each Asterisk (MOH) outgoing media packet.
n.b. I see similar behaviour if I run both Asterisk and client locally. i.e. no NAT.
The bug seems to be related to Asterisk-26523.
Reverting the ASTERISK-26523 change so that it only updates the lastrtprx if the frame isn't AST_FRAME_NULL gives me a chan_sip that honours the sip.conf rtptimeout.
> chan_sip doesn't honour rtptimeout setting
> ------------------------------------------
>
> Key: ASTERISK-27251
> URL: https://issues.asterisk.org/jira/browse/ASTERISK-27251
> Project: Asterisk
> Issue Type: Bug
> Security Level: None
> Components: Channels/chan_sip/General
> Affects Versions: 13.17.1
> Environment: Ubuntu 16.04 (64-bit)
> Reporter: Ian Gilmour
>
> I have Asterisk running in the cloud with sip.conf configured with:
> {noformat}
> rtptimeout=60
> rtpholdtimeout=300
> rtpkeepalive=20
> {noformat}
> I have a confbridge managed conference room, configured to play MOH if there is only 1 participant.
> I have a SIP client (Jitsi) configured to use TLS+SRTP. It registers with Asterisk on a chan_sip managed IP address + port.
> The SIP client is behind a NAT.
> If the SIP client enters the conference room MOH plays as expected. If I then disconnect the SIP client from the network (so it doesn't deregister, or issue a SIP BYE) I see Asterisk sending media (MOH) to the SIP client's orginal IP address and port indefinitely (>24hours+) and the CLI shows the user as present in the conf call. Media output only terminates if I kick the user manually out of the conf call via the CLI.
> Once the SIP client is disconnected from the network wireshark shows ICMP Destination unreachable messages being returned in response to each Asterisk (MOH) outgoing media packet.
> n.b. I see similar behaviour if I run both Asterisk and client locally. i.e. no NAT.
> The bug seems to be related to ASTERISK-26523.
> Reverting the ASTERISK-26523 change so that it only updates the lastrtprx if the frame isn't AST_FRAME_NULL gives me a chan_sip that honours the sip.conf rtptimeout.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
More information about the asterisk-bugs
mailing list