Summary: | ASTERISK-05115: 486 is now congestion rather than busy in sip_chan.c version 1.841 | ||
Reporter: | Andrew Lindh (andrew) | Labels: | |
Date Opened: | 2005-09-19 13:10:08 | Date Closed: | 2011-06-07 14:10:49 |
Priority: | Minor | Regression? | No |
Status: | Closed/Complete | Components: | Channels/chan_sip/General |
Versions: | Frequency of Occurrence | ||
Related Issues: | |||
Environment: | Attachments: | ( 0) 5257.patch.txt | |
Description: | When I call out using SIP and dial a truely busy POTS number I get a 486 Busy Here from my cisco gateway...that's correct. Asterisk used to return BUSY in the DIALSTATUS to the dial plan, now it returns CONGESTION even when the gateway says "user busy". Looking at the code (channels/chan_sip.c) there was a change in version 1.841 The SIP return codes 486/488/603/606 used to set AST_CONTROL_BUSY, but now are falling through to contestion (about line 9500 or so in chan_sip.c). I call this a bug as it no longer returns busy when the called number is busy. I'm not sure response code 488 should be busy, it looks more like congestion to me. But 486, 600, 603 look like (and cisco says 486 and 600) should be user busy. ****** ADDITIONAL INFORMATION ****** fragment that I think should be in chan_sip.c case 486: /* Busy here */ case 488: /* Not acceptable here - codec error */ case 603: /* Decline */ case 600: /* Busy everywhere */ if (owner) ast_queue_control(p->owner, AST_CONTROL_BUSY); break; | ||
Comments: | By: Michael Jerris (mikej) 2005-09-25 21:35:59 oej- it appears to me that this portion of the invite response patch was in error: case 486: /* Busy here */ + case 488: /* Not acceptable here - codec error */ case 600: /* Busy everywhere */ case 603: /* Decline */ - if (p->owner) - ast_queue_control(p->owner, AST_CONTROL_BUSY); - break; Thoughts? By: Michael Jerris (mikej) 2005-09-25 21:54:30 tossed a patch out there that adds AST_CONTROL_BUSY in for 480, 486, 600, and 603, and adds a break in after the 482 so it does not get an AST_CONTROL_CONGESTION or BUSY which appears to be unintended. Somone is going to need to do some testing and verification on this please. By: Mark Spencer (markster) 2005-09-28 00:10:19 I agree it's an error, fixed in CVS head. By: Digium Subversion (svnbot) 2008-01-15 15:49:26.000-0600 Repository: asterisk Revision: 6674 U trunk/channels/chan_sip.c ------------------------------------------------------------------------ r6674 | markster | 2008-01-15 15:49:26 -0600 (Tue, 15 Jan 2008) | 2 lines 486 (and friends) are really BUSY (bug ASTERISK-5115) ------------------------------------------------------------------------ http://svn.digium.com/view/asterisk?view=rev&revision=6674 |