http: response body often missing after specific request
This patch works around a problem with the HTTP body
being dropped from the response to a specific client
and under specific circumstances:
a) Client request comes from node.js user agent
"Shred" via use of swagger-client library.
b) Asterisk and Client are *not* on the same
host or TCP/IP stack
In testing this problem, it has been determined that
the write of the HTTP body is lost, even if the data
is written using low level write function. The only
solution found is to instruct the TCP stack with the
shutdown function to flush the last write and finish
the transmission. See review for more details.
Matthew Jordan [Fri, 28 Mar 2014 14:19:20 +0000 (14:19 +0000)]
contrib/realtime: Remove empty SQL script files
Since the relatime scripts are now managed by Alembic, the previous realtime
scripts were previously removed. However, the removal process messed up, as
the files were still in the repository. The contents were just empty.
This removes the files from the tree.
........
Merged revisions 411442 from http://svn.asterisk.org/svn/asterisk/branches/12
Matthew Jordan [Fri, 28 Mar 2014 03:55:26 +0000 (03:55 +0000)]
chan_sip: Add MESSAGE request to allowed methods
The allowed methods advertised by chan_sip did not previously note the MESSAGE
request. Even in Asterisk 1.8, we do accept in-dialog MESSAGE requests; we
should advertise that we support MESSAGE requests.
ASTERISK-23504 #close
ASTERISK-23504 #comment Reported by: Martin Kontsek
ASTERISK-23504 #comment Patch sip.h_patch.diff uploaded by Martin Kontsek (license 6587)
Mark Michelson [Thu, 27 Mar 2014 14:21:15 +0000 (14:21 +0000)]
Give sorcery instances a reference to their wizards.
On graceful shutdown, sorcery wizards are all killed off, but it is
possible for sorcery instances to still have dangling pointers after
this, possibly causing a crash. Giving the sorcery instances a reference
to their wizards ensures that the wizard reference will remain valid for
the lifetime of the sorcery instance.
Joshua Colp [Wed, 26 Mar 2014 22:45:10 +0000 (22:45 +0000)]
say: Fix a bug where SayNumber in Polish tries to play incorrect sound.
This change fixes a bug where calling SayNumber with a number divisible by
100 using the Polish language would cause the code to attempt to play a
sound file with an empty name.
Jonathan Rose [Wed, 26 Mar 2014 16:15:12 +0000 (16:15 +0000)]
chan_sip: Send real CallerID information with P-Assserted-Identity (RFC-3325)
Prior too this patch, the P-Asserted-Identity header would include anonymous
caller id information which seems to go against the point of the
P-Asserted-Identity header. Now the real caller ID information will be
included in this header. Also, no privacy header would be included.
This patch adds 'Privacy: id' to outgoing SIP messages that include the
P-Asserted-Identity header.
(closes issue AST-1301)
........
Merged revisions 411189 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 411190 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 411193 from http://svn.asterisk.org/svn/asterisk/branches/12
Sean Bright [Tue, 25 Mar 2014 18:44:57 +0000 (18:44 +0000)]
ARI: Don't complain about missing ARI users when we aren't enabled
Currently, if ARI is not enabled it will still complain that there are no
configured users. This patch checks to see if ARI is enabled before logging and
error or iterating the container to validate the users.
Richard Mudgett [Tue, 25 Mar 2014 16:57:41 +0000 (16:57 +0000)]
res_pjsip: Fix contact authenticate_qualify endpoint lookup when qualifing a contact.
* Fixed bad use of ao2_find() in on_endpoint().
* Replaced use of find_endpoints() with find_an_endpoint() since only the
first found endpoint is ever needed.
* Fixed qualify_contact_cb() to update the contact with the aor
authenticate_qualify setting. Otherwise, permanent contacts in the aor
type sections would have a config line order dependancy.
* Fixed off nominal path contact ref leak in qualify_contact(). The
comment saying the unref is not needed was wrong.
* Fixed off nominal path use of the endpoint parameter if it is NULL in
send_out_of_dialog_request().
