David Vossel [Fri, 24 Apr 2009 21:22:31 +0000 (21:22 +0000)]
TLS/SSL private key option
Adds option to specify a private key .pem file when configuring TLS or SSL in AMI, HTTP, and SIP. Before this, the certificate file was used for both the public and private key. It is possible for this file to hold both, but most configurations allow for a separate private key file to be specified. Clarified in .conf files how these options are to be used. The current conf files do not explain how the private key is handled at all, so without knowledge of Asterisk's TLS implementation, it would be hard to know for sure what was going on or how to set it up.
Richard Mudgett [Fri, 24 Apr 2009 17:59:01 +0000 (17:59 +0000)]
There is no need to use the struct ast_party_connected_line.source update values.
The messages sent by a technology when a connected line update is received
are best determined by the current call state of the channel. The struct
ast_party_connected_line.source value is really only useful as a possible
tracing aid.
Russell Bryant [Fri, 24 Apr 2009 14:04:26 +0000 (14:04 +0000)]
Convert the ast_channel data structure over to the astobj2 framework.
There is a lot that could be said about this, but the patch is a big
improvement for performance, stability, code maintainability,
and ease of future code development.
The channel list is no longer an unsorted linked list. The main container
for channels is an astobj2 hash table. All of the code related to searching
for channels or iterating active channels has been rewritten. Let n be
the number of active channels. Iterating the channel list has gone from
O(n^2) to O(n). Searching for a channel by name went from O(n) to O(1).
Searching for a channel by extension is still O(n), but uses a new method
for doing so, which is more efficient.
The ast_channel object is now a reference counted object. The benefits
here are plentiful. Some benefits directly related to issues in the
previous code include:
1) When threads other than the channel thread owning a channel wanted
access to a channel, it had to hold the lock on it to ensure that it didn't
go away. This is no longer a requirement. Holding a reference is
sufficient.
2) There are places that now require less dealing with channel locks.
3) There are places where channel locks are held for much shorter periods
of time.
4) There are places where dealing with more than one channel at a time becomes
_MUCH_ easier. ChanSpy is a great example of this. Writing code in the
future that deals with multiple channels will be much easier.
Some additional information regarding channel locking and reference count
handling can be found in channel.h, where a new section has been added that
discusses some of the rules associated with it.
Mark Michelson also assisted with the development of this patch. He did the
conversion of ChanSpy and introduced a new API, ast_autochan, which makes it
much easier to deal with holding on to a channel pointer for an extended period
of time and having it get automatically updated if the channel gets masqueraded.
Mark was also a huge help in the code review process.
Thanks to David Vossel for his assistance with this branch, as well. David
did the conversion of the DAHDIScan application by making it become a wrapper
for ChanSpy internally.
The changes come from the svn/asterisk/team/russell/ast_channel_ao2 branch.
Labels are sometimes (most of the time?) NULL for extensions.
(closes issue #14895)
Reported by: chris-mac
Patches:
20090423__bug14895__2.diff.txt uploaded by tilghman (license 14)
Tested by: lmadsen
If both sides of a Local channel were hung up at around the same time it was
possible for one thread to destroy the local private structure and have the other thread
immediately try to remove the already freed structure from the local channel list.
........
Fix a double free issue with the Pickup dialplan application.
As part of the pickup process the connected line information is updated.
Part of this process does a shallow copy of the target channel's connected line
information to a local structure. Once complete the structure contents are freed.
As a result any information in the target channel's connected line information
structure is no longer valid. This change will now set the contents back to a clean
state so that the freeing of the target channel's connected line information structure
when the channel is destroyed will no longer try to double free things.
Detect availability of pthread_rwlock_timedwrlock() before using it.
(closes issue #14930)
Reported by: tilghman
Patches:
20090420__bug14930.diff.txt uploaded by tilghman (license 14)
Tested by: mvanbaak, tilghman
........
Jeff Peeler [Wed, 22 Apr 2009 21:15:55 +0000 (21:15 +0000)]
Fix building of chan_h323 with gcc-3.3
There seems to be a bug with old versions of g++ that doesn't allow a structure
member to use the name list. Rename list member to group_list in ast_group_info
and change the few places it is used.
Terry Wilson [Wed, 22 Apr 2009 20:07:41 +0000 (20:07 +0000)]
Add funcs for manipulating delimited lists in the dialplan
Adds PUSH and POP for appending to and retrieving/removing from the
end of a list and UNSHIFT and SHIFT for insert to and retrieiving/
removing from the beginning of a list.
