This patch fixes a regression of sorts that was introduced in
rev 24425.
It basically fixes AST-190/ABE-1782.
What was wrong: the user has 6000 extensions in one context; and
then 6000 contexts, one per extension. The parser could only handle
about 4893 of the 6000 extens in the single context.
This was due to the regression I mentioned. To get rid of
shift/reduce conflicts, Luigi set up right-recursive lists
for globals, context elements, switch lists, and statements.
Right recursive lists got rid of the warnings, but instead, they
use up a tremendous amount of stack space when the lists are long.
I saw this a few years back, and resolved not to fix it until
someone complained. That day has arrived!
After the changes were made, I ran the regression test suite,
and there were no problems.
I took the test case the user provided, and added 100,000
extensions to the single context, that already had 6,000 extens
in it. (I'll see your 6, and raise you 100!) It takes a few minutes
to read it all in, check it and generate code for it, but no
problems.
So, I think I can say that fundamentally, there are no longer
any limits on the number of items you can place in contexts,
statement blocks, switches, or globals, beyond your virt mem
constraints.
........
................
fix two very minor bugs: if anyone ever uses SLINEAR16 as a format in RTP, ensure that the samples are byte-swapped to network order if needed. also, when a smoother is operating on a format that has a sample rate other than 8000 samples per second, use the proper sample rate for computing delivery timestamps.
........
action_bridge() and bridge_exec() both search for the channels to bridge to, and then immediately drop the lock. Instead, they should hold the lock until the masquerade is complete. This will guarantee the channel remains and prevent any other weirdness from occurring. In action_bridge() some more weirdness comes into play. Both channels are needlessly locked at the same time and perform the exact same logic. It makes sense from a coding organizational standpoint, but could cause a theoretical deadlock so I split the code up. There is an issue associated with this, but since its a rather complicated thing to reproduce I'm not certain this alone will close it.
Modify h323 to build against PTLib as well as the older PWLib
Several changes in PTLib have occurred requiring build time detection. Changes
accounted for include the library name change, config option change, install
location change, and a boolean type change which is handled by ast_ptlib.h.
Also, the sed check has been modified to properly work with autoconf >= 2.62.
This patch fixes merge_contexts_and_delete so it does not deadlock when hints are present.
Reason: when I re-engineered the merge_and_delete func to
reduce its lock time, I failed to notice that the
functions it calls still also do locking as before.
This leads to deadlocks on dialplan reloads, when
there are actually living, subscribed hints registered
in the system.
While the reporter come across this problem while using
AEL, I might note that these deadlocks should also happen
if extensions.conf were used.
All of the above routines are static and restricted
to be used only within pbx.c, and more specifically
within the merge_contexts_and_delete routine.
They are pretty much the same as their counterparts
except they don't lock contexts or hints.
Most of them now do the real work of their
name-alike, with optional locking via extra arguments,
and are called by their name-alike. The goal was to
have the original functions so they would behave
exactly as before.
Both PJ and I tested these fixes, and the deadlocking
problem is no longer encountered.
Russell Bryant [Wed, 18 Feb 2009 06:05:09 +0000 (06:05 +0000)]
Merged revisions 176901 via svnmerge from
https://origsvn.digium.com/svn/asterisk/trunk
........
r176901 | russell | 2009-02-18 00:00:40 -0600 (Wed, 18 Feb 2009) | 9 lines
Fix a number of incorrect uses of strncpy().
The big problem here is that the 3rd argument provided in these uses of strncpy()
did not reserve a byte for the null terminator, leaving the potential for writing
one byte past the end of the buffer.
Aside from this, there were coding guidelines violations with regards to spacing,
as well as hard coded lengths being used instead of sizeof().
T38 faxdetect should jump to the 'fax' extension for incoming calls only
The previous implementation of T38 faxdetect resulted in both sides of the
call jumping to a fax extension when both sides had 't38pt_udptl=yes' and
'faxdetect=yes' in sip.conf and a 'fax' extension in the current context.
This revision will jump to a 'fax' extension on incoming calls only.
........
Several changes to codec_dahdi to play nice with G723.
This commit brings in the changes that were living out on the
svn/asterisk/team/sruffell/asterisk-trunk-transcoder branch. codec_dahdi.c now
always uses signed linear as the simple codec so that a soft g729 codec will
not end up being preferred to the hardware codec. There are also changes to
allow codec_dahdi.c to feed packets to the hardware in the native sample size of
the codec. This solves problems with choppy audio when using G723.