* Added missing off nominal path unref of pjsip tdata in
send_out_of_dialog_request().
* Fixed off nominal path failing to call the callback in send_request_cb()
when the request is challenged for authentication.
* Eliminated silly RAII_VAR() use in qualify_contact_cb().
* Updated ast_sip_send_request() doxygen to better reflect reality.
Kinsey Moore [Tue, 25 Mar 2014 16:06:57 +0000 (16:06 +0000)]
chan_sip: Fix incorrect use of timers
If update_provisional_keepalive() is called while
send_provisional_keepalive_full() is waiting on the PVT lock, then
pvt->provisional_keepalive_sched_id will be changed to a new sched_id
value by update_provisional_keepalive(), but that new sched_id then may
be overwritten with -1 by send_provisional_keepalive_full(), killing
the pvt's reference to a schedule and "leaking" the reference.
(closes issue ASTERISK-22079)
Review: https://reviewboard.asterisk.org/r/3368/
Reported by: Jamuel Starkey, Matteo, Leif Madsen, Steve Davies
Patches:
provisional_keepalive_fix.diff uploaded by Steve Davies (license 5012)
........
Merged revisions 411088 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 411089 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 411091 from http://svn.asterisk.org/svn/asterisk/branches/12
Jonathan Rose [Tue, 25 Mar 2014 15:56:05 +0000 (15:56 +0000)]
ARI: Resolve a subscription leak against implicit bridge subscriptions
When a channel in a stasis application is joined to a bridge, a subscription
for that bridge is created implicitly for the stasis application serving the
channel. Prior to this patch, subsequent removals of the channel from the
bridge would leave the subscription open.
Richard Mudgett [Thu, 20 Mar 2014 16:35:57 +0000 (16:35 +0000)]
assigned-uniqueids: Miscellaneous cleanup and fixes.
* Fix memory leak in ast_unreal_new_channels(). Made it generate the ;2
uniqueid on a stack variable instead of mallocing it.
* Made send error response to ARI and AMI requests instead of just logging
excessive uniqueid length and allowing truncation. action_originate() and
ari_channels_handle_originate_with_id().
* Fixed minor truncating uniqueid hole when generating the ;2 uniqueid
string length. Created public and internal lengths of uniqueid. The
internal length can handle a max public uniqueid plus an appended ;2.
* free() and ast_free() are NULL tolerant so they don't need a NULL test
before calling.
* Made use better struct initialization format instead of the position
dependent initialization format. Also anything not explicitly initialized
in the struct is initialized to zero by the compiler.
* Made ast_channel_internal_set_fake_ids() use the safer
ast_copy_string() instead of strncpy().
Mark Michelson [Wed, 19 Mar 2014 17:27:57 +0000 (17:27 +0000)]
PJSIP: Allow for identify sections to be specified in sorcery.conf.
"identify" is a special type of configuration object in PJSIP because
unlike the other objects, it is not provided by the base res_pjsip module.
Instead, it is provided by the res_pjsip_endpoint_identifier_ip module. If
using the default sorcery wizard (config,criteria=type=identify) then things
work because the module that applies the default wizard is the correct module.
However, if attempting to use sorcery.conf to apply an alternate wizard, it
was not possible. If you attempted to specify the identify object type in the
res_pjsip section, then the object could not be registered since the object
was undocumented for the res_pjsip module. There was no alternate configuration
section defined for it, so you were out of luck if you wanted to override the
default wizard.
With this change, the identify section will properly have a sorcery.conf-based
wizard applied when the identify definition is within the res_pjsip_endpoint_identifier_ip
section.
........
Merged revisions 410933 from http://svn.asterisk.org/svn/asterisk/branches/12
Joshua Colp [Wed, 19 Mar 2014 12:54:25 +0000 (12:54 +0000)]
res_stasis: Extend bridge type to be a comma separated list of bridge attributes.
This change turns the bridge type field into a comma separated list of attributes.
