Russell Bryant [Tue, 26 May 2009 18:30:04 +0000 (18:30 +0000)]
Merged revisions 196843 via svnmerge from
https://origsvn.digium.com/svn/asterisk/trunk
................
r196843 | russell | 2009-05-26 13:20:57 -0500 (Tue, 26 May 2009) | 16 lines
Merged revisions 196826 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4
........
r196826 | russell | 2009-05-26 13:14:36 -0500 (Tue, 26 May 2009) | 9 lines
Resolve a file handle leak.
The frames here should have always been freed. However, out of luck, there was
never any memory leaked. However, after file streams became reference counted,
this code would leak the file stream for the file being read.
In sip.conf the transport option allows for the configuration of what transport types (udp, tcp, and tls) a peer will accept, but only the first type listed was used for outbound connections. This patch changes this. Now the default transport type is only used until the peer registers. When registration takes place the transport type is parsed out of the Contact header. If the Contact header's transport type is equal to one that the peer supports, the peer's default transport type for outbound connections is set to match the Contact header's type. If the Contact header's transport type is not present, then the peer's default transport type is set to match the one the peer registered with. When a peer unregisters or the registration expires, the default transport type for that peer is reset.
Sign problem calculating timestamp for iax frame leads to no audio on the receiving peer.
There are rare cases in which a frame's delivery timestamp is slightly less than the iax2_pvt's offset. This causes the pvt's timestamp to be a small negative number, but since the timestamp value is unsigned it looks like a huge positive number. This patch checks for this negative case and sets the ms to zero. A similar check is already done right below this one in the 'else' statement.
This commit prevents cdr records with AST_CDR_FLAG_ANSLOCKED and AST_CDR_FLAG_LOCKED from being updated in certain cases.
This is accomplished by adding two functions to update the answer time and disposition of calls that checks for the proper lock flags. These functions are used in the ast_bridge_call() function so that ForkCDR(A) calls are respected.
This patch also modifies the way ast_bridge_call() chooses the cdr record to base the bridged_cdr on. Previously the first unlocked cdr record would be chosen, now instead the first cdr record is chosen and forked cdr records are moved to the bridge_cdr. This allows the original cdr record and any forked cdr records to be properly updated with answer and end times.
If a variable had a blank value upon the initial setting, then it would do nothing.
Identified by Dmitry Andrianov via private email, fixed by me.
........
Fix the CLI command 'manager show command' documentation and functionality.
The CLI command 'manager show command' supports passing multiple action names in
the same line, but it was not allowing that because of a incorrect check in the
argumentes counter. Also the documentation was updated to show that this usage
of the command is possible.
........
Add a similar dependency on SMDI for voicemail as already exists for ADSI.
(closes issue #14846)
Reported by: pj
Patches:
20090413__bug14846__1.4.diff.txt uploaded by tilghman (license 14)
20090507__issue14846__1.6.0.diff.txt uploaded by tilghman (license 14)
20090507__issue14846__1.6.1.diff.txt uploaded by tilghman (license 14)
........
................
Move the spawn of astcanary down, until after the call to daemon(3).
This avoids possible conflicts with the internal implementation of
daemon(3).
(closes issue #15093)
Reported by: tzafrir
Patches:
20090513__issue15093__2.diff.txt uploaded by tilghman (license 14)
Tested by: tzafrir
........
Fix externalivr's setvariable command so that it properly sets multiple variables.
The command had a for loop that was guaranteed to only execute once since
the continuation operation of the loop would set the input buffer NULL. I rewrote
the loop so that its operation was more obvious, and it would set multiple variables
correctly.
I also reduced stack space required for the function, constified the input string,
and modified the function so that it would not modify the input string while I was
at it.
IAX was not sending REGREJ to terminate invalid registrations. Instead it sent another REGAUTH if the authentication challenge failed. This caused a loop of REGREQ and REGAUTH frames.
