The `ast_geoloc_datastore_add_eprofile` function does not return 0 on
success, it returns the size of the underlying datastore. This means
that the datastore will be freed and its pointer set to NULL when no
error occured at all.
George Joseph [Mon, 10 Oct 2022 14:35:54 +0000 (08:35 -0600)]
pjsip_transport_events: Fix possible use after free on transport
It was possible for a module that registered for transport monitor
events to pass in a pjsip_transport that had already been freed.
This caused pjsip_transport_events to crash when looking up the
monitor for the transport. The fix is a two pronged approach.
1. We now increment the reference count on pjsip_transports when we
create monitors for them, then decrement the count when the
transport is going to be destroyed.
2. There are now APIs to register and unregister monitor callbacks
by "transport key" which is a string concatenation of the remote ip
address and port. This way the module needing to monitor the
transport doesn't have to hold on to the transport object itself to
unregister. It just has to save the transport_key.
* Added the pjsip_transport reference increment and decrement.
* Changed the internal transport monitor container key from the
transport->obj_name (which may not be unique anyway) to the
transport_key.
* Added a helper macro AST_SIP_MAKE_REMOTE_IPADDR_PORT_STR() that
fills a buffer with the transport_key using a passed-in
pjsip_transport.
* Added the following functions:
ast_sip_transport_monitor_register_key
ast_sip_transport_monitor_register_replace_key
ast_sip_transport_monitor_unregister_key
and marked their non-key counterparts as deprecated.
* Updated res_pjsip_pubsub and res_pjsip_outbound_register to use
the new "key" monitor functions.
NOTE: res_pjsip_registrar also uses the transport monitor
functionality but doesn't have a persistent object other than
contact to store a transport key. At this time, it continues to
use the non-key monitor functions.
Naveen Albert [Thu, 27 Oct 2022 11:32:12 +0000 (11:32 +0000)]
tcptls: Prevent crash when freeing OpenSSL errors.
write_openssl_error_to_log has been erroneously
using ast_free instead of free, which will
cause a crash when MALLOC_DEBUG is enabled since
the memory was not allocated by Asterisk's memory
manager. This changes it to use the actual free
function directly to avoid this.
Mike Bradeen [Thu, 13 Oct 2022 16:19:17 +0000 (10:19 -0600)]
res_pjsip: prevent crash on websocket disconnect
When a websocket (or potentially any stateful connection) is quickly
created then destroyed, it is possible that the qualify thread will
destroy the transaction before the initialzing thread is finished
with it.
Depending on the timing, this can cause an assertion within pjsip.
To prevent this, ast_send_stateful_response will now create the group
lock and add a reference to it before creating the transaction.
While this should resolve the crash, there is still the potential that
the contact will not be cleaned up properly, see:ASTERISK~29286. As a
result, the contact has to 'time out' before it will be removed.
res_pjsip_outbound_registration: Allow to use multiple proxies for registration
Current registration code use pjsip_parse_uri to verify outbound_proxy
that is different from the reading this option for the endpoint. This
made value with multiple proxies invalid for registration pjsip settings.
Removing URI validation helps to use registration through multiple proxies.
res_pjsip: return all codecs on a re-INVITE without SDP
Currently chan_pjsip on receiving a re-INVITE without SDP will only
return the codecs that are previously negotiated and not offering
all enabled codecs.
This causes interoperability issues with different equipment (e.g.
from Cisco) for some of our customers and probably also in other
scenarios involving 3PCC infrastructure.
According to RFC 3261, section 14.2 we SHOULD return all codecs
on a re-INVITE without SDP
The PR proposes a new parameter to configure this behaviour:
all_codecs_on_empty_reinvite. It includes the code, documentation,
alembic migrations, CHANGES file and example configuration additions.
Expands the pjsip logger to support the ability to filter
by SIP message method. This can make certain types of SIP debugging
easier by only logging messages of particular method(s).
ASTERISK-30146 #close
Co-authored-by: Sean Bright <sean@seanbright.com>
Change-Id: I9c8cbb6fc8686ef21190eb42e08bc9a9b147707f
Dialing API: Cancel a running async thread, may not cancel all calls
race condition: ast_dial_join() may not cancel outgoing call, if
function is called just after called party answer and before
application execution (bit is_running_app not yet set).
This fix adds ast_softhangup() calls in addition to existing
pthread_kill() when is_running_app is not set.
This fixes dahdi_request to properly set the cause
code to CONGESTION instead of BUSY if no channels
were actually available.