These attributes include: mixing, holding, dtmf_events, and proxy_media. By setting
the various attributes a user can control the type of bridge created with the
behavior they need for their application.
(closes issue ASTERISK-23437)
Reported by: Matt Jordan
ARI: allow json content type with zero length body
When a request was received with a Content-type of json,
the body was sent for json parsing - even if it was zero
length. This resulted in ARI requests failing that were
valid, such as a channel DELETE with no parameters. The
code has now been changed to skip json parsing with zero
content length.
(closes issue SWP-6748)
Reported by: Samuel Galarneau
Review: https://reviewboard.asterisk.org/r/3360/
........
Merged revisions 410858 from http://svn.asterisk.org/svn/asterisk/branches/12
Matthew Jordan [Tue, 18 Mar 2014 15:28:45 +0000 (15:28 +0000)]
cdr: Add asserts for when we don't know about a CDR for a channel
In the CDR core, every channel should either be filtered out (due to being an
'internal' channel used as an implementation detail, such as playing media
back into a bridge) or it should get a CDR. Even if that CDR ends up being
discarded, we still give the channel a CDR in case we end up needing it. If we
hit a situation where a channel does not have a CDR, we should blow up in
-dev-mode. Asserts are appropriate for that.
This patch adds those asserts, as they would have quickly caught the error
fixed by r410814.
........
Merged revisions 410861 from http://svn.asterisk.org/svn/asterisk/branches/12
g711_free() was introduced in spandsp 0.0.6pre4 and g711_release() became a
noop. I opted not to remove the call to g711_release() since it is harmless
and to call g711_free() if we have a sufficiently recent version of spandsp.
(issue ASTERISK-20149)
Reported by: Alexandr Gordeev
........
Merged revisions 410829 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 410830 from http://svn.asterisk.org/svn/asterisk/branches/12
Joshua Colp [Mon, 17 Mar 2014 22:54:32 +0000 (22:54 +0000)]
res_pjsip: Enable PJSIP DNS client support.
This change enables DNS client support within PJSIP. System
nameservers are automatically discovered using res_init or
res_ninit. If this fails then PJSIP will resort to using
gethostbyname for resolution.
By enabling this support we gain SRV support, failover, and
weight support.
(closes issue ASTERISK-23435)
Reported by: Matt Jordan
Joshua Colp [Mon, 17 Mar 2014 22:46:56 +0000 (22:46 +0000)]
res_pjsip_multihomed: Make address replacement less aggressive.
This change makes the res_pjsip_multihomed module less aggressive when
changing the address in messages. It will now only occur if the transport
in use is bound to the any address OR if the system determined source
address matches the bound address of the transport in use.
Russ Meyerriecks [Mon, 17 Mar 2014 22:24:03 +0000 (22:24 +0000)]
callerid: Logic error in checksum processing
Callerid checksum-ing was being handled incorrectly here. When the checksum is
calculated to be 0x00, it will perform 0x100-0x00 which results in 0x100. This
value will then fail the otherwise correct callerid message.
This patch changes the logic to simply add the calculated checksum to the
transmitted 2's compliment checksum.
........
This is a merge of merged revisions 410750 410747 from http://svn.asterisk.org/svn/asterisk/branches/12
I didn't want a broken patch to be comitted to trunk so I pre-merge merged them.
Mark Michelson [Mon, 17 Mar 2014 19:35:17 +0000 (19:35 +0000)]
Revert changes to sorcery that accidentally got committed.
These changes were still up for review and have not been approved
yet. I must have had the changes in my working copy when making
a different change.
........
Merged revisions 410696 from http://svn.asterisk.org/svn/asterisk/branches/12
Mark Michelson [Mon, 17 Mar 2014 17:22:12 +0000 (17:22 +0000)]
Fix stuck channel in ARI through the introduction of synchronous bridge actions.
Playing back a file to a channel in an ARI bridge would attempt to wait until
the playback concluded before returning. The method used involved signaling the
waiting thread in the ARI custom playback function.
