Since adding the AST_CONTROL_SRCUPDATE frame type,
there are places where ast_rtp_new_source may be called
where the tech_pvt of a channel may not yet have an
rtp structure allocated. This caused a crash in chan_skinny,
which was fixed earlier, but now the same crash has been
reported against chan_h323 as well. It seems that the best
solution is to modify ast_rtp_new_source to not attempt to
set the marker bit if the rtp structure passed in is NULL.
This change to ast_rtp_new_source also allows the removal
of what is now a redundant pointer check from chan_skinny.
The purpose of this branch was to take into account
"burps" which could cause jitterbuffers to misbehave.
One such example is if the L option to Dial() were used
to inject audio into a bridged conversation at regular
intervals. Since the audio here was not passed through
the jitterbuffer, it would cause a gap in the jitterbuffer's
timestamps which would cause a frames to be dropped for a
brief period.
Now ast_generic_bridge will empty and reset the jitterbuffer
each time it is called. This causes injected audio to be handled
properly.
ast_generic_bridge also will empty and reset the jitterbuffer
if it receives an AST_CONTROL_SRCUPDATE frame since the change
in audio source could negatively affect the jitterbuffer.
All of this was made possible by adding a new public API call
to the abstract_jb called ast_jb_empty_and_reset.
I discovered that also, in the previous bug fixes and changes,
the cdr.conf 'unanswered' option is not being obeyed, so
I fixed this.
And, yes, there are two 'answer' times involved in this
scenario, and I would agree with you, that the first
answer time is the time that should appear in the CDR.
(the second 'answer' time is the time that the bridge
was begun).
I made the necessary adjustments, recording the first
answer time into the peer cdr, and then using that to
override the bridge cdr's value.
To get the 'unanswered' CDRs to appear, I purposely
output them, using the dial cmd to mark them as
DIALED (with a new flag), and outputting them if
they bear that flag, and you are in the right mode.
I also corrected one small mention of the Zap device
to equally consider the dahdi device.
I heavily tested 10-sec-wait macros in dial, and
without the macro call; I tested hangups while the
macro was running vs. letting the macro complete
and the bridge form. Looks OK. Removed all the
instrumentation and debug.
In a conversion to use ast_strlen_zero, the meaning of the flag IAX_HASCALLERID
was perverted. This change reverts IAX2 to the original meaning, which was,
that the callerid set on the client should be overridden on the server, even if
that means the resulting callerid is blank. In other words, if you set
"callerid=" in the IAX config, then the callerid should be overridden to blank,
even if set on the client. Note that there's a distinction, even on realtime,
between the field not existing (NULL in databases) and the field existing, but
set to blank (override callerid to blank).
make datastore creation and destruction a generic API since it is not really channel related, and add the ability to add/find/remove datastores to manager sessions
Russell Bryant [Mon, 4 Aug 2008 15:02:12 +0000 (15:02 +0000)]
Be explicit that we don't want a result from this callback. The callback would
never indicate a match, so nothing would have been returned anyway, but it was
still a poor example of proper usage.
The changes I introduce here seem to clear up the problem
for me. However, if they do not for you, please reopen this
bug, and we'll keep digging.
The root of this problem seems to be a subtle memory corruption
introduced when creating an extension with an empty extension
name. While valgrind cannot detect it outside of DEBUG_MALLOC
mode, when compiled with DEBUG_MALLOC, this is certain death.
The code in main/features.c is a puzzle to me. On the initial
module load, the code is attempting to add the parking extension
before the features.conf file has even been opened!
I just wrapped the offending call with an if() that will not
try to add the extension if the extension name is empty. THis
seems to solve the corruption, and let the "memory show allocations"
work as one would expect.
But, really, adding an extension with an empty name is a seriously
bad thing to allow, as it will mess up all the pattern matching
algorithms, etc. So, I added a statement to the add_extension2 code to return
a -1 if this is attempted.
Sean Bright [Fri, 1 Aug 2008 19:29:26 +0000 (19:29 +0000)]
Remove some code that used to do something but does not anymore, mainly
to get rid of a shadow warning (but this seemed legitimate enough to fix
here instead of in my branch).
Russell Bryant [Fri, 1 Aug 2008 18:16:24 +0000 (18:16 +0000)]
Merge changes from team/bbryant/keyrotation
This set of changes enhances IAX2 encryption support by adding key rotation
to provide enhanced security. The key used for encryption is rotated right
after the call gets set up, and then again every few minutes. This was
discussed at the last AstriDevCon. For interoperability with older versions
of Asterisk, there is an option that disables key rotation.
Tilghman Lesher [Fri, 1 Aug 2008 16:39:51 +0000 (16:39 +0000)]
SIP should use the transport type set in the Moved Temporarily for the next
invite.
(closes issue #11843)
Reported by: pestermann
Patches:
20080723__issue11843_302_ignores_transport_16branch.diff uploaded by bbryant (license 36)
20080723__issue11843_302_ignores_transport_trunk.diff uploaded by bbryant (license 36)
Tested by: pabelanger
Mark Michelson [Fri, 1 Aug 2008 14:29:48 +0000 (14:29 +0000)]
IMAP storage functioned under the assumption that folders
such as "Work" and "Family" would be subfolders of the
INBOX. This is an invalid assumption to make, but it could
be desirable to set up folders in this manner, so a new
option for voicemail.conf, "imapparentfolder" has been
added to allow for this.
