Sean Bright [Tue, 27 Jul 2010 21:57:07 +0000 (21:57 +0000)]
Add ability to use system libedit and update bundled libedit.
The version of libedit that is bundled with asterisk is old and has some bugs.
This patch updates the bundled version of libedit within asterisk, and also
updates asterisk to use the system libedit instead if one is available (and
pkg-config is available). This review integrates several patches from other
users specifically kkm and tzafrir.
If a channel has an audiohook write list created on it, that
list stays on the channel until the channel is destroyed. There
is no reason to keep that list on the channel if it becomes empty.
If it is empty that just means we are doing needless translating
for every ast_read and ast_write. This patch removes the audiohook
list from the channel once it is detected to be empty on either a
read or write. If a audiohook is added back to the channel after
this list is destroyed, the list just gets recreated as if it never
existed to begin with.
Russell Bryant [Tue, 27 Jul 2010 19:50:56 +0000 (19:50 +0000)]
Fix inband DTMF detection on outgoing ISDN calls.
This is a regression from the sig_pri split from chan_dahdi. When a call is
first initiated, the inband DTMF detector is not enabled if it's an outgoing
ISDN call. However, it needs to be turned on once the media path starts up.
This handling was put back in the open_media() callback of chan_dahdi. In
sig_pri, open_media() calls were added to a few places where it was needed,
including handling of PRI_EVENT_RINGING, PRI_EVENT_PROGRESS, and
PRI_EVENT_PROCEEDING.
Mark Michelson [Tue, 27 Jul 2010 18:54:07 +0000 (18:54 +0000)]
Fix parsing error in sip_sipredirect().
The code was written in a way that did a bad job of
parsing the port out of a URI. Specifically, it would
do badly when dealing with an IPv6 address. In this
particular scenario, there was no value from parsing
the port out, so I just removed that logic. And while
I was messing around in the function, I changed some
variable names to be more descriptive.
Fix bad behavior of dynamic_exclude_static option in sip.conf.
We were attempting to create a contactdeny rule based on the peer's
IP address before the peer's IP address had been set. By moving the
processing further down in the function, we can ensure stuff works
as we expect for it to.
David Vossel [Mon, 26 Jul 2010 19:59:03 +0000 (19:59 +0000)]
transaction matching using top most Via header
This patch modifies the way chan_sip.c does transaction to dialog
matching. Asterisk now stores information in the top most Via header
of the initial incoming request and compares that against other Requests
that have the same call-id. This results in Asterisk being able to
detect a forked call in which it has received multiple legs of the
fork. I completely stripped out the previous matching code and made
the comparisons a little more explicit and easier to understand. My
comments in the code should offer all the details involving this patch.
This patch also fixes a bug with the usage of the OBJ-MULTIPLE flag to
find multiple dialogs with the same call-id. Since the callback
function was returning (CMP_MATCH | CMP_STOP) only the first item
found was being returned. I fixed this by making a new callback
function for finding multiple dialogs that only returns (CMP_MATCH)
on a match allowing for multiple items to be returned.
Mark Michelson [Mon, 26 Jul 2010 16:04:09 +0000 (16:04 +0000)]
Allow for systems without locale support to be usable.
A recent change to SIP URI comparison code added a locale-specific
string comparison to the mix, and certain systems do not support
such functions. This fix allows for those systems to still use
Asterisk 1.8
SIP promiscuous redirect could fail to dial the redirect.
The ast_channel was created with one variable to ast_request() but the
call to ast_call() that initiates the outgoing call was using a different
variable. The two variables are not equivalent if the call_forward string
included a channel technology specifier. e.g., SIP/200
........
................
Mark Michelson [Fri, 23 Jul 2010 16:33:52 +0000 (16:33 +0000)]
SIP URI comparison fixes.
This initially was created to work around the issue of
using a string comparison instead of a binary comparison
for IP addresses. It evolved a bit when test cases were
created and it was discovered that comparison of URI
parameters was not working exactly as it should.
sip_uri_cmp() and its helpers have been moved to reqresp_parser.c
and a new test has been added.