........
create a UDPTL structure in create_addr_from_peer() if it does not already exist for T38
This is required to create a UDPTL structure in create_addr_from_peer() to handle the
scenario where 't38pt_udptl=yes' is not defined in the [general] section of sip.conf but
is defined the peer's context. I tested this patch by enabling t38pt_udptl in the
[general] section on one system and only enabling t38pt_udptl in a peer's context on
the system sending a fax. Without the patch, the sending system will fail to initiate
T38 negotiation with the warning message, "No way to add SDP without an UDPTL structure".
When this patch is applied the sending side will successfully initiate T38 negotiation.
........
Modify bridging to properly evaluate DTMF after first warning is played
The main problem is currently if the Dial flag L is used with a warning sound,
DTMF is not evaluated after the first warning sound. To fix this, a flag has
been added in ast_generic_bridge for playing the warning which ensures that if
a scheduled warning is missed, multiple warrnings are not played back (due to a
feature evaluation or waiting for digits). ast_channel_bridge was modified to
store the nexteventts in the ast_bridge_config structure as that information
was lost every time ast_channel_bridge was reentered, causing a hangup due to
incorrect time calculations.
(closes issue #14315)
Reported by: tim_ringenbach
Reviewed on reviewboard:
http://reviewboard.digium.com/r/163/
........
................
Add assertions in the quest to track down a refcount leak.
(closes issue #14485)
Reported by: davevg
........
r176642 | tilghman | 2009-02-17 15:14:18 -0600 (Tue, 17 Feb 2009) | 8 lines
Prior to masquerade, move the group definitions to the channel performing the
masq, so that the group count lingers past the bridge.
(closes issue #14275)
Reported by: kowalma
Patches:
20090216__bug14275.diff.txt uploaded by Corydon76 (license 14)
Tested by: kowalma
........
Russell Bryant [Tue, 17 Feb 2009 21:05:17 +0000 (21:05 +0000)]
Blocked revisions 176639 via svnmerge
........
r176639 | russell | 2009-02-17 15:04:08 -0600 (Tue, 17 Feb 2009) | 9 lines
Significantly improve scheduler performance under high load.
This patch changes the scheduler to use a max-heap to store pending scheduler
entries instead of a fully sorted doubly linked list. When the number of
entries in the scheduler gets large, this will perform much better. For much
more detailed information on this change, see the review request.
Russell Bryant [Tue, 17 Feb 2009 20:42:04 +0000 (20:42 +0000)]
Blocked revisions 176627 via svnmerge
........
r176627 | russell | 2009-02-17 14:41:24 -0600 (Tue, 17 Feb 2009) | 37 lines
Merge a large set of updates to the Asterisk indications API.
This patch includes a number of changes to the indications API. The primary
motivation for this work was to improve stability. The object management
in this API was significantly flawed, and a number of trivial situations could
cause crashes.
The changes included are:
1) Remove the module res_indications. This included the critical functionality
that actually loaded the indications configuration. I have seen many people
have Asterisk problems because they accidentally did not have an
indications.conf present and loaded. Now, this code is in the core,
and Asterisk will fail to start without indications configuration.
There was one part of res_indications, the dialplan applications, which did
belong in a module, and have been moved to a new module, app_playtones.
2) Object management has been significantly changed. Tone zones are now
managed using astobj2, and it is no longer possible to crash Asterisk by
issuing a reload that destroys tone zones while they are in use.
3) The API documentation has been filled out.
4) The API has been updated to follow our naming conventions.
5) Various bits of code throughout the tree have been updated to account
for the API update.
6) Configuration parsing has been mostly re-written.
7) "Code cleanup"
The code is from svn/asterisk/team/russell/indications/.
Russell Bryant [Tue, 17 Feb 2009 17:38:47 +0000 (17:38 +0000)]
Merged revisions 176557 via svnmerge from
https://origsvn.digium.com/svn/asterisk/trunk
........
r176557 | russell | 2009-02-17 11:33:38 -0600 (Tue, 17 Feb 2009) | 12 lines
Fix a race condition that caused device states to become incorrect for hints.
The problem here is that the hint processing code was subscribed to the wrong
event type. So, it started processing state for a hint too soon, before the
device state cache had been updated.
Also, fix a similar bug in app_queue, as it was also subscribed to the wrong
event type.
After a 'sip reload', qualifies for realtime peers weren't immediately
restarted, instead waiting until the next registration. We're now
caching the qualify across a reload/restart and starting the qualify
immediately upon loading the peer.