The problem with this is that there were some corner cases that were not accounted for:
* If a bridge channel could not be found, then we never would attempt the playback but
would still attempt to wait for the playback to complete.
* If the bridge playfile action failed to queue, we would still attempt to wait for the
playback to complete.
* If the bridge playfile action were queued but some circumstance caused the playback
not to occur (the bridge dies, the channel is removed from the bridge), then we would
never be notified.
The solution to this is to move the waiting logic into the bridge code. A new bridge
API function is added to queue a synchronous action on a bridge. The waiting thread
is notified when the queued frame has been freed, either due to an error occurring
or due to successful playback. As a failsafe, the waiting thread has a 10 minute
timeout just in case there is a frame leak somewhere.
Matthew Jordan [Sun, 16 Mar 2014 20:27:28 +0000 (20:27 +0000)]
stasis/app.c: Add some extra debugging for subscription counts
Events are sent to a connected ARI application based on the things that ARI
application cares about. These subscriptions can be set up implicitly - such
as when that ARI application creates a new object - or explicitly, via the
application resource's subscription operations. Debugging *why* something was
being sent to an application - or why something was not being sent to an
application - was a bit tricky, as there was no debug information for the
subscriptions.
This patch adds some debug level 3 statements that show the subscription counts
for applications. (Level 3 was chosen as it matches the verbose level 3
statements elsewhere)
........
Merged revisions 410650 from http://svn.asterisk.org/svn/asterisk/branches/12
Mark Michelson [Fri, 14 Mar 2014 18:11:55 +0000 (18:11 +0000)]
Handle the return values of realtime updates and stores more accurately.
Realtime backends' update and store callbacks return the number of rows affected,
or -1 if there was a failure. There were a couple of issues:
* The config API was treating 0 as a successful return, and positive values as
a failure. Now the config API treats anything >= 0 as a success.
* res_sorcery_realtime was treating 0 as a successful return from the store
procedure, and any positive values as a failure. Now sorcery treats anything
> 0 as a success. It still considers 0 a "failure" since there is no change
to report to observers.
Mark Michelson [Fri, 14 Mar 2014 18:05:04 +0000 (18:05 +0000)]
Prevent conflicts regarding unsolicited and solicited MWI to an endpoint.
If an endpoint is receiving unsolicited MWI for a mailbox and then attempts
to subscribe to an AOR that provides MWI for the same mailbox, then the SUBSCRIBE
is rejected with a 500 response.
Mark Michelson [Fri, 14 Mar 2014 16:19:21 +0000 (16:19 +0000)]
Prevent delayed astdb syncs.
The syncing thread sleeps for a second before waiting to be
told to attempt to sync again. If a signal were sent during this
sleeping period, we would end up having to wait until the next
sync signal occurred in order to sync up the astdb.
This code rearrangement also ensures that any pending transactions
will be synced prior to Asterisk shutting down.
Patches: db_sync.patch by John Hardin (License #6512)
........
Merged revisions 410556 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 410559 from http://svn.asterisk.org/svn/asterisk/branches/12
This fixes an issue where a Stasis application running over ARI and
subscribed to ari/events could miss the ChannelEnteredBridge event
because it did not subscribe to the new bridge fast enough.
To accomplish this, it subscribes the application controlling the
channel to the new bridge before adding it to that bridge which
required the stasis_app_control structure to maintain a reference to
the stasis_app.
res_pjsip_multihomed: Fix a bug where the 200 OK for a REGISTER would contain the wrong contact.
........
r410510 | file | 2014-03-13 06:24:17 -0700 (Thu, 13 Mar 2014) | 2 lines
res_pjsip_multihomed: Remove change for testing fix.
........
Merged revisions 410509-410510 from http://svn.asterisk.org/svn/asterisk/branches/12
Richard Mudgett [Wed, 12 Mar 2014 19:06:52 +0000 (19:06 +0000)]
res_musiconhold.c: Generate MOH start/stop events whenever the MOH stream is started/stopped.