Jeff Peeler [Wed, 22 Apr 2009 19:23:49 +0000 (19:23 +0000)]
Make chan_h323 respect packetization settings and fix small reload issue.
Previously, packetization settings were ignored and now they are not. A new
config option 'autoframing' has been added to mirror the way chan_sip handles
it. Turning on the autoframing option (available both as a global option or per
peer) overrides the local settings with the remote packetization settings.
Testing was performed with varying packetization levels with the following
codecs: ulaw, alaw, gsm, and g729.
Also, an unrelated config reload issue has been fixed in the case of the config
file not changing.
(closes issue #12415)
Reported by: pj
Patches:
2009012200_h323packetization.diff.txt uploaded by mvanbaak (license 7),
modified by me
Previously, packetization settings were ignored and now they are not. A new
config option 'autoframing' has been added to mirror the way chan_sip handles
it. Turning on the autoframing option (available both as a global option or per
peer) overrides the local settings with the remote packetization settings.
Testing was performed with varying packetization levels with the following
codecs: ulaw, alaw, gsm, and g729.
(closes issue #12415)
Reported by: pj
Patches:
2009012200_h323packetization.diff.txt uploaded by mvanbaak (license 7),
modified by me
........
Do not continue to receive DTMF, when the channel is hungup and about to be destroyed.
(closes issue #14858)
Reported by: barryf
Patches:
20090421__bug14858.diff.txt uploaded by tilghman (license 14)
Tested by: barryf
David Vossel [Tue, 21 Apr 2009 20:28:37 +0000 (20:28 +0000)]
Fixes segfault when switching UDP to TCP in sip.conf after reload.
If transport in sip.conf is switched from UDP to TCP, Asterisk segfaults right after issuing a sip reload. The problem is the socket type is changed to TCP but the fd may still be present for UDP. Later, when the TCP session should be created or set using an existing one, it isn't because the old file descriptor is still present. Now every time transport is changed during a sip.conf reload, the file descriptor is set to -1, signifying it must be created or found.
Richard Mudgett [Tue, 21 Apr 2009 17:44:01 +0000 (17:44 +0000)]
Added CCBS/CCNR Party A support and enhanced COLP support.
This change adds the following features to chan_misdn:
* CCBS/CCNR Party A support for PTMP and PTP modes.
* Enhances COLP support for call diversion and explicit call transfer.
These enhanced features require a modified version of mISDN.
The latest modified mISDN v1.1.x based version is available at:
http://svn.digium.com/svn/thirdparty/mISDN/trunk
http://svn.digium.com/svn/thirdparty/mISDNuser/trunk
Taged versions of the modified mISDN code are available under:
http://svn.digium.com/svn/thirdparty/mISDN/tags
http://svn.digium.com/svn/thirdparty/mISDNuser/tags
Add check in configure script to check for GLOB_NOMAGIC and GLOB_BRACE in glob.h
This allows config.c to compile when linked against uclibc that does not support these parameters
........
Add a workaround for func_odbc/ARRAY() for problems that occur with certain special characters.
In certain cases, due to the way Set() works in 1.4, values may not get set
properly. This is a workaround for 1.4 only that corrects for these issues,
without making func_odbc more difficult to use properly.
(closes issue #14614)
Reported by: wdoekes
Patches:
20090309__bug14614__2.diff.txt uploaded by tilghman (license 14)
double_set_unescape_workaround_for_func_odbc.osso-and-tilghman-1.diff uploaded by wdoekes (license 717)
Tested by: wdoekes, tilghman
........
AEL was not handling the case of a device hint containing an @ symbol, which
caused parking hints (e.g. hint(park:exten@context)) to error out the parser.
This patch makes AEL treat the @ the same way it treats colon and ampersand
now, meaning the characters are included in verbatim.
Clean up problem with manager implementation of mmap where it was not testing against MAP_FAILED response.
Got rid of shadowed variable used in processign the mmap results.
Change test of mmap results to compare against MAP_FAILED
........
Fix a bug with non-UDP connections that caused dialogs to not get freed.
This issue crept up because of a reference count issue on non-UDP based dialogs.
The dialog reference count was increased when transmitting a packet reliably but never
decreased. This caused the dialog structure to hang around despite being unlinked from
the dialogs container.
Fixed autologoff in agents.conf not working when agent logs in via AgentLogin app
An agent logs in by calling an extension that calls the AgentLogin app. In agents.conf ackcall=always is set, so when they get a call they have the choice to either acknowledge it or ignore it. autologoff=10 is set as well, so if the agent ignores the call over 10sec one may assume that the agent should be logged out (and in this case hungup on as well), but this was not happening.