Currently, the cause is erroneously set to busy
if the channel itself is found, regardless of its
current state. However, if the channel is not available
(e.g. T1 down, card not operable, etc.), then the
channel itself may not be in a functional state,
in which case CHANUNAVAIL is the correct cause to use.
This adds a simple check to ensure that busy tone
is only returned if a channel is encountered that
has an owner, since that is the only possible way
that a channel could actually be busy.
Naveen Albert [Sun, 16 Oct 2022 20:35:29 +0000 (20:35 +0000)]
res_pjsip_pubsub: Prevent removing subscriptions.
pjproject does not provide any mechanism of removing
event packages, which means that once a subscription
handler is registered, it is effectively permanent.
pjproject will assert if the same event package is
ever registered again, so currently unloading and
loading any Asterisk modules that use subscriptions
will cause a crash that is beyond our control.
For that reason, we now prevent users from being
able to unload these modules, to prevent them
from ever being loaded twice.
Naveen Albert [Wed, 28 Sep 2022 12:38:37 +0000 (12:38 +0000)]
say: Don't prepend ampersand erroneously.
Some logic in say.c for determining if we need
to also add an ampersand for file seperation was faulty,
as non-successful files would increment the count, causing
a leading ampersand to be added improperly.
This is fixed, and a unit test that captures this regression
is also added.
Mike Bradeen [Thu, 29 Sep 2022 20:55:20 +0000 (14:55 -0600)]
audiohook: add directional awareness
Add enum to allow setting optional direction. If set to only one
direction, only feed matching-direction frames to the associated
slin factory.
This prevents mangling the transcoder on non-mixed frames when the
READ and WRITE frames would have otherwise required it. Also
removes the need to mute or discard the un-wanted frames as they
are no longer added in the first place.
res_stasis_snoop is changed to use this addition to set direction
on audiohook based on spy direction.
If no direction is set, the ast_audiohook_init will init this enum
to BOTH which maintains existing functionality.
Naveen Albert [Wed, 1 Jun 2022 16:06:14 +0000 (16:06 +0000)]
cdr: Allow bridging and dial state changes to be ignored.
Allows bridging, parking, and dial messages to be globally
ignored for all CDRs such that only a single CDR record
is generated per channel.
This is useful when CDRs should endure for the lifetime of
an entire channel and bridging and dial updates in the
dialplan should not result in multiple CDR records being
created for the call. With the ignore bridging option,
bridging changes have no impact on the channel's CDRs.
With the ignore dial state option, multiple Dials and their
outcomes have no impact on the channel's CDRs. The
last disposition on the channel is preserved in the CDR,
so the actual disposition of the call remains available.
These two options can reduce the amount of "CDR hacks" that
have hitherto been necessary to ensure that CDR was not
"spoiled" by these messages if that was undesired, such as
putting a dummy optimization-disabled local channel between
the caller and the actual call and putting the CDR on the channel
in the middle to ensure that CDR would persist for the entire
call and properly record start, answer, and end times.
Enabling these options is desirable when calls correspond
to the entire lifetime of channels and the CDR should
reflect that.
"fname" is passed in as a const char *, but strstr() mangles that
into a char *, and we were attempting to modify the string in place.
This is an unwanted (and undocumented) side-effect.
This patch adds support for mediasec SIP headers and SDP attributes.
These are defined in RFC 3329, 3GPP TS 24.229 and
draft-dawes-sipcore-mediasec-parameter. The new features are
implemented so that a backbone for RFC 3329 is present to streamline
future work on RFC 3329.
With this patch, Asterisk can communicate with Deutsche Telekom trunks
which require these fields.
Avoid crashing by skipping invisible bridges and checking the
snapshot for a null pointer. In effect this is how the bridges
are enumerated in res/ari/resource_bridges.c already.
Naveen Albert [Mon, 19 Sep 2022 17:35:40 +0000 (17:35 +0000)]
res_pjsip_geolocation: Change some notices to debugs.
If geolocation is not in use for an endpoint, the NOTICE
log level is currently spammed with messages about this,
even though nothing is wrong and these messages provide
no real value. These log messages are therefore changed
to debugs.
Naveen Albert [Wed, 21 Sep 2022 23:17:13 +0000 (23:17 +0000)]
func_logic: Don't emit warning if both IF branches are empty.