* Made res_musiconhold.c always post the MusicOnHoldStart/MusicOnHoldStop
events when it actually starts/stops the music streams. This allows the
events to always happen when MOH starts/stops. The event posting code was
moved to the MOH alloc/release routines.
* Made channel_do_masquerade() stop any MOH on the original channel before
masquerading so the original channel will get a stop event with correct
information.
* Cleaned up a couple odd codings in moh_files_alloc() and moh_alloc()
dealing with the music state variable.
(issue ASTERISK-23311)
Reported by: Benjamin Keith Ford
Richard Mudgett [Wed, 12 Mar 2014 18:47:10 +0000 (18:47 +0000)]
app_confbridge: Make explicitly stop MOH if a user is kicked or hangs up while MOH is playing.
When MOH is playing to a user in a conference and the user is kicked or
hangs up from the conference then the AMI MusicOnHoldStop events didn't
happen. (Asterisk v11 AMI event: MusicOnHold, state:Stop)
(closes issue ASTERISK-23311)
Reported by: Benjamin Keith Ford
Joshua Colp [Wed, 12 Mar 2014 12:51:34 +0000 (12:51 +0000)]
res_pjsip_multihomed: Fix a bug where outgoing messages for TCP would go out using UDP.
This change fixes a bug where the code which changes the transport did not check whether
the message is going out over UDP or not before changing it. For TCP and TLS transports
we don't need to change the transport as the correct one is already chosen.
........
Merged revisions 410471 from http://svn.asterisk.org/svn/asterisk/branches/12
Joshua Colp [Tue, 11 Mar 2014 16:07:42 +0000 (16:07 +0000)]
res_pjsip_multihomed: Add module which places the correct address within messages.
Due to how messages are handled within PJSIP it is not until a message is actually
sent that the destination is reliably known. This means that the addresses placed
within the message may not be of the interface the message is being sent out on.
This module determines what interface a message is being sent on and updates the
message to contain the correct address if applicable.
This module was tested by myself in a virtualized environment with multiple interfaces
and also by Kinsey Moore in the following configuration:
Networks:
* 10.24.16.0/21
** hard phone
** default gateway
* 10.24.64.0/21
** softphone with pjsip-based stack
Transport details:
bind address: 0.0.0.0
protocol: UDP
All endpoints were tested with explicitly configured transports and unconfigured transports.
This was tested with inbound and outbound calls, both of which were experiencing detrimental
effects from incorrect IP addresses in SIP messages. These effects were only experienced by the
soft phone on the 10.24.64.0 network since the messages to the hard phone on the 10.24.16.0
network had the correct IP address.
Richard Mudgett [Mon, 10 Mar 2014 17:21:01 +0000 (17:21 +0000)]
AST-2014-001: Stack overflow in HTTP processing of Cookie headers.
Sending a HTTP request that is handled by Asterisk with a large number of
Cookie headers could overflow the stack.
Another vulnerability along similar lines is any HTTP request with a
ridiculous number of headers in the request could exhaust system memory.
(closes issue ASTERISK-23340)
Reported by: Lucas Molas, researcher at Programa STIC, Fundacion; and Dr. Manuel Sadosky, Buenos Aires, Argentina
........
Merged revisions 410380 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 410381 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 410383 from http://svn.asterisk.org/svn/asterisk/branches/12
Kinsey Moore [Mon, 10 Mar 2014 13:30:51 +0000 (13:30 +0000)]
AST-2014-002: chan_sip: Exit early on bad session timers request
This change allows chan_sip to avoid creation of the channel and
consumption of associated file descriptors altogether if the inbound
request is going to be rejected anyway.
George Joseph [Sat, 8 Mar 2014 16:50:36 +0000 (16:50 +0000)]
pjsip_cli: Create pjsip show channel and contact, and general cli code cleanup.
Created the 'pjsip show channel' and 'pjsip show contact' commands.
Refactored out the hated ast_hashtab. Replaced with ao2_container.
Cleaned up function naming. Internal only, no public name changes.
Cleaned up whitespace and brace formatting in cli code.
Changed some NULL checking from "if"s to ast_asserts.