Mark Michelson [Fri, 17 Apr 2009 20:20:23 +0000 (20:20 +0000)]
Prevent a crash when SIP blonde transferring an unbridged call.
If one attempts to use the attended transfer button on a SIP phone
to transfer an unbridged call (such as a call to an IVR) but hangs
up while the target of the transfer is still ringing, we need to not
crash.
The problem was that ast_hangup was called from outside the channel
thread.
Fix a bug where a value used to create the channel name was bogus.
This commit fixes the scenario where an incoming call is authenticated
using a peer entry. Previously the channel name was created using either
the username setting from the sip.conf entry or the IP address that the
call came from. Now the channel name will be created using the peer name
itself. This commit will not change the way the channel name is generated
for users or friends.
(closes issue #14256)
Reported by: Nick_Lewis
Patches:
chan_sip.c-chname.patch uploaded by Nick (license 657)
Tested by: Nick_Lewis, file
........
Mark Michelson [Fri, 17 Apr 2009 13:29:33 +0000 (13:29 +0000)]
Several fixes to the extenpatternmatchnew logic.
1. Differentiate between literal characters in an extension
and characters that should be treated as a pattern match. Prior to
these fixes, an extension such as NNN would be treated as a pattern,
rather than a literal string of N's.
2. Fixed the logic used when matching an extension with a bracketed
expression, such as 2[5-7]6.
3. Removed all areas of code that were executed when NOT_NOW was
#defined. The code in these areas had the potential to crash, for
one thing, and the actual intent of these blocks seemed counterproductive.
4. Fixed many many coding guidelines problems I encountered while looking
through the corresponding code.
5. Added failure cases and warning messages for when duplicate extensions
are encountered.
6. Miscellaneous fixes to incorrect or redundant statements.
David Vossel [Thu, 16 Apr 2009 19:30:23 +0000 (19:30 +0000)]
SIP state notify reorganization
What I've done here is simply break up how a state NOTIFY is built. Originally both the XML and sip header information were built within the same function. While this does work, it does not allow for the creation of multipart/related message bodies that can contain multiple XML entries with only one sip header. Now a separate function builds the XML for each notify. This patch also makes maintaining and modifying state notifications in the future much less of a pain.
Fix a bug with the dahdi_setoption callback in chan_dahdi.
This function incorrectly reported success even if the option was
unsupported. This was exposed by the options to change the underlying
channel format. The function now returns a failure if the option
is unsupported.
National prefix inserted even when caller ID not available
When the caller ID is restricted, the expected behavior is for the caller id to be blank. In chan_dahdi, the national prefix is placed onto the callers number even if its restricted (empty) causing the caller id to be the national prefix rather than blank.
Mark Michelson [Wed, 15 Apr 2009 15:24:50 +0000 (15:24 +0000)]
Make the cancellation of the dial timeout on a call forward optional.
This introduces the 'z' option to app_dial. With it set, a call forward
will cancel any timeout originally set for this instance of the Dial
application.
Jeff Peeler [Tue, 14 Apr 2009 15:54:16 +0000 (15:54 +0000)]
Add service maintenance message support
This is the companion commit to libpri r732. Service messages are now supported
for switch types 4ess/5ess. A new option service_message_support has been added
to chan_dahdi.conf and is noted in the sample config file. The service message
support is turned off by default. The current implementation relies on AstDB
to keep track of channel state, which allows the statuses to be preserved
across Asterisk restarts. Below is a description of the storage format.
The state and reason for the service state are in the form <state>:<reason>,
where:
<state> ::= { 'O' } // 'O' – Out Of Service
<reason> ::= { '0' | '1' | '2' | '3' }, where:
'0' – No reason (backwards compatibility)
'1' – NEAR END
'2' – FAR END
'3' – both NEAR and FAR END
The new CLI commands to handle channel service state are:
pri service disable channel <chan>
pri service enable channel <chan>
Many people contributed to the development of this functionality. Because I
entered at the very end I do not know the exact history. Special thanks to
all who moved the bug forward one way or another:
cmaj, PCadach, markster, mattf, drmac, MikeJ, serge-v, murf, kanelbullar, Seb7,
tilghman, lmadsen, and especially dhubbard (he answered lots of my questions
and did a large portion of the work)
Fix a bug with the change I made yesterday to outbound proxy support.
Per discussion with oej on IRC we need the actual IP address, not the
outbound proxy IP address, in the sa field. This change matches the already
existing code for all other uses of the outbound proxy setting.