There is a bug tracker issue where people are reporting "Ghost" channels in their 'iax2 show channels' output. The confusion is caused by channels being listed as "(NONE)" with format "unknown". These are not channels of coarse. They are usually just pending registration or poke requests, but it is confusing output. To help make sense of this I have added two columns to 'iax2 show channels'. One shows the first message which started the transaction, and the second shows the last message sent by either side of the call. This helps diagnose why the entry exists and why it may not go away.
Fix a race condition where a reinvite could trigger a 482 response.
The loop detection/spiral detection code in chan_sip used the owner
channel's state as a criterion for determining if the incoming INVITE
is a looped request. The problem with this is that the INVITE-handling
code happens in a different thread than the thread that marks the owner
channel as being up. As a result, if a reinvite were to come in very quickly,
say from another Asterisk on the same LAN, it was possible for the reinvite
to arrive before the owner channel had been set to the up state.
This patch corrects the problem by using the invitestate of the sip_pvt
instead, since that can be guaranteed to be set correctly by the time
the reinvite arrives. Since there is a switch statement further in the
INVITE-handling code, the AST_STATE_RINGING state also checks the invitestate
of the sip_pvt in case we should actually be treating the channel as if it were
up already.
Remove an extraneous unlocking operation from ast_channel_free.
In the case that we could not remove the desired channel from the
list of channels, there was an extra call to unlock the channel list.
Since we unlock the list later on in the function anyway, this results
in the list being unlocked twice yet only being locked once.
(closes issue #15098)
Reported by: tim_ringenbach
Patches:
remove_extra_unlock.diff uploaded by tim (license 540)
........
................
Fix logic for how to proceed with a single digit extension.
(closes issue #15091)
Reported by: andrew
Patches:
20090512__issue15091.diff.txt uploaded by tilghman (license 14)
Tested by: andrew
........
................
This change modifies app_queue to properly generate CDR records in failure
situations.
This involves setting a proper cdr disposition coresponding to the given
failure condition and ensuring the proper information is stored in the cdr
record.
Update spiral support in trunk and 1.6.X to match what is in 1.4.
In 1.4, a SIP spiral is treated the same way as a call forward. This
works much better than what is currently in trunk and 1.6.X. The code
in trunk and 1.6.X did not create a new call to the recipient of the spiral,
instead trying to continue the same call. In addition to just being plain
wrong, this also had the side effect of only being able to spiral calls
to other SIP channels.
With this in place, as long as call forwards are honored, SIP spirals
will work properly. This means that it will work for outbound calls
made by the Queue, Dial, and Page applications. For originated calls and
spool calls, however, the spiral will not work properly until a generic
call forward mechanism is introduced into Asterisk.
Set the invitestate to INV_CANCELLED only if we are actually sending a SIP CANCEL.
The problem was that the hangup code was setting the invitestate too early. The result of
this was that we would always send a CANCEL request, even if it was not an appropriate
time to do so (e.g. we have not yet received a provisional response for our INVITE).
Note that this same fix had been applied to trunk and the 1.6.X branches starting with
revision 155467. This is why you will see this revision being blocked from those places.
Move 300 bytes around on the stack, to make more room for an extension buffer.
This allows more concurrent extensions to be copied for a single voicemail,
without creating a possibility of upsetting existing users, where a dialplan
could run out of stack space where it had run fine before. Alternatively,
we could have allocated off the heap, but that is a larger change and would
have increased the chance for instability introduced by this change.
This is really solved starting in 1.6.0.11, as the use of an ast_str buffer
allows an unlimited number of extensions (up to available memory). We
additionally create a new warning message when the buffer length is exceeded,
permitting administrators to see an issue after the fact, whereas previously
the list was silently truncated.
(closes issue #14739)
Reported by: p_lindheimer
Patches:
20090417__bug14739.diff.txt uploaded by tilghman (license 14)
Tested by: p_lindheimer
........
................
Sent wrong message to clear a call we started if the other end has not responed yet.
