Summary: | ASTERISK-21011: Asterisk to Asterisk IAX2 trunk registration does not register | ||||
Reporter: | David Brillert (aragon) | Labels: | |||
Date Opened: | 2013-01-29 10:15:53.000-0600 | Date Closed: | 2013-02-15 10:35:04.000-0600 | ||
Priority: | Major | Regression? | |||
Status: | Closed/Complete | Components: | Channels/chan_iax2 | ||
Versions: | 11.2.1 | Frequency of Occurrence | Constant | ||
Related Issues: |
| ||||
Environment: | Attachments: | ( 0) 192.168.192.1.txt ( 1) 192.168.192.74.txt ( 2) iax2_asterisk_to_asterisk_failure.txt ( 3) iax2_configurations.txt ( 4) iax2_debug_11.0.1.txt | |||
Description: | Asterisk dynamic server does not update iax2 show peers or register remote/static Asterisk server.
iax2 set debug on trace uploaded from both servers. | ||||
Comments: | By: David Brillert (aragon) 2013-01-29 10:20:15.678-0600 iax.conf from both systems By: David Brillert (aragon) 2013-01-29 10:20:55.553-0600 FYI Asterisk to Asterisk fails registration. But Voixphone to Asterisk registration succeeds. By: David Brillert (aragon) 2013-01-29 10:24:00.732-0600 This could be a duplicate of ASTERISK-20963 By: Matt Jordan (mjordan) 2013-01-29 12:27:24.834-0600 Fairly sure this is. Does this occur if both iax2 peers are defined statically? By: David Brillert (aragon) 2013-01-29 12:58:44.446-0600 If I set 192.168.192.1 to specific and remove registration from 192.168.192.74 the peers become available after a reload. I am not using realtime. By: Rusty Newton (rnewton) 2013-02-05 18:40:44.550-0600 The reporter in ASTERISK-20963 saw their issue in 1.8.19.0-rc1 and later. It looks like 11.1.0-rc1 was released at the same time as 1.8.19.0-rc1. Can you revert to a version previous to 11.1.0-rc1 ? and see if the issue no longer occurs? Perhaps 11.0.2 ? By: David Brillert (aragon) 2013-02-06 07:58:20.465-0600 Downgrading to 11.0.1 fixes the IAX2 registration problem. Downgrading to 1.8.18 fixes the IAX2 registration problem. By: David Brillert (aragon) 2013-02-07 11:44:25.897-0600 iax2 debug 11.0.1.txt shows a working trace and iax.conf if I downgrade to to 11.0.1. The same configs do not work in 11.2 By: Jonathan Rose (jrose) 2013-02-07 15:36:44.828-0600 The symptoms and causes are similar. For the symptoms at least though, they aren't quite the same. In ASTERISK-20963, there is no traffic from Asterisk to the consumer. Here, the registration just isn't working and the actual flow follows much more closely to what I would expect in general. I think I'm going to shift my focus over here, at least until I get some more feedback on 20963. By: Jonathan Rose (jrose) 2013-02-07 16:42:17.579-0600 I have a couple things to request from you as well. 1. Complete iax.conf including general sections, register lines, etc for both machines. Just the peer definitions won't cut it unfortunately. I assumed that you included a line like: register => iax2testprotocol:scrubbed@192.168.192.1 in 192.168.192.74's config since the peer definition for 192.168.192.74 on the other machine was dynamic, but so far I still haven't been able to reproduce the bug using these configurations. 2. Output of the following command on each machine: module show like res_timing -- there were only two changes that really touched chan_iax2 over the revisions between the one that the bug is reported as present in and the one that it isn't. One of them seemed really innocuous and self-contained. The other was timing related, so aligning which timing modules we are both using might to isolate it. By: David Brillert (aragon) 2013-02-07 21:11:22.413-0600 Here is all the information you asked for but there is a catch... With the same configuration files and after reinstalling 11.2.1 I can no longer reproduce the problem :P As far as the timing module is concerned we normally delete all but res_timing_dahdi.so so, sip*CLI> module show like res_timing Module Description Use Count res_timing_dahdi.so DAHDI Timing Interface 1 1 modules loaded By: Rusty Newton (rnewton) 2013-02-15 10:35:04.377-0600 We believe this is a duplicate of ASTERISK-20963. Since it can't be reproduced now, we'll close this and David can watch the progress on ASTERISK-20963. |