Fixed some register/unregister ordering to reduce deadlock potential.
Fixed ast_sip_location_add_contact where the 'name' buffer was too short.
Fixed some self-assignment issues in res_pjsip_outbound_registration.
Matthew Jordan [Sat, 8 Mar 2014 15:45:59 +0000 (15:45 +0000)]
resource_channels: Check if a passed in ID is NULL before checking its length
Calling strlen on a NULL string is explosive. This patch checks whether or not
the passed in string is NULL or zero length before checking to see if the
string is too long.
........
Merged revisions 410274 from http://svn.asterisk.org/svn/asterisk/branches/12
Matthew Jordan [Fri, 7 Mar 2014 21:54:01 +0000 (21:54 +0000)]
config_options: Display the see-also information for CLI config option help
The config option help information has always parsed the <see-also> tags in the
XML documentation. Unfortunately, it just never bothered displaying them on
the CLI. With this patch, when you execute 'config show help [module] [obj]
[option]', it will display what other options are useful to you.
(closes issue ASTERISK-22008)
Reported by: Richard Mudgett
........
Merged revisions 410209 from http://svn.asterisk.org/svn/asterisk/branches/12
Matthew Jordan [Fri, 7 Mar 2014 21:28:12 +0000 (21:28 +0000)]
res_pjsip: Fix documentation for one touch recording see-also links
The one touch recording options have several see-also links between the
various configuration options. These were 'broken' by the snake casing
of those options. This patch corrects the see-also links such that they
reference the correct option names.
........
Merged revisions 410194 from http://svn.asterisk.org/svn/asterisk/branches/12
Mark Michelson [Fri, 7 Mar 2014 21:23:39 +0000 (21:23 +0000)]
Make res_sorcery_realtime filter unknown retrieved results.
When retrieving data from a database or other realtime backend, it's quite
possible to retrieve variables that Asterisk does not care about but that
are legitimate to exist. Asterisk does not need to throw a hissy fit when
these variables are encountered but rather just filter them out.
In order to prevent confusion over the allow and disallow
list of codecs being the same an option for registering a
field as an alias is added. The alias field will be read
from the configuration file, but afterwards is not listed
as a known field. With disallow set as an alias, the CLI
command pjsip show endpoint # will list the allow= field,
but not the disallow field.
Richard Mudgett [Fri, 7 Mar 2014 20:41:13 +0000 (20:41 +0000)]
stasis cache: Enhance to keep track of an item from different entities.
A stasis cache entry now contains more than a single message/snapshot. It
contains messages/snapshots for the local entity as well as any remote
entities that post to the cached item. In addition callbacks can be
supplied when the cache is created to compute and post the aggregate
message/snapshot representing all entities stored in the cache entry.
* All stasis messages now have an eid to indicate what entity posted it.
* The stasis cache enhancements allow device state to cache and aggregate
the device states from local and remote entities in a single operation.
The cached aggregate device state is available immediately after it is
posted to the stasis bus. This improves performance by eliminating a
cache dump and associated ao2 container traversals to calculate the
aggregate state.
(closes issue ASTERISK-23204)
Reported by: Mark Michelson
uniqueid: channel linkedid, ami, ari object creation with id's
Much needed was a way to assign id to objects on creation, and
much change was necessary to accomplish it. Channel uniqueids
and linkedids are split into separate string and creation time
components without breaking linkedid propgation. This allowed
the uniqueid to be specified by the user interface - and those
values are now carried through to channel creation, adding the
assignedids value to every function in the chain including the
channel drivers. For local channels, the second channel can be
specified or left to default to a ;2 suffix of first. In ARI,
bridge, playback, and snoop objects can also be created with a
specified uniqueid.
Along the way, the args order to allocating channels was fixed
in chan_mgcp and chan_gtalk, and linkedid is no longer lost as
masquerade occurs.
Matthew Jordan [Fri, 7 Mar 2014 05:04:32 +0000 (05:04 +0000)]
chan_sip: Allow static realtime members to be qualified during module load.