In the state MISDN_CALLING (i.e. SETUP was sent but no answer has arrived yet),
it is not allowed to clear the call with RELEASE_COMPLETE. It must be
cleared with DISCONNECT. A RELEASE_COMPLETE is only allowed as an answer
to a SETUP. (See Q.931 ch. 5.3.2, 5.3.2.a, 5.3.2.b)
Patches:
chan-misdn-ccstate7.patch uploaded by customer.
find_peer() does not find a valid peer when using pvt->recv as the sockaddr_in argument. Because of the way TCP works, the port number in pvt->recv is not what we're looking for at all. There is currently only one place that find_peer searches for a peer using the sockaddr_in argument. If the peer is not found after using pvt->recv (works for UDP since the port number will be correct), a temp sockaddr_in struct is made using the Contact header in the sip_request. This has the correct port number in it.
Make absolute paths for logger channels work properly
(Note: This is not a new feature, it was previously undocumented and broken.)
The Asterisk logger has a feature to support absolute pathnames for logger channels, but the code implementing the feature was broken. This has been fixed, and the absolute path feature is now documented in the sample logger.conf.
........
................
Fix Background within a Macro for FreePBX.
If the single digit DTMF is an extension in the specified context, then
go there and signal no DTMF. Otherwise, we should exit with that DTMF.
If we're in Macro, we'll exit and seek that DTMF as the beginning of an
extension in the Macro's calling context. If we're not in Macro, then
we'll simply seek that extension in the calling context. Previously,
someone complained about the behavior as it related to the interior of a
Gosub routine, and the fix (#14011) inadvertently broke FreePBX
(#14940). This change should fix both of these situations, but with the
possible incompatibility that if a single digit extension does not exist
(but a longer extension COULD have matched), it would have previously
gone immediately to the "i" extension, but will now need to wait for a
timeout.
(closes issue #14940)
Reported by: p_lindheimer
Patches:
20090420__bug14940.diff.txt uploaded by tilghman (license 14)
Tested by: p_lindheimer
........
................
Give a more helpful message when an incoming call's dialed extension does not match.
Added the dialed extension and context to the chan_misdn messages warning
that the dialed number cannot be matched in the dialplan.
........
................
Second result should not contain data from the first result.
(closes issue #15039)
Reported by: jims
Patches:
20090506__issue15039.diff.txt uploaded by tilghman (license 14)
Tested by: jims
........
Eliminate repetition of fullcontact during reconstruction.
If the fullcontact field appears in both the sippeers and the
sipregs table, then during reconstruction of the field, it will
otherwise be doubled.
(closes issue #14754)
Reported by: Alexei Gradinari
Patches:
20090506__bug14754.diff.txt uploaded by tilghman (license 14)
Tested by: lmadsen
........
................
Make ParkedCall application stop execution of the dialplan after hang up
Just changed park_exec to always return non-zero. I really wasn't entirely sure
at first if this was a bug. Decided it was since it would be surprising when
not using ParkedCall in the dialplan to hang up and have dialplan execution
continue.
Fix a bug where a timer would be created but not acknowledged.
This scenario crept up if chan_iax2 was loaded with no configuration file present.
It would create a timer and tell it to go at an interval but the thread that normally
acknowledges it would not be created because no configuration file was present. The timer
will now be closed if no configuration file is present.
Fix an incorrect assumption that certain values on the channel will always exist when they may not.
The CDR code involved with bridges wrongly assumed that the currently executing application and data
values will always exist. It is possible for this to be false when call forwarding is involved.
Fix a bug with setting t38pt_udptl at the user or peer level.
If an incoming call authenticated as a user or peer and t38pt_udptl was
not set to yes in general then no UDPTL session would be present and any
T38 related things would fail. This commit changes it so that if after
authenticating T38 is enabled but no UDPTL session is present one will be
created.
Correct some flaws in the memory accounting code for stringfields and ao2 objects
Under some conditions, the memory allocation for stringfields and ao2 objects would not have supplied valid file/function names for MALLOC_DEBUG tracking, so this commit corrects that.
........
Properly account for memory allocated for channels and datastores
As in previous commits, when channels are allocated (with ast_channel_alloc) or datastores are allocated (with ast_datastore_alloc) properly account for the memory being owned by the caller, instead of the allocator function itself.
........