The IF function currently emits warnings if both IF branches
are empty. However, there is no actual necessity that either
branch be non-empty as, unlike other conditional applications/
functions, nothing is inherently done with IF, and both
sides could legitimately be empty. The warning is thus turned
into a debug message.
Naveen Albert [Sun, 11 Sep 2022 22:13:02 +0000 (22:13 +0000)]
features: Add no answer option to Bridge.
Adds the n "no answer" option to the Bridge application
so that answer supervision can not automatically
be provided when Bridge is executed.
Additionally, a mechanism (dialplan variable)
is added to prevent bridge targets (typically the
target of a masquerade) from answering the channel
when they enter the bridge.
This patch adds a new option to the 100rel parameter for pjsip
endpoints called "peer_supported". When an endpoint with this option
receives an incoming request and the request indicated support for the
100rel extension, then Asterisk will send 1xx responses reliably. If
the request did not indicate 100rel support, Asterisk sends 1xx
responses normally.
manager: be more aggressive about purging http sessions.
If we find that n_max (currently hard wired to 1) sessions were purged,
schedule the next purge for 1ms into the future rather than 5000ms (as
per current). This way we will purge up to 1000 sessions per second
rather than 1 every 5 seconds.
This mitigates a build-up of sessions should http sessions gets
established faster than 1 per 5 seconds.
Ben Ford [Tue, 2 Aug 2022 17:15:36 +0000 (12:15 -0500)]
res_pjsip: Add TEL URI support for basic calls.
This change allows TEL URI requests to come through for basic calls. The
allowed requests are INVITE, ACK, BYE, and CANCEL. The From and To
headers will now allow TEL URIs, as well as the request URI.
Support is only for TEL URIs present in traffic from a remote party.
Asterisk does not generate any TEL URIs on its own.
encrypting a block of data with RSA
decrypting a block of data with RSA
signing a block of data with RSA
verifying a signature with RSA
encrypting a block of data with AES-ECB
encrypting a block of data with AES-ECB
Naveen Albert [Thu, 8 Sep 2022 14:12:34 +0000 (14:12 +0000)]
func_frame_trace: Remove bogus assertion.
The FRAME_TRACE function currently asserts if it sees
a MASQUERADE_NOTIFY. However, this is a legitimate thing
that can happen so asserting is inappropriate, as there
are no clear negative ramifications of such a thing. This
is adjusted to be like the other frames to print out
the subclass.
Naveen Albert [Wed, 27 Jul 2022 19:54:43 +0000 (19:54 +0000)]
lock.c: Add AMI event for deadlocks.
Adds an AMI event to indicate that a deadlock
has likely started, when Asterisk is compiled
with DETECT_DEADLOCKS enabled. This can make
it easier to perform automated deadlock detection
and take appropriate action (such as doing a core
dump). Unlike the deadlock warnings, the AMI event
is emitted only once per deadlock.
George Joseph [Thu, 25 Aug 2022 13:00:33 +0000 (07:00 -0600)]
res_geolocation: Add two new options to GEOLOC_PROFILE
Added an 'a' option to the GEOLOC_PROFILE function to allow
variable lists like location_info_refinement to be appended
to instead of replacing the entire list.
Added an 'r' option to the GEOLOC_PROFILE function to resolve all
variables before a read operation and after a Set operation.
Added a few missing parameters to the ones allowed for writing
with GEOLOC_PROFILE.
Fixed a bug where calling GEOLOC_PROFILE to read a parameter
might actually update the profile object.
George Joseph [Thu, 18 Aug 2022 12:29:09 +0000 (06:29 -0600)]
res_geolocation: Allow location parameters on the profile object
You can now specify the location object's format, location_info,
method, location_source and confidence parameters directly on
a profile object for simple scenarios where the location
information isn't common with any other profiles. This is
mutually exclusive with setting location_reference on the
profile.
Updated appdocsxml.dtd to allow xi:include in a configObject
element. This makes it easier to link to complete configOptions
in another object. This is used to add the above fields to the
profile object without having to maintain the option descriptions
in two places.
Added profile parameter "suppress_empty_ca_elements" that
will cause Civic Address elements that are empty to be
suppressed from the outgoing PIDF-LO document.
Fixed a possible SEGV if a sub-parameter value didn't have a
value.