(closes issue #14196)
Reported by: pdf
Patches:
20090120__bug14196_1.4.diff.txt uploaded by pdf (license 663)
Tested by: pdf
........
................
Fixes issue with AST_CONTROL_SRCUPDATE not being relayed correctly during bridging
This should have been committed with rev176247, but I missed it. srcupdate frames no longer break out of the native bridge, but are not being sent to the other call leg either. This fixs that.
Use the correct list macros for deleting an item from the middle of a list.
(issue #13777)
Reported by: pj
Patches:
20090203__bug13777.diff.txt uploaded by Corydon76 (license 14)
Tested by: pj
........
fix a flaw in the ast_string_field_build() family of API calls; these functions made no attempt to reuse the space already allocated to a field, so every time the field was written it would allocate new space, leading to what appeared to be a memory leak.
........
r176254 | kpfleming | 2009-02-16 15:41:46 -0600 (Mon, 16 Feb 2009) | 3 lines
correct a logic error in the last stringfields commit... don't mark additional space as allocated if the string was built using already-allocated space
........
................
Open the DAHDI pseudo device and set it to be nonblocking atomically
Apparently on FreeBSD, attempting to set the O_NONBLOCKING flag separately
from opening the file was causing an "inappropriate ioctl for device" error.
While I cannot fathom why this would be happening, I certainly am not opposed
to making the code a bit more compact/efficient if it also fixes a bug.
Turns key rotation back on by default. Added bit into encryption IE to indicate whether or not key rotation is supported or not. If it is not supported then it is not enabled, which insures backwards compatibility. This eliminates the need for the keyrotate option in iax.conf, so it has been removed.
........
................
Assist proper thread synchronization when stopping the logger thread.
I was finding that on my dev box, occasionally attempting to "stop now" in
trunk would cause Asterisk to hang. I traced this to the fact that the logger
thread was waiting on a condition which had already been signalled. The logger
thread also need to be sure to check the value of the close_logger_thread variable.
The close_logger_thread variable is only checked when the list of logmessages is empty.
This allows for the logger thread to print and free any pending messages before exiting.
........
Don't have the Via header stored as a stringfield as it can change often during the lifetime of a dialog.
This issue crept up with subscriptions on the AA50. When an outgoing NOTIFY is sent a new branch value
is created and the Via header is changed to reflect it. Since this was a stringfield a new spot in the
pool was used for the value while the old was left untouched/unused. If the current pool was full a new
pool was created. This would cause memory usage to increase steadily.
From a user point-of-view, this adds new CLI commands and Manager Actions to
better facilitate the reloading of queues and the resetting of their statistics.
The new CLI commands are the "queue reload" and "queue reset stats" commands.
The new manager actions are the QueueReload and QueueReset commands.
Turns key rotation back on by default. Added bit into encryption IE to indicate whether or not key rotation is supported or not. If it is not supported then it is not enabled, which insures backwards compatibility. This eliminates the need for the keyrotate option in iax.conf, so it has been removed.
........
Fix a potential crash situation when using IMAP voicemail
If calling into VoiceMailMain when using IMAP storage, it was
possible to crash Asterisk by hanging up the phone when prompted
for a voicemail mailbox. This patch fixes the issue.
While it may appear that this patch is superficial, it allows code
execution to continue to the failure case just below the IMAP_STORAGE
code block where this patch has been applied
Add basic (passthrough, playback, record) support for ITU G.722.1 and G.722.1C (also known as Siren7 and Siren14)
This patch adds passthrough, file recording and file playback support for the codecs listed above, with negotiation over SIP/SDP supported. Due to Asterisk's current limitation of treating a codec/bitrate combination as a unique codec, only G.722.1 at 32 kbps and G.722.1C at 48 kbps are supported.
Along the way, some related work was done:
1) The rtpPayloadType structure definition, used as a return result for an API call in rtp.h, was moved from rtp.c to rtp.h so that the API call was actually usable. The only previous used of the API all was chan_h323.c, which had a duplicate of the structure definition instead of doing it the right way.
2) The hardcoded SDP sample rates for various codecs in chan_sip.c were removed, in favor of storing these sample rates in rtp.c along with the codec definitions there. A new API call was added to allow retrieval of the sample rate for a given codec.
3) Some basic 'a=fmtp' parsing for SDP was added to chan_sip, because chan_sip *must* decline any media streams offered for these codecs that are not at the bitrates that we support (otherwise Bad Things (TM) would result).