Ensure that string pools allocated to hold stringfields are properly accounted in MALLOC_DEBUG mode
This commit modifies the stringfield pool allocator to remember the 'owner' of the stringfield manager the pool is being allocated for, and ensures that pools allocated in the future when fields are populated are owned by that file/function.
........
Ensure that astobj2 memory allocations are properly accounted for when MALLOC_DEBUG is used
This commit ensures that all astobj2 allocated objects are properly accounted for in MALLOC_DEBUG mode by passing down the file/function/line information from the module/function that actually called the astobj2 allocation function.
........
Ensure that by default only one console channel driver is loaded
This configuration file was changed to ensure that only one console channel driver
(chan_oss) is loaded by default, but the change would only work if chan_console
was not built. Now it will work as expected; if chan_alsa or chan_console are built
and installed, they will not be loaded unless explicity requested.
........
Fix an error in queue_log file rotation optimization code
This code was copy-and-pasted without properly changing references to event_rotate into queue_rotate, so under some conditions the log rotation would rotate queue_log even though it was not necessary.
........
Move static buffers to outside for loops in app_chanspy.
Similar to seanbright's commit 191422, this moves some static buffers
to be defined outside of for loops since it is undefined if memory
will be re-used or if the stack will grow with each iteration of the
loop.
........
r191629 | mmichelson | 2009-05-02 05:45:24 -0500 (Sat, 02 May 2009) | 3 lines
Kevin has informed me that thi sort of thing is not necessary.
........
................
Fix DTMF not being sent to other side after a partial feature match
This fixes a regression from commit 176701. The issue was that
ast_generic_bridge never exited after the feature digit timeout had elapsed,
which prevented the queued DTMF from being sent to the other side.
This issue was reported to me directly.
........
................
Sean Bright [Fri, 1 May 2009 16:26:32 +0000 (16:26 +0000)]
Fix a crash in app_queue with very long member lists.
A user reported via #asterisk that with very long lists of members, a crash
occurs in ast_strdupa, so just use a single buffer and ast_copy_string instead
of stack allocating copys of each interface name.
Change working directory to / under certain conditions.
If backgrounding and no core will be produced, then changing the directory
won't break anything; likewise, if the CWD isn't accessible by the current
user, then a core wasn't possible anyway.
(closes issue #14831)
Reported by: chris-mac
Patches:
20090428__bug14831.diff.txt uploaded by tilghman (license 14)
20090430__bug14831.diff.txt uploaded by tilghman (license 14)
Tested by: chris-mac
........
Build XML documention from *only* the source files that have docs in them
Change the build process so that doc/core-en_US.xml is dependent solely on the source files that have documentation in them, not on all source files.
........
We never, ever want these files to processed automatically, because we store the output files in Subversion and users should never need to rebuild them.
........
Fix 'inconsistent line endings' when autoconf 2.63 is used
Attempt to make configure script regeneration 'safe' using autoconf 2.63, which embeds a bare CR into the script, thus making Subversion complain about inconsistent line endings
This commit changes the MIME type of the configure script to be 'binary' thus making Subversion no longer inspect line endings, and as a bonus 'svn diff' will no longer try to generate diff output for it, which is not generally useful anyway.
........
................
Resolve a crash in res_smdi when used with chan_dahdi.
When chan_dahdi goes to get an SMDI message, it provides no search criteria.
It just grabs the next message that arrives. This code was written with the
SMDI dialplan functions in mind, since that is now the preferred method of
using SMDI. However, this broke support of it being used from chan_dahdi.
Allow for a position to be specified when entering a queue.
This would allow for one to add a caller to a specific place in the
queue instead of just placing the caller in the back every time. To help
facilitate some interesting manipulations, a new channel variable called
QUEUEPOSITION has been added. When a caller is removed from a queue, his
position in that queue is stored in the QUEUEPOSITION variable. One such
strategy an administrator can employ is to allow for the removal of a caller
from one queue followed by the insertion of the same caller into a separate
queue in the same position.
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.
........
................
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.
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
........
projects / thirdparty / asterisk.git / log