George Joseph [Tue, 16 Aug 2022 12:25:10 +0000 (06:25 -0600)]
res_geolocation: Add built-in profiles
The trigger to perform outgoing geolocation processing is the
presence of a geoloc_outgoing_call_profile on an endpoint. This
is intentional so as to not leak location information to
destinations that shouldn't receive it. In a totally dynamic
configuration scenario however, there may not be any profiles
defined in geolocation.conf. This makes it impossible to do
outgoing processing without defining a "dummy" profile in the
config file.
This commit adds 4 built-in profiles:
"<prefer_config>"
"<discard_config>"
"<prefer_incoming>"
"<discard_incoming>"
The profiles are empty except for having their precedence
set and can be set on an endpoint to allow processing without
entries in geolocation.conf. "<discard_config>" is actually the
best one to use in this situation.
Joshua C. Colp [Tue, 30 Aug 2022 13:01:29 +0000 (10:01 -0300)]
res_pjsip_sdp_rtp: Skip formats without SDP details.
When producing an outgoing SDP we iterate through the configured
formats and produce SDP information. It is possible for some
configured formats to not have SDP information available. If this
is the case we skip over them to allow the SDP to still be
produced.
Naveen Albert [Tue, 3 May 2022 12:53:28 +0000 (12:53 +0000)]
cli: Prevent assertions on startup from bad ao2 refs.
If "core show channels" is run before startup has completed, it
is possible for bad ao2 refs to occur because the system is not
yet fully initialized. This will lead to an assertion failing.
To prevent this, initialization of CLI builtins is moved to be
later along in the main load sequence. Core CLI commands are
loaded at the same time, but channel-related commands are loaded
later on.
Joshua C. Colp [Fri, 19 Aug 2022 13:24:33 +0000 (10:24 -0300)]
pjsip: Add TLS transport reload support for certificate and key.
This change adds support using the pjsip_tls_transport_restart
function for reloading the TLS certificate and key, if the filenames
remain unchanged. This is useful for Let's Encrypt and other
situations. Note that no restart of the transport will occur if
the certificate and key remain unchanged.
Naveen Albert [Thu, 25 Aug 2022 11:51:03 +0000 (11:51 +0000)]
res_tonedetect: Fix typos referring to wrong variables.
Fixes two typos that cause fax detection to not work.
One refers to the wrong frame variable, and the other
refers to the subclass.integer instead of the frametype
as it should.
Sean Bright [Fri, 19 Aug 2022 16:02:07 +0000 (12:02 -0400)]
chan_dahdi.c: Resolve a format-truncation build warning.
With gcc (Ubuntu 11.2.0-19ubuntu1) 11.2.0:
> chan_dahdi.c:4129:18: error: ‘%s’ directive output may be truncated
> writing up to 255 bytes into a region of size between 242 and 252
> [-Werror=format-truncation=]
This removes the error-prone sizeof(...) calculations in favor of just
doubling the size of the base buffer.
Naveen Albert [Sat, 5 Feb 2022 12:13:56 +0000 (12:13 +0000)]
features: Add transfer initiation options.
Adds additional control options over the transfer
feature functionality to give users more control
in how the transfer feature sounds and works.
First, the "transfer" sound that plays when a transfer is
initiated can now be customized by the user in
features.conf, just as with the other transfer sounds.
Secondly, the user can now specify the transfer extension
in advance by using the TRANSFER_EXTEN variable. If
a valid extension is contained in this variable, the call
will automatically be transferred to this destination.
Otherwise, it will fall back to collecting the extension
from the user as is always done now.
George Joseph [Fri, 5 Aug 2022 13:50:49 +0000 (07:50 -0600)]
res_geolocation: Address user issues, remove complexity, plug leaks
* Added processing for the 'confidence' element.
* Added documentation to some APIs.
* removed a lot of complex code related to the very-off-nominal
case of needing to process multiple location info sources.
* Create a new 'ast_geoloc_eprofile_to_pidf' API that just takes
one eprofile instead of a datastore of multiples.
* Plugged a huge leak in XML processing that arose from
insufficient documentation by the libxml/libxslt authors.
* Refactored stylesheets to be more efficient.
* Renamed 'profile_action' to 'profile_precedence' to better
reflect it's purpose.
* Added the config option for 'allow_routing_use' which
sets the value of the 'Geolocation-Routing' header.
* Removed the GeolocProfileCreate and GeolocProfileDelete
dialplan apps.
* Changed the GEOLOC_PROFILE dialplan function as follows:
* Removed the 'profile' argument.
* Automatically create a profile if it doesn't exist.
* Delete a profile if 'inheritable' is set to no.
* Fixed various bugs and leaks
* Updated Asterisk WiKi documentation.