Add dynamic fax buffer configuration option to chan_dahdi.conf
When the 'faxdetect' configuration option is used, one may also want to use
the 'faxbuffers' configuration option in chan_dahdi.conf. This option will
dynamically use the configured 'faxbuffers' buffer policy on a channel for
the life of the call following the detection of fax tones. The faxbuffers
buffer policy will be reverted during call teardown.
An example use of 'faxbuffers' is below. This example would switch to using
6 buffers with a full buffer policy.
This patch adds forceencryption=yes as an iax.conf option. When force encryption is enabled, no unencrypted connections are allowed. This insures all connections are encrypted. This is a new feature, so CHANGES and iax.conf.sample are updated as well.
Fix crashes when receiving certain T.38 packets. Also, increase the maximum
size of T.38 packets and warn users when they try to set the limits above those
maximums.
(closes issue #13050)
Reported by: schern
Patches:
20090212__bug13050.diff.txt uploaded by Corydon76 (license 14)
Tested by: schern
........
................
Fix ParkedCall event information for From field in the case of a blind transfer
If the parker information can not be obtained from the peer, try and see if
the BLINDTRANSFER channel variable has been set. Previously, a blind transfer
to the ParkAndAnnounce app would return nothing for the From.
Fix crash in event of failed attempt to transfer to parking
The peer may not necessarily exist, such as in the case of a transfer to
ParkAndAnnounce. In this case don't try to play a sound to it.
........
................
Key rotation breaks compatibility between (trunk/1.6.1) and (1.2/1.4/1.6.0). As a follow up to this, I am investigating possible ways to allow key rotation to be on by default and not affect the other branches, but for now it must be turned off.
........
Russell Bryant [Thu, 12 Feb 2009 17:08:25 +0000 (17:08 +0000)]
Merged revisions 175125 via svnmerge from
https://origsvn.digium.com/svn/asterisk/trunk
................
r175125 | russell | 2009-02-12 10:57:25 -0600 (Thu, 12 Feb 2009) | 35 lines
Merged revisions 175124 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4
........
r175124 | russell | 2009-02-12 10:51:13 -0600 (Thu, 12 Feb 2009) | 27 lines
Don't send DTMF for infinite time if we do not receive an END event.
I thought that this was going to end up being a pretty gnarly fix, but it turns
out that there was actually already a configuration option in rtp.conf,
dtmftimeout, that was intended to handle this situation. However, in between
Asterisk 1.2 and Asterisk 1.4, the code that processed the option got lost.
So, this commit brings it back to life.
The default timeout is 3 seconds. However, it is worth noting that having
this be configurable at all is not really the recommended behavior in RFC 2833.
From Section 3.5 of RFC 2833:
Limiting the time period of extending the tone is necessary
to avoid that a tone "gets stuck". Regardless of the
algorithm used, the tone SHOULD NOT be extended by more than
three packet interarrival times. A slight extension of tone
durations and shortening of pauses is generally harmless.
Three seconds will pretty much _always_ be far more than three packet
interarrival times. However, that behavior is not required, so I'm going to
leave it with our legacy behavior for now.
Make lock information for ao2_trylock be more useful and gnarly
Core show locks information involving an ao2_trylock did not
show the function that called ao2_trylock, but would instead
show ao2_trylock as the source of the lock. This is not useful
when trying to debug locking issues.
One bizarre note is that this logic is already in 1.4 but somehow
did not get merged to trunk or the 1.6.X branches.
........
Fix odd "thank you" sound playing behavior in app_queue.c
If someone has configured the queue to play an position or holdtime
announcement, then it is odd and potentially unexpected to hear a
"Thank you for your patience" sound when no position or holdtime
was actually announced.
This fixes the announcement so that the "thanks" sound is only played
in the case that a position or holdtime was actually announced.
There is a way that the "thank you" sound can be played without a
position or holdtime, and that is to set announce-frequency to a value
but keep announce-position and announce-holdtime both turned off.
Fix 'd' option for app_dial and add new option to Answer application
The 'd' option would not work for channel types which use RTP to transport
DTMF digits. The only way to allow for this to work was to answer the channel
if we saw that this option was enabled.
I realized that this may cause issues with CDRs, specifically with giving false
dispositions and answer times. I therefore modified ast_answer to take another
parameter which would tell if the CDR should be marked answered.
I also extended this to the Answer application so that the channel may be answered
but not CDRified if desired.
I also modified app_dictate and app_waitforsilence to only answer the channel if it
is not already up, to help not allow for faulty CDR answer times.
All of these changes are going into Asterisk trunk. For 1.6.0 and 1.6.1, however, all
the changes except for the change to the Answer application will go in since we do
not introduce new features into stable branches
Tell the device state core a change happened when a channel is freed but not a specific state.
