On 09/22/2014 02:20 PM, Trever L. Adams wrote:
This morning I upgraded my test system to F21 Alpha. I need to start working with it for various reasons. The FreeSWITCH version is largely the same as was working Friday/Saturday under F20 up to det.
Now, I get similar to the following whenever I do a SIP -> FreeTDM or FreeTDM -> SIP call.
SIP or FreeTDM to 9664 (music test) yields no problems. SIP to 9787 (zrtp enrollment) yields no problems.
2014-09-22 14:06:30.189605 [DEBUG] switch_ivr_bridge.c:660 BRIDGE THREAD DONE [sofia/internal/trevera@highcountry.snowyriver.sapphiresunday.org] 2014-09-22 14:06:30.189605 [DEBUG] switch_core_state_machine.c:498 (FreeTDM/1:5/) State Change CS_REPORTING -> CS_DESTROY 2014-09-22 14:06:30.189605 [DEBUG] switch_ivr_bridge.c:690 Send signal FreeTDM/1:5/ [BREAK] 2014-09-22 14:06:30.189605 [DEBUG] switch_core_session.c:1388 Send signal FreeTDM/1:5/ [BREAK] 2014-09-22 14:06:30.189605 [DEBUG] switch_core_session.c:1615 Session 2 (FreeTDM/1:5/) Locked, Waiting on external entities 2014-09-22 14:06:30.189605 [DEBUG] switch_core_session.c:908 Send signal sofia/internal/trevera@highcountry.snowyriver.sapphiresunday.org [BREAK] 2014-09-22 14:06:30.189605 [NOTICE] switch_core_session.c:1633 Session 2 (FreeTDM/1:5/) Ended 2014-09-22 14:06:30.189605 [NOTICE] switch_core_session.c:1637 Close Channel FreeTDM/1:5/ [CS_DESTROY] 2014-09-22 14:06:30.189605 [DEBUG] switch_core_state_machine.c:626 (FreeTDM/1:5/) Running State Change CS_DESTROY 2014-09-22 14:06:30.189605 [DEBUG] switch_core_state_machine.c:636 (FreeTDM/1:5/) State DESTROY 2014-09-22 14:06:30.189605 [INFO] switch_core.c:2555 Shutting down 2014-09-22 14:06:30.189605 [INFO] switch_core.c:2555 Shutting down 2014-09-22 14:06:30.189605 [INFO] switch_core.c:2555 Shutting down
The last continues ad infititium. The FreeTDM (FXS) is effectively unusable after this as no dialtone, etc.
This problem exists from remotes/origin/v1.4 04789ae35f18d0d498fc52d2448bc9ae17d9309c and master 4dc7f92aff804447154fde712a312eda67a2e97b.
gcc version 4.9.1 20140912 (Red Hat 4.9.1-9) (GCC) glibc-2.20-2.fc21.x86_64 kernel-3.16.3-300.fc21.x86_64 openvox_dahdi-linux-complete-2.9.2+2.9.2 w/ patch in https://issues.asterisk.org/jira/browse/DAHLIN-340 for 3.16.2 and later.
Any one else seeing this and figure it out? I did have to remove the warning from features.h about _BSD_SOURCE and _SVID_SOURCE so that it would compile (all warnings treated as errors). This is due to those no longer being supported. The macro, besides warning, defines _DEFAULT_SOURCE automatically so that all compilation should stay the same. I only removed the warning.
Thank you for any help. Hopefully, this will be a heads up to those who can hold off on moving one or more systems over.
Trever
I was just enabling toll free numbers for certain sip accounts and noticed as part of testing the following: FXS -> FXO (some external number) doesn't go weird on hangup. SIP -> FXO also doesn't go weird on hangup. The problem exists pure sip (internal both) and with SIP or FXS calling SIP or FXS. I suppose this means the problem is either not DAHDI alone, or is very complex. Thanks, Trever