Naveen Albert [Mon, 1 Aug 2022 00:30:00 +0000 (00:30 +0000)]
app_confbridge: Fix memory leak on updated menu options.
If the CONFBRIDGE function is used to dynamically set
menu options, a memory leak occurs when a menu option
that has been set is overridden, since the menu entry
is not destroyed before being freed. This ensures that
it is.
Additionally, logic that duplicates the destroy function
is removed in lieu of the destroy function itself.
Naveen Albert [Thu, 28 Jul 2022 11:10:35 +0000 (11:10 +0000)]
manager: Remove documentation for nonexistent action.
The manager XML documentation documents a "FilterList"
action, but there is no such action. Therefore, this can
lead to confusion when people try to use a documented
action that does not, in fact, exist. This is removed
as the action never did exist in the past, nor would it
be trivial to add since we only store the regex_t
objects, so the filter list can't actually be provided
without storing that separately. Most likely, the
documentation was originally added (around version 10)
in anticipation of something that never happened.
Naveen Albert [Fri, 22 Jul 2022 20:57:05 +0000 (20:57 +0000)]
general: Improve logging levels of some log messages.
Adjusts some logging levels to be more or less important,
that is more prominent when actual problems occur and less
prominent for less noteworthy things.
Naveen Albert [Wed, 27 Jul 2022 18:34:57 +0000 (18:34 +0000)]
cdr.conf: Remove obsolete app_mysql reference.
The CDR sample config still mentions that app_mysql
is available in the addons directory, but this is
incorrect as it was removed as of 19. This removes
that to avoid confusion.
Naveen Albert [Sat, 23 Jul 2022 22:17:28 +0000 (22:17 +0000)]
pbx_functions.c: Manually update ast_str strlen.
When ast_func_read2 is used to read a function using
its read function (as opposed to a native ast_str read2
function), the result is copied directly by the function
into the ast_str buffer. As a result, the ast_str length
remains initialized to 0, which is a bug because this is
not the real string length.
This can cascade and have issues elsewhere, such as when
reading substrings of functions that only register read
as opposed to read2 callbacks. In this case, since reading
ast_str_strlen returns 0, the returned substring is empty
as opposed to the actual substring. This has caused
the ast_str family of functions to behave inconsistently
and erroneously, in contrast to the pbx_variables substitution
functions which work correctly.
This fixes this issue by manually updating the ast_str length
when the result is copied directly into the ast_str buffer.
Additionally, an assertion and a unit test that previously
exposed these issues are added, now that the issue is fixed.
build: fix bininstall launchd issue on cross-platform build
configure script detects /sbin/launchd, but the result of this
check is not used in Makefile (bininstall). Makefile also detects
/sbin/launchd file to decide if it is required to install
safe_asterisk.
configure script correctly detects cross compile build and sets
PBX_LAUNCHD=0
In case of building asterisk on MacOS host for Linux target using
external toolchain (e.g. OpenWrt toolchain), bininstall does not
install safe_asterisk (due to /sbin/launchd detection in Makefile),
but it is required on target (Linux).
This patch adds HAVE_SBIN_LAUNCHD=@PBX_LAUNCHD@ to makeopts.in to
use the result of /sbin/launchd detection from configure script in
Makefile.
Also this patch uses HAVE_SBIN_LAUNCHD in Makefile (bininstall) to
decide if it is required to install safe_asterisk.
Naveen Albert [Mon, 11 Jul 2022 11:32:39 +0000 (11:32 +0000)]
db: Add AMI action to retrieve DB keys at prefix.
Adds the DBGetTree action, which can be used to
retrieve all of the DB keys beginning with a
particular prefix, similar to the capability
provided by the database show CLI command.
Naveen Albert [Tue, 12 Jul 2022 21:38:12 +0000 (21:38 +0000)]
manager: Fix incomplete filtering of AMI events.
The global event filtering code was only in one
possible execution path, so not all events were
being properly filtered out if requested. This moves
that into the universal AMI handling code so all
events are properly handled.
Additionally, the CLI listing of disabled events can
also get truncated, so we now print out everything.
res_pjsip: delay contact pruning on Asterisk start
Move the call to ast_sip_location_prune_boot_contacts() *after* the call
to ast_res_pjsip_init_options_handling() so that
res/res_pjsip/pjsip_options.c is informed about the contact deletion and
updates its sip_options_contact_statuses list. This allows for an AMI
event to be sent by res/res_pjsip/pjsip_options.c if the endpoint
registers again from the same remote address and port (i.e., same URI)
as used before the Asterisk restart.