We need to do this because while we know that the freeing of the channel may cause something to become
not in use we do not know this for sure. There may be another channel that is still up which would cause
it to be in use.
(closes issue #13238)
Reported by: kowalma
Patches:
20090121__bug13238.diff.txt uploaded by Corydon76 (license 14)
Tested by: alecdavis
........
Fix potential for stack overflows in app_chanspy.c
When using the 'g' or 'e' options, the stack allocations that
were used could cause a stack overflow if a spyer stayed on the
line long enough without actually successfully spying on anyone.
The problem has been corrected by using static buffers and copying
the contents of the appropriate strings into them instead of using
functions like alloca or ast_strdupa
Fix an fd leak that would occur in HTTP AMI sessions
The explanation behind this fix is a bit complicated, and I've already
typed it up in the code as a huge comment inside of manager.c, so I'll
give the abridged version here.
We needed a way to separate action-specific data from session-specific data.
Unfortunately, the only way to maintain API compatibility and to not have to
change every single manager action was to rename the current mansession structure
and wrap it inside a new mansession structure which actually contains action-
specific data.
Improve behavior of jitterbuffer when maxjitterbuffer is set.
This change improves the way the jitterbuffer handles maxjitterbuffer and
dramatically reduces the number of frames dropped when maxjitterbuffer is
exceeded. In the previous jitterbuffer, when maxjitterbuffer was exceeded, all
new frames were dropped until the jitterbuffer is empty. This change modifies
the code to only drop frames until maxjitterbuffer is no longer exceeded.
Also, previously when maxjitterbuffer was exceeded, dropped frames were not
tracked causing stats for dropped frames to be incorrect, this change also
addresses that problem.
Make the logic for inuse and inringing manipluation match that of 1.4. The old broken logic would reset the values back to 0 during certain scenarios causing the wrong state to be reported.
(closes issue #14399)
Reported by: caspy
(issue #13238)
Reported by: kowalma
........
Fixes issue with hangups not being sent and external process never terminating.
The ignore_hangup, run_dead, and noanswer flags were never initilized to zero causing hangups to never be issued. If the external script expects to be notified of a hangup and never receives one, it runs indefinitely.
Don't overwrite our pointer to the music class when music on hold stops. We will use this if it starts again to see if we can resume the music where it left off.
(closes issue #14407)
Reported by: mostyn
........
................
check ast_strlen_zero() before calling ast_strdupa() in sip_uri_headers_cmp()
and sip_uri_params_cmp()
The reporter didn't actually upload a properly-formed patch, instead a
modified chan_sip.c file was uploaded. I created a patch to determine the
changes, then modified the suggested changes to create a proper fix. The
summary above is a complete description of the changes.
This is very similar to the DAHDI immediate=yes option. When the phone is picked up, instead of giving a dialtone it connects directly to the "s" extension. Changes where implemented in chan_iax2.c to directly connect to the "s" extension in the appropriate context when this option is enabled. Examples explaining its use are added to iax2.conf.sample. CHANGES has been updated as well.
Don't subscribe to a mailbox on pseudo channels. It is futile. This solves an issue where duplicated pseudo channels would cause a crash because the first one would unsubscribe and the next one would also try to unsubscribe the same subscription.
(closes issue #14322)
Reported by: amessina
........
Some clients do not put the call-id for replaces at the beginning, so support it being anywhere in the string.
(closes issue #14350)
Reported by: fhackenberger
........
r173968 | file | 2009-02-06 13:15:01 -0400 (Fri, 06 Feb 2009) | 2 lines
Remove a debug message I put in by accident.
........
................
Always detach and destroy the whisper and barge audiohooks. Additionally also allow an audiohook to be detached if it has not been attached.
(closes issue #14414)
Reported by: bluecrow76
........
Russell Bryant [Fri, 6 Feb 2009 10:57:30 +0000 (10:57 +0000)]
Blocked revisions 173858 via svnmerge
........
r173858 | russell | 2009-02-06 04:55:35 -0600 (Fri, 06 Feb 2009) | 13 lines
Add a common implementation of a scheduler context with a dedicated thread.
This commit expands the Asterisk scheduler API to include a common implementation
of a scheduler context being processed by a dedicated thread. chan_iax2 has been
updated to use this new code. Also, as a result, this resolves some race
conditions related to the previous chan_iax2 scheduler handling.
Related to rev 171452 which resolved the same issues in 1.4.
projects / thirdparty / asterisk.git / log