Paul Belanger [Fri, 23 Jul 2010 13:26:41 +0000 (13:26 +0000)]
Portability updates for Makefiles.
When possible, use $(INSTALL). This allows us to use the functionality within
install for setting directory / file permissions, a requirement for unprivileged
installation.
Also move any directory we plan to create within the installdirs macro. Plus
various other formatting issues.
Alec L Davis [Thu, 22 Jul 2010 23:14:50 +0000 (23:14 +0000)]
Support FXS module Polarity Reversal on remote party Answer and Hangup
FXS lines normally connect to a telephone. However, when FXS lines are routed
to an external PBX or Key System to act as "external" or "CO" lines, it is
extremely difficult, if not impossible for the external PBX to know when
the call has been disconnected without receiving a polarity reversal on the line.
Now using answeronpolarityswitch and hanguponpolarityswitch keywords that
previously were used only for FXO ports, now applies like functionality for
an FXS port, but from the connected equipment's point of view.
Richard Mudgett [Thu, 22 Jul 2010 21:16:04 +0000 (21:16 +0000)]
DNID not cleared when channel hang up (Affects PRI and SS7)
The "dahdi show channels" CLI command still reports the DNID of the
previous call even if the call is already hang up. The "dahdi show
channels" command of older releases clear the DNID once the channel is
hang up.
Regression from the sig_analog/sig_pri extraction from chan_dahdi.
Jeff Peeler [Thu, 22 Jul 2010 19:45:30 +0000 (19:45 +0000)]
Add method for finding XML doc files for systems that don't support GLOB_BRACE.
In particular, Solaris and perhaps others do not support the above mentioned
GNU extension. In this case the paths are simply expanded without the braces
and the calls to glob are made separately.
Note: I could not explain memory allocation failures that were being reported
from within libxml itself when making calls to glob without using GLOB_NOCHECK.
This is the only reason why that flag is being used.
(closes issue #15402)
Reported by: snuffy
Patches:
bug_xmlpatt-v3.diff uploaded by snuffy (license 35),
modified by me
Allow PLC to function properly when channels use SLIN for audio.
If a channel involved in a bridge was using SLIN audio, then translation
paths were not guaranteed to be set up properly since in all likelihood
the number of translation steps was only 1.
This patch enforces the transcode_via_slin behavior if transcode_via_slin
or generic_plc is enabled and one of the formats to make compatible is
SLIN.
David Vossel [Wed, 21 Jul 2010 18:52:14 +0000 (18:52 +0000)]
send "423 Interval too small" Response to Subscribe with Expires less that min allowed
[RFC3265]3.1.6.1....
The notifier MAY also check that the duration in the "Expires" header
is not too small. If and only if the expiration interval is greater
than zero AND smaller than one hour AND less than a notifier-
configured minimum, the notifier MAY return a "423 Interval too
small" error which contains a "Min-Expires" header field. The "Min-
Expires" header field is described in SIP [1].
This fixes some cases of no outgoing calls on FXO before an incoming call.
Remove an unnecessary testing of an "off-hook" bit from DAHDI for FXO
(KS/GS) channels.In some cases the bit would not be initialized properly
before the first inbound call and thus prevent an outgoing call.
If those tests are actually required by anybody, they should define
DAHDI_CHECK_HOOKSTATE in channels/sig_analog.c .
(closes issue #14577)
Reported by: jkroon
Patches:
asterisk_chan_dahdi_hookstate_fix_trunk_new.diff uploaded by frawd (license 610)
Tested by: frawd
Russell Bryant [Wed, 21 Jul 2010 16:15:00 +0000 (16:15 +0000)]
Use poll() instead of select() in res_timing_pthread to avoid stack corruption.
This code did not properly check FD_SETSIZE to ensure that it did not try to
select() on fds that were too large. Switching to poll() removes the limitation
on the maximum fd value.
Russell Bryant [Wed, 21 Jul 2010 13:02:46 +0000 (13:02 +0000)]
Update documentation for 'comebacktoorigin' in featuers.conf.