ASTERISK-30109 Reported-by: Michael Neuhauser
Change-Id: I1ba4478019e4931a7085f62708d9b66837e901a8
Naveen Albert [Fri, 1 Apr 2022 20:17:15 +0000 (20:17 +0000)]
chan_dahdi: Add POLARITY function.
Adds a POLARITY function which can be used to
retrieve the current polarity of an FXS channel
as well as set the polarity of an FXS channel
to idle or reverse at any point during a call.
Naveen Albert [Tue, 29 Mar 2022 01:35:43 +0000 (01:35 +0000)]
chan_dahdi: Fix buggy and missing Caller ID parameters
There are several things wrong with analog Caller ID
handling that are fixed by this commit:
callerid.c's Caller ID generation function contains the
logic to use the presentation to properly send the proper
Caller ID. However, currently, DAHDI does not pass any
presentation information to the Caller ID module, which
means that presentation is completely ignored on all calls.
This means that lines could be getting Caller ID information
they aren't supposed to.
Part of the reason this has been obscured is because the
simple switch logic for handling the built in *67 and *82
is completely wrong. Rather than modifying the presentation
for the call accordingly (which is what it's supposed to do),
it simply blanks out the Caller ID or fills it in. This is
wrong, so wrong that it makes a mockery of the specification.
Additionally, it would leave to the "UNAVAILABLE" disposition
being used for Caller ID generation as opposed to the "PRIVATE"
disposition that it should have been using. This is now fixed
to only update the presentation and not modify the number and
name, so that the simple switch *67/*82 work correctly.
Next, sig_analog currently only copies over the name and number,
nothing else, when it is filling in a duplicated caller id
structure. Thus, we also now copy over the presentation
information so that is available for the Caller ID spill.
Additionally, this meant that "valid" was implicitly 0,
and as such presentation would always fail to "Unavailable".
The validity is therefore also copied over so it can be used
by ast_party_id_presentation.
As part of this fix, new API is added so that all the relevant
Caller ID information can be passed in to the Caller ID generation
functions. Parameters that are also completely missing from the
Caller ID spill have also been added, to enhance the compatibility,
correctness, and completeness of the Asterisk Caller ID implementation.
Mike Bradeen [Thu, 2 Jun 2022 02:03:06 +0000 (20:03 -0600)]
Makefile: Avoid git-make user conflict
make_version now silently checks if the required git commands will
fail. If they do, then return UNKNOWN__git_check_fail to
distinguish this failure from other UNKNOWN__ version failures
Makefile checks for this value on install and exits out with
instructions
Naveen Albert [Sat, 18 Jun 2022 12:17:50 +0000 (12:17 +0000)]
app_confbridge: Always set minimum video update interval.
Currently, if multiple video-enabled ConfBridges are
conferenced together, we immediately get into a scenario
where an infinite sequence of video updates fills up
the taskprocessor queue and causes memory consumption
to climb unabated until Asterisk is killed. This is due
to the core bridging mechanism that provides video updates
(softmix_bridge_write_control in bridge_softmix.c)
continously updating all the channels in the bridge with
video updates.
The logic to do so in the core is that the video updates
should be provided if the video_update_discard property
for the bridge is 0, or if enough time has elapsed since
the last video update. Thus, we already have a safeguard
built in to ensure the scenario described above does not
happen. Currently, however, this safeguard is not being
adequately ensured.
In app_confbridge, the video_update_discard property
defaults to 2000, which is a healthy value that should
completely prevent this issue. However, this value is
only set onto the bridge in the SFU video mode. This
leaves video modes such as follow_talker completely
vulnerable, since video_update_discard will actually
be 0, since the default or set value was never applied.
As a result, the core bridging mechanism will always
try to provide video updates regardless of when the last
one was sent.
To prevent this issue from happening, we now always
set the video_update_discard property on the bridge
with the value from the bridge profile. The app_confbridge
defaults will thus ensure that infinite video updates
no longer happen in any video mode.
Sean Bright [Tue, 5 Jul 2022 15:24:58 +0000 (11:24 -0400)]
pbx.c: Simplify ast_context memory management.
Allocate all of the ast_context's character data in the structure's
flexible array member and eliminate the clunky fake_context. This will
simplify future changes to ast_context.
projects / thirdparty / asterisk.git / log
