[asterisk-bugs] [JIRA] (ASTERISK-26048) Asterisk crashes with PJSIP Assertion "Invalid transport name"
Erico Mattos (JIRA)
noreply at issues.asterisk.org
Fri May 20 15:08:56 CDT 2016
Erico Mattos created ASTERISK-26048:
---------------------------------------
Summary: Asterisk crashes with PJSIP Assertion "Invalid transport name"
Key: ASTERISK-26048
URL: https://issues.asterisk.org/jira/browse/ASTERISK-26048
Project: Asterisk
Issue Type: Bug
Security Level: None
Components: pjproject/pjsip
Affects Versions: 13.9.1, 13.7.1
Environment: Kernel: 2.6.32-504.8.1.el6.x86_64
FreePBX 13.0.120
Before update:
FreePBX Distro: 10.13.66-9
Asterisk 13.7.1
After update:
FreePBX Distro: 10.13.66-12
Asterisk 13.9.1
Reporter: Erico Mattos
Severity: Critical
Relevant scenario:
- An incomming call is redirected to a group with an extension using transport TCP or TLS
The following message showes on stderr and then crashes asterisk:
asterisk: ../src/pjsip/sip_transport.c:299: pjsip_transport_get_type_from_name: Assertion `!"Invalid transport name"' failed.
On the stdout (before the stderr message) using TCP:
Executing [s at macro-dial:17] Dial("PJSIP/NET-6238770660-00000003", "PJSIP/401/sip:401 at 10.62.10.151:5061;transport=CP,20,trM(auto-blkvm)b(func-apply-sipheaders^s^1),") in new stack
On the stdout (before the stderr message) using TLS:
Executing [s at macro-dial:17] Dial("PJSIP/NET-6238770660-00000002", "SIP/401&PJSIP/402/sip:402 at 191.190.6.203:5061;transport=LS,20,trM(auto-blkvm)b(func-apply-sipheaders^s^1),") in new stack
On both situation the first letter "T" is not shown.
If I use UDP the error doesn't occurs since que transport is not appended.
Full scenario (to understand the log attached):
- Incomming call cames from a trunk on a SIP UDP gateway
- Call is redirected to a IVR with a announcement
- The option selected (4) is associated to a Ring Group (400)
- The Ring group has two extensions, one offline and one online with TCP transport
- Crashes when option 4 is selected
--
This message was sent by Atlassian JIRA
(v6.2#6252)
More information about the asterisk-bugs
mailing list