Check for errors from fseek() when loading config file, properly abort on errors from fread(), and supply a traceback for errors generated when loading the config file.
Also, prepend a newline to traceback output so that the main error message is on it's own line.
........
Richard Mudgett [Fri, 4 Mar 2011 15:22:04 +0000 (15:22 +0000)]
Get real channel of a DAHDI call.
Starting with Asterisk v1.8, the DAHDI channel name format was changed for
ISDN calls to: DAHDI/i<span>/<number>[:<subaddress>]-<sequence-number>
There were several reasons that the channel name had to change.
1) Call completion requires a device state for ISDN phones. The generic
device state uses the channel name.
2) Calls do not necessarily have B channels. Calls placed on hold by an
ISDN phone do not have B channels.
3) The B channel a call initially requests may not be the B channel the
call ultimately uses. Changes to the internal implementation of the
Asterisk master channel list caused deadlock problems for chan_dahdi if it
needed to change the channel name. Chan_dahdi no longer changes the
channel name.
4) DTMF attended transfers now work with ISDN phones because the channel
name is "dialable" like the chan_sip channel names.
For various reasons, some people need to know which B channel a DAHDI call
is using.
* Added CHANNEL(dahdi_span), CHANNEL(dahdi_channel), and
CHANNEL(dahdi_type) so the dialplan can determine the B channel currently
in use by the channel. Use CHANNEL(no_media_path) to determine if the
channel even has a B channel.
* Added AMI event DAHDIChannel to associate a DAHDI channel with an
Asterisk channel so AMI applications can passively determine the B channel
currently in use. Calls with "no-media" as the DAHDIChannel do not have
an associated B channel. No-media calls are either on hold or
call-waiting.
Fix usage of "hasvoicemail=yes" and "mailbox=" in users.conf for SIP.
Since it's a duplicate, nothing is going to be done, so delme doesn't need to
be set at all. Strangely, when this was added, this was being set to 1 in 1.6,
and 0 in trunk.
A later version of flex already includes the fwrite workaround code, which if used twice causes a compilation error.
Detect whether Flex will compile without the workaround; if so, suppress our workaround code.
........
r309034 | tilghman | 2011-02-28 05:07:52 -0600 (Mon, 28 Feb 2011) | 2 lines
Don't broadcast FullyBooted to every AMI connection
The FullyBooted event should not be sent to every AMI connection every
time someone connects via AMI. It should only be sent to the user who
just connected.
Use remotesecret to authenticate with a remote party
The remotesecret option was only being used for outbound registration
and not for placing calls. This patch uses remotesecret on outbound
calls if it is set, otherwise secret is still used.
Properly check the bounds of arrays when decoding UDPTL packets. Also, remove broken support for receiving UDPTL packets larger than 16k. That shouldn't ever happen anyway.
Fix regression that changed behavior of queues when ringing a queue member.
This reverts r298596, which was to fix a highly bizarre and contrived issue
with a queue member that called into his own queue being transferred back
into his own queue. I couldn't reproduce that issue in any way. I think one
of the other recent transfer fixes actually fixed this.
Richard Mudgett [Tue, 15 Feb 2011 16:13:55 +0000 (16:13 +0000)]
No response sent for SIP CC subscribe/resubscribe request.
Asterisk does not send a response if we try to subscribe for call
completion after we have received a 180 Ringing. You can only subscribe
for call completion when the call has been cleared.
When we receive the 180 Ringing, for this call, its call-completion state
is 'CC_AVAILABLE'. If we then send a subscribe message to Asterisk, it
trys to change the call-completion state to 'CC_CALLER_REQUESTED'.
Because this is an invalid state change, it just ignores the message. The
only state Asterisk will accept our subscribe message is in the
'CC_CALLER_OFFERED' state.
Asterisk will go into the 'CC_CALLER_OFFERED' when the SIP client clears
the call by sending a CANCEL.
Asterisk should always send a response. Even if its a negative one.
The fix is to allow for the CCSS core to notify a CC agent that a failure
has occurred when CC is requested. The "ack" callback is replaced with a
"respond" callback. The "respond" callback has a parameter indicating
either a successful response or a specific type of failure that may need
to be communicated to the requester.
Tilghman Lesher [Mon, 14 Feb 2011 06:50:23 +0000 (06:50 +0000)]
Calling a gosub routine defined in AEL from Dial/Queue ceased to work.
A bug in AEL did not distinguish between the "s" extension generated by
AEL and an "s" extension that was required to exist by the chan_dahdi
(or another channel) that was not supplied with a starting extension.
Therefore, AEL made incorrect assumptions about what commands were
permissable in the context. This was fixed by making AEL generate a
different extension name. However, Dial and Queue make additional
assumptions about the name of the default gosub extension. Therefore,
they needed to be brought into line with a "macro" rendered by AEL (as
a gosub), without breaking traditional dialplans written without the
aid of AEL.
Alexandr Anikin [Thu, 10 Feb 2011 18:50:50 +0000 (18:50 +0000)]
Corrections for properly work with H.323v2 (older) endpoints and other
small fixes.
Interpret remote side H.225 version.
Corrections for H.323v2 endpoints:
don't start TCS and MSD before connect,
don't start TCS and MSD by accepting H.245 connection,
start TCS and MSD by StartH245 facility message.
Other fixes:
fix non zeroended remoteDisplayName issue, small fixes in call clearing
by closing H.245 connection, tcp keepalive introduced on TCP
connections (now is hardcoded, will be configurable in the future),
don't force H.245tunneling if FastStart is active, don't send Alerting
singal more than once per call.
Jason Parker [Tue, 8 Feb 2011 21:24:01 +0000 (21:24 +0000)]
Fix issue with verbose messages not showing on remote console.
This code was reworked recently, and since the logchannel list hadn't been
created yet at this point, and it was a verbose message, it was being dropped
on the floor. Now it'll continue on to where it should be handled.
Don't try to pickup a call in the middle of a masquerade
If A calls B which doesn't answer and C & D both try to do a call pickup, it is
possible for ast_pickup_call to answer the call, then fail to masquerade one of
the calls because the other one is already in the process of masquerading. This
patch checks to see if the channel is in the process of masquerading before
call before selecting it for a pickup.
Don't allow a REFER w/replaces to replace its own dialog
Asterisk currently accepts a REFER with a Refer-To with an embedded Replaces
header that matches the dialog of the REFER. This would be a situation like A
calls B, A calls C, A transfers B to A, which is just silly. This patch makes
the transfer fail instead of making Asterisk freak out and forget to hang other
channels up.
Mark Michelson [Mon, 7 Feb 2011 17:36:56 +0000 (17:36 +0000)]
Rearrange a bit of code in the generic CC recall operation.
By waiting to call the callback macro after the CC_INTERFACES,
extension, priority, and context have been set, this information
can be accessed more easily within the callback macro.
Richard Mudgett [Fri, 4 Feb 2011 18:53:06 +0000 (18:53 +0000)]
Don't send redirecting updates to the caller if the dialplan forked the call.
Each fork in the dial could be redirected and confuse the caller. For
ISDN the DivLeg1 and DivLeg3 messages would get confused because ISDN
redirects calls in sequence not in parallel.
* Also fixed a formatting inconsistency in app_dial.c and make a warning
message more useful about what frame type could not be written.
Jeff Peeler [Thu, 3 Feb 2011 23:49:28 +0000 (23:49 +0000)]
Fix SIP deadlock involving state changes.
Once again a call to pbx_builtin_getvar_helper (and pbx_builtin_setvar_helper)
has caused locking problems. Both of these functions lock the channel when
the channel argument is passed in!
In this case, the suspected problem (the backtrace makes it impossible to tell)
was the private being locked in sip_set_rtp_peer and then:
transmit_reinvite_with_sdp
try_suggested_sip_codec
pbx_builtin_getvar_helper
(Traced to verify that the fix was only required in 1.8 and later.)
When a call involves a local channel (like SIP -> Local -> SIP), the hangup
cause was not being set. This resulted in SIP channels sometimes getting a
503 error instead of a 486 when the far side sent a busy. In Asterisk 1.8+
this also can cause issues with CCSS that involve a local channel. This patch
sets the hangupcause for one side of the local channel to the other in
local_hangup for outbound calls.
........
................
Set exception on channel in parking thread when POLLPRI event detected.
This is done just to make the code be equivalent to the old select code. As
noted in 303106 the same issue was already fixed in this branch, but the
exception was not set on the channel in the case of POLLPRI. The reason that
this did not cause a problem here is because in 122923 the check in __ast_read
to check the exception flag was removed.
* Include the null terminator in the buffer length. When the frame is
queued it is copied. If the null terminator is not part of the frame
buffer length, the receiver could see garbage appended onto it.
* Add channel lock protection with ast_sendtext().
* Fixed AMI SendText action ast_sendtext() return value check.
........
................
Andrew Latham [Tue, 1 Feb 2011 18:02:06 +0000 (18:02 +0000)]
doc/tex dir removed, but corresponding entries still exists
Update README, CHANGES, and Makefile. Direct users to
http://wiki.asterisk.org for documentation or to the
AST.txt and AST.pdf included in the tarball.
(closes issue #18443)
Reported by: bas
Patches:
changes.diff uploaded by lathama (license 1028)
readme.diff uploaded by lathama (license 1028)
Tested by: lathama bas
Prevent a crash when dialing a technology with no destination (ex: Dial(SIP/))
chan_iax2 and other channel drivers already had code to prevent this. The
attempt that app_dial was making to prevent it was not correct, so I fixed that.
Tilghman Lesher [Mon, 31 Jan 2011 06:41:36 +0000 (06:41 +0000)]
Change mutex tracking so that it only consumes memory in the core mutex object when it's actually being used.
This reduces the overall size of a mutex which was 3016 bytes before this back
down to 216 bytes (this is on 64-bit Linux with a glibc-implemented mutex).
The exactness of the numbers here may vary slightly based upon how mutexes are
implemented on a platform, but the long and short of it is that prior to this
commit, chan_iax2 held down 98MB of memory on a 64-bit system for nothing more
than a table of 32767 locks. After this commit, the same table occupies a mere
7MB of memory.
Andrew Latham [Sun, 30 Jan 2011 00:11:56 +0000 (00:11 +0000)]
Add Function and Application Relationships to documentation
Add and extend the see-also sections to the documentation for applications
and functions in an effort to expand the online documentation of the wiki.
Also check for and update any links to moved documentation in the doc folder.
When we pass the S() or L() options to MeetMe, make sure that we honor C as well.
Without this patch, if the user was kicked from the conference via the S() or L()
mechanism, we would just hang up on them even if we also passed C (continue in
dialplan when kicked). With this patch we honor the C flag in those cases.
Make sure that we unref the correct object when ejecting the most recent caller.
Currently, when we kick the last user to enter, we decrement our own reference
count which results in a crash when we kick another user or when we exit the
conference ourselves.
This will fix #18225 in 1.8 and trunk, but that particular bug does not exist in
1.6.2.
Don't leak references if we can't create a pseudo channel for mixing in MeetMe.
If there was a problem allocating a pseudo channel when building our meetme, we
weren't destroying our user container or destroying the mutexes that we created.
........
r304682 | seanbright | 2011-01-28 17:38:05 -0500 (Fri, 28 Jan 2011) | 2 lines
Revert part of the previous commit that snuck in.
........
Fix default prefix=/usr regression on non-Linux systems.
This partially reverts a change made in branches/1.4/ r267759, which will
cause issue #17013 to be reopened. This issue was pointed out by a user
on #asterisk, who helpfully discovered that paths were being set incorrectly.
To truly understand what was wrong, one should run:
svn diff --force -c<this revision> configure
........
................
This patch modifies chan_sip to route responses to the address the request came from. It also modifies chan_sip to respect the maddr parameter in the Via header.
Per the man page, setvbuf() must be called before any other operation on an open file.
We use setvbuf() to associate a buffer with a stream, but we have already written
to the open file. This works (by chance) on Linux, but fails on other platforms,
such as OpenSolaris.
DTMF attended transfers sometimes fail for no apparent reason.
The loop in feature_request_and_dial() can exit when Party C has answered
without processing an AST_CONTROL_ANSWER. Also sometimes an
AST_CONTROL_ANSWER never happens even though Party C has answered.
Don't hangup Party C if he is up or we receive an AST_CONTROL_ANSWER.
........
................
In the case of an attended transfer (A calls B, A atxfers to C) where
A becomes unreachable before replying to Asterisk's BYE, Asterisk can
sometimes retransmit the BYE indefinitely. This is because
__sip_autodestruct tests p->refer && !ast_test_flag(&p->flags[0],
SIP_ALREADYGONE and will then transmit a BYE. When this BYE times out,
it will not ever be marked as ALREADYGONE, so when __sip_autodestruct
is called again, we end up starting the cycle over.
This patch adds a call to sip_alreadygone(pkt->owner) in retrans_pkt
in the case of a BYE that has timed out. This should prevent Asterisk
from trying to transmit new BYE messages in the future.
Sending out unnecessary PROCEEDING messages breaks overlap dialing.
Issue #16789 was a good idea. Unfortunately, it breaks overlap dialing
through Asterisk. There is not enough information available at this point
to know if dialing is complete. The ast_exists_extension(),
ast_matchmore_extension(), and ast_canmatch_extension() calls are not
adequate to detect a dial through extension pattern of "_9!".
Workaround is to use the dialplan Proceeding() application early in
non-dial through extensions.
* Effectively revert issue #16789.
* Allow outgoing overlap dialing to hear dialtone and other early media.
A PROGRESS "inband-information is now available" message is now sent after
the SETUP_ACKNOWLEDGE message for non-digital calls. An
AST_CONTROL_PROGRESS is now generated for incoming SETUP_ACKNOWLEDGE
messages for non-digital calls.
* Handling of the AST_CONTROL_CONGESTION in chan_dahdi/sig_pri was
inconsistent with the cause codes.
* Added better protection from sending out of sequence messages by
combining several flags into a single enum value representing call
progress level.
(closes issue #18509)
Reported by: wimpy
Patches:
issue18509_early_media_v1.8_v3.patch uploaded by rmudgett (license 664)
Expanded upon issue18509_early_media_v1.8_v3.patch to include analog
and SS7 because of backporting requirements.
Tested by: wimpy, rmudgett
........
................
A previous change was made to account for when the number of voicemail messages
exceeds the max limit to be handled properly, but it caused gaps in the messages
to not be properly handled. This has now been resolved.
In later non 1.4 branches, it appears that resequencing wasn't even occurring
due from what appears and accidental code removal.
projects / thirdparty / asterisk.git / log