The documentation for this option did not match the code. Fix that along with
some minor cleanups to the code along the way. Document a slight change in
behavior (to something that was previously undocumented) in UPGRADE.txt.
David Vossel [Tue, 20 Jul 2010 21:41:21 +0000 (21:41 +0000)]
fixes sip CANCEL race condition
If Asterisk sends a 4xx error and the other side sends a CANCEl
before receiving the 4xx and responding with the ACK, Asterisk
will process the CANCEL and send a 487 Request Terminated as
a new final response to the INVITE. Since we are issuing a new
final response to the INVITE, the old one must be pretend_acked
else it will keep retransmitting.
This commit contains several changes to the way output channel variables are handled.
FAX output channel variables will now match the values reported by FAXOPT() and should be set in all failure and success cases.
This commit also contains a few modifications to the way FAXOPT() variables are populated in a few spots and fixes for some reference count leaks of the session details structure in some failure cases.
Also found and fixed more cases where FAXOPT(status) may not have gotten set.
Avoid trying to pickup a parked extension before the park operation is completed.
A crash could occur if the extension is picked up while the parking extension is
being announced. Testing pu->notquiteyet while searching for a parked extension
resolves this crash.
Mark Michelson [Mon, 19 Jul 2010 17:16:23 +0000 (17:16 +0000)]
Fix port setting of external address in SIP.
There are two changes here:
1. Since the externip setting can now have a port attached
to it, calling it "externip" is misleading. The option is now
documented and parsed as "externaddr." This also extends to the
"matchexterniplocally" setting. It is now documented and parsed
as "matchexternaddrlocally." The old names for the options may
still be used, but they are no longer used in the sip.conf.sample
file.
2. If no port is set for the externaddr, and UDP is the transport
to be used, then we will set the port of the externaddr to that of
the udpbindaddr. This was how things worked prior to the IPv6 merge,
so this is a regression fix.
Mark Michelson [Mon, 19 Jul 2010 17:10:00 +0000 (17:10 +0000)]
Remove the fe80:1234::1234 test case from test_acl.c
The ACL test was failing on Mac OS X because it would
convert the above invalid link-local address into
fe80::1234 while reporting no error from getaddrinfo().
Linux does not do this.
Jeff Peeler [Mon, 19 Jul 2010 14:39:07 +0000 (14:39 +0000)]
Fix regression with distinctive ring detection.
The issue here is that passing an array to a function prohibits the ARRAY_LEN
macro from returning the real size. To avoid this the size is now defined and
use of ARRAY_LEN is avoided.
Mark Michelson [Mon, 19 Jul 2010 14:17:16 +0000 (14:17 +0000)]
Make ACLs IPv6-capable.
ACLs can now be configured to match IPv6 networks. This is only
relevant for ACLs in chan_sip for now since other channel drivers
do not support IPv6 addressing. However, once those channel drivers
are outfitted to support IPv6 addressing, the ACLs will already be
ready for IPv6 support.
Save and restore AST_FLAG_BRIDGE_HANGUP_DONT on attended transfer.
ast_bridge_call() clears AST_FLAG_BRIDGE_HANGUP_DONT. But during an attended
transfer, ast_bridge_call() is called for a second bridge on the same channel,
and it clears that flag, which still needs to get set for when the original
ast_bridge_call() gets control back and checks it.
Default to no udptl error correction so that error correction will be disabled in the event that the remote end indicates that they do not support the error correction mode we requested.
priexclusive in chan_dahdi.conf ignored when reloading dahdi module
During a reload, the priexclusive and outsignalling parameters are not
read in from the config file as intended. Unfortunately, they get set to
defaults as a result. This patch makes sure that they do not get set to
defaults during a reload.
(closes issue #17441)
Reported by: mtryfoss
Patches:
issue17441_v1.4.patch uploaded by rmudgett (license 664)
issue17441_v1.6.2.patch uploaded by rmudgett (license 664)
issue17441_trunk.patch uploaded by rmudgett (license 664)
Tested by: rmudgett
........
projects / thirdparty / asterisk.git / log