Fix a bug where using an outbound proxy would cause the local address to be 127.0.0.1.
Copy the outbound proxy IP address into the SIP dialog structure as the IP address we will
be sending to. This has to be done because the logic that determines what local IP address to use
in the SIP messages is not aware of an outbound proxy being in place. It only knows what IP address
we are sending to.
David Vossel [Fri, 10 Apr 2009 15:49:16 +0000 (15:49 +0000)]
Even more changes concerning r187426. Revised where locks are placed yet once again. ast_call() should not be called with a channel locked. could cause deadlock issues with local channels.
Mark Michelson [Thu, 9 Apr 2009 21:06:26 +0000 (21:06 +0000)]
Add a new option, mwi_from, to sip.conf.
This allows for you to change the From header for outgoing MWI
NOTIFY requests. Prior to this, the best you could do was to
set a callerid in the general section of sip.conf. The problem
was that this was used for all outbound requests, not just
MWI NOTIFY requests.
Jeff Peeler [Thu, 9 Apr 2009 19:10:02 +0000 (19:10 +0000)]
Add ability for dialplan execution to continue when caller hangs up.
The F option to app_dial has been modified to accept no parameters and perform
the above functionality. I don't see anywhere else that is doing function
overloading, but this really is the best place for this operation because:
- It makes it close to the 'g' option in the argument list which provides
similar functionality.
- The existing code to support the current F option provides a very
convienient location to add this new feature.
Handle a SIP race condition (reinvite before an ACK) properly.
RFC 5047 explains the proper course of action to take if a
reINVITE is received before the ACK from a previous invite
transaction. What we are to do is to treat the reINVITE as
if it were both an ACK and a reINVITE and process it normally.
Later, when we receive the ACK we had been expecting, we will
ignore it since its CSeq is less than the current iseqno of
the sip_pvt representing this dialog.
Race condition between ast_cli_command() and 'module unload' could cause a deadlock.
Add lock timeouts to avoid this potential deadlock.
(closes issue #14705)
Reported by: jamessan
Patches:
20090320__bug14705.diff.txt uploaded by tilghman (license 14)
Tested by: jamessan
........
David Vossel [Thu, 9 Apr 2009 17:39:10 +0000 (17:39 +0000)]
Fixes deadlock caused by calling get_cid_name with chan locked.
get_cid_name should not be called with a channel lock. get_cid_name calls ast_get_hint which eventually calls pbx_find_extension. pbx_find_extension starts and stops autoservice which should not be done with a channel lock, so get_cid_name should not be called with one.
Mark Michelson [Thu, 9 Apr 2009 17:30:39 +0000 (17:30 +0000)]
Fix a crash in res_musiconhold when using cached realtime moh.
The moh_register function links an mohclass and then immediately
unrefs the class since the container now has a reference. The problem
with using realtime music on hold is that the class is allocated,
registered, and started in one fell swoop. The refcounting logic
resulted in the count being off by one. The same problem did not
happen when using a static config because the allocation and registration
of an mohclass is a separate operation from starting moh. This also did
not affect non-cached realtime moh because the classes are not registered
at all.
I also have modified res_musiconhold to use the _t_ variants of the ao2_
functions so that more info can be gleaned when attempting to trace the
refcounts. I found this to be incredibly helpful for debugging this issue
and there's no good reason to remove it.
Add support for allowing the channel driver to handle transcoding.
This was accomplished using a set of options and the setoption channel callback.
The core calls into the channel driver using these options and the channel driver
either returns success or failure.
add a dedicated log channel for modules to be able report security-related events, so that they can be fed into external processes for analysis and possible mitigation efforts
(inspired by this evening's Toronto Asterisk Users Group meeting and previous dicussions amongst various community members)
Jeff Peeler [Wed, 8 Apr 2009 21:00:39 +0000 (21:00 +0000)]
Add timer for features so that backup bridge config can go away
The biggest change done here was elimination of the backup_config for use with
features. Previously, the bridging code upon detecting a feature would set the
start time of the bridge to the start time of the feature. Then after the
feature had either expired or timed out the start time would be reset to the
true bridge start time from the backup_config. Now, the time differences are
calculated with respect to the newly added feature_start_time timeval instead.
There should be no behavior changes from the previous functionality aside from
the bridge timing being unaffected by either valid or partial feature matches.
Previously the timing would be increased by the length of time configured for
featuredigittimeout, which was probably never noticed.
projects / thirdparty / asterisk.git / log