Specify codecs in callfiles and manager, to allow video calls to be set up
from callfiles and AMI.
(closes issue #9531)
Reported by: Geisj
Patches:
20080715__bug9531__1.4.diff.txt uploaded by Corydon76 (license 14)
20080715__bug9531__1.6.0.diff.txt uploaded by Corydon76 (license 14)
Tested by: Corydon76
OK, a few days of debugging, a bunch of instrumentation
in chan_sip, main/channel.c, main/pbx.c, etc. and 5 solid
notebook pages of notes later, I have made the small
tweek necc. to get the start time right on the second
CDR when:
A Calls B
B answ.
A hits Xfer button on sip phone,
A dials C and hits the OK button,
A hangs up
C answers ringing phone
B and C converse
B and/or C hangs up
But does not harm the scenario where:
A Calls B
B answ.
B hits xfer button on sip phone,
B dials C and hits the OK button,
B hangs up
C answers ringing phone
A and C converse
A and/or C hangs up
The difference in start times on the second CDR is because
of a Masquerade on the B channel when the xfer number is
sent. It ends up replacing the CDR on the B channel with
a duplicate, which ends up getting tossed out. We keep
a pointer to the first CDR, and update *that* after the
bridge closes. But, only if the CDR has changed.
I hope this change is specific enough not to muck
up any current CDR-based apps. In my defence, I
assert that the previous information was wrong,
and this change fixes it, and possibly other
similar scenarios.
I wonder if I should be doing the same thing
for the channel, as I did for the peer, but
I can't think of a scenario this might affect.
I leave it, then, as an exersize for the users,
to find the scenario where the chan's CDR
changes and loses the proper start time.
........
and as to 1.4 to trunk; have I expressed my
feelings about code shifting from one file
to another? Good.
Two errors:
1) If a function returns SQLITE_LOCKED, no recovery is possible.
2) An error message can be allocated, even when no error is signalled.
(closes issue #13109)
Reported by: gknispel_proformatique
Optimize frame cache by realloc'ing the smallest frame when the cache is full.
This ensures that we don't just keep a cache of tiny frames, continually doing
an alloc/free for each data frame, thus negating the point of having a cache.
Add more timeout checks into app_queue, specifically
targeting areas where an unknown and potentially
long time has just elapsed. Also added a check
to try_calling() to return early if the timeout
has elapsed instead of potentially setting a negative
timeout for the call (thus making it have *no* timeout
at all).
Qwell pointed out, via IRC, that the previous fix only worked when explicitly
set. When nothing is set, and the option is implied, it breaks, because
configure sets the prefix to 'NONE'. Fixing.
launch_netscript sometimes returns -1, which fails to set AGISTATUS. Map
failure to -1, so that AGISTATUS is always set.
(closes issue #13199)
Reported by: smw1218
Move implementation of an attended-transfer-complete sound from one channel
driver into a common place for multiple channel drivers.
(closes issue #13152)
Reported by: caio1982
Patches:
atxfer_complete_sound3.diff uploaded by caio1982 (license 22)
Add %u and %g to the ASTERISK_PROMPT settings, for username and group,
respectively. Also, take the opportunity to clean up the CLI prompt
generation code.
(closes issue #13175)
Reported by: eliel
Patches:
cliprompt.patch uploaded by eliel (license 64)
Minor changes to reduce packaging changes made by the Fedora maintainer.
(closes issue #12974)
Reported by: jcollie
Patches:
0001-Don-t-override-duplicate-optimization-flags.patch uploaded by jcollie (license 412)
Detect when sox fails to raise the volume, because sox can't read the file.
(closes issue #12939)
Reported by: rickbradley
Patches:
20080728__bug12939.diff.txt uploaded by Corydon76 (license 14)
Tested by: rickbradley
Mark Michelson [Mon, 28 Jul 2008 19:53:56 +0000 (19:53 +0000)]
This commit compensates for buggy poll(2)
implementations. Asterisk has, for a long time,
had its own implementation of poll(2) which
just used the input arguments to call select(2).
In 1.4, this internal implementation was used
for Darwin systems. This was removed in Asterisk
trunk at some point, but it seems as though this
was not the right move to make.
On Mac OS X, it appears as though the poll used
to gather CLI input does not respond properly
when connecting via a remote Asterisk console.
Reverting to the use of Asterisk's poll fixed
the issue.
Also, there is now an option for the configure
script, --enable-internal-poll, which will allow
for anyone to use Asterisk's internal poll
implementation in case they suspect that their
system's poll implementation is buggy.
Change SendImage() to output a more consistent status variable.
(closes issue #13134)
Reported by: eliel
Patches:
app_image.c.patch uploaded by eliel (license 64)
UPGRADE.patch uploaded by eliel (license 64)
Russell Bryant [Sat, 26 Jul 2008 15:15:14 +0000 (15:15 +0000)]
ast_device_state() gets called in two different ways. The first way is when
called from elsewhere in Asterisk to find the current state of a device. In
that case, we want to use the cached value if it exists. The other way is when
processing a device state change. In that case, we do not want to check the
cache because returning the last known state is counter productive.
projects / thirdparty / asterisk.git / log