When a static realtime peer with qualify=yes is loaded, Asterisk will fail to
send an OPTIONS request due to the lastms being equal to 0. This results in
the peer being unable to receive calls from Asterisk because the status is
permanently UNKNOWN.
This patch allows an OPTIONS request to be sent during module load by
ignoring the lastms value on startup only.
Review: https://reviewboard.asterisk.org/r/3294/
(closes issue ASTERISK-17523)
Reported by: Maciej Krajewski
Tested by: wushumasters
patches:
realtime_fix_11.7.0.txt uploaded by Trevor Peirce (license 6112)
........
Merged revisions 410105 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 410106 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 410107 from http://svn.asterisk.org/svn/asterisk/branches/12
Russell Bryant [Thu, 6 Mar 2014 23:43:34 +0000 (23:43 +0000)]
moh: fix a refcount error with realtime MOH
I observed a crash in res_musiconhold on an Asterisk 11 system using realtime
MOH. Investigation of the backtrace showed a corrupt mohclass, implying that
it got destroyed before the code expected it to. I went looking for reference
counting errors that could have caused this crash and this patch this result.
It contains 2 changes.
1) Remove a usless block of code that was impossible to reach. There was even
a comment indicating that it was impossible to reach. The conditional includes
"!ast_test_flag(global_flags, MOH_CACHERTCLASSES)" and it's inside of an if
block with the opposite check "ast_test_flag(global_flags,
MOH_CACHERTCLASSES)". There's no good reason to keep it around.
2) A similar block to #1 contained a reference counting error. It stores
state->class in the local variable mohclass without increasing its reference
count. The reference count on mohclass is decremented at the end of the
function. This block of code probably very rarely runs, which would help
explain why this system was working fine for many months before experiencing a
crash.
George Joseph [Thu, 6 Mar 2014 22:39:54 +0000 (22:39 +0000)]
sorcery: Create AST_SORCERY dialplan function.
This patch creates the AST_SORCERY dialplan function which allows someone to
retrieve any value from a sorcery-based config file. It's similar to
AST_CONFIG.
The creation of the function itself was fairly straightforward but it required
changes to the underlying sorcery infrastructure that rippled into individual
sorcery objects. The changes stemmed from inconsistencies in how sorcery
created ast_variable objectsets from sorcery objects and the inconsistency
in how individual objects used that feature especially when it came to
parameters that can be specified multiple times like contact in aor and match
in identify. You can read more here...
http://lists.digium.com/pipermail/asterisk-dev/2014-February/065202.html
So, what this patch does, besides actually creating the AST_SORCERY function,
is the following...
* Creates ast_variable_list_append which is a helper to append one ast_variable
list to another.
* Modifies the ast_sorcery_object_field_register functions to accept the
already-defined sorcery_fields_handler callback.
* Modifies ast_sorcery_objectset_create to accept a parameter indicating return
type preference...a single ast_variable with all values concatenated or an
ast_variable list with multiple entries. Also fixed a few bugs.
* Modifies individual sorcery object implementations to use the new function
definition of the ast_sorcery_object_field_register functions.
* Modifies location.c and res_pjsip_endpoint_identifier_ip.c to implement
sorcery_fields_handler handlers so they return multiple occurrences as an
ast_variable_list.
* Added a whole bunch of tests to test_sorcery.
Jonathan Rose [Thu, 6 Mar 2014 19:04:58 +0000 (19:04 +0000)]
pjsip configuration: Make transport TOS values consistent with endpoints
Transport TOS values were interpreted as DSCP values without being documented
as such. Endpoint TOS values (tos_audio/tos_video) behaved normally as TOS
values have historically. This patch makes the transport TOS values behave as
TOS values and makes all TOS values readable as string values (e.g. AF11).
In addition, alembic scripts have been updated to use the proper field types
for all TOS/COS values.
(issue ASTERISK-23235)
Reported by: George Joseph
Review: https://reviewboard.asterisk.org/r/3304/
........
Merged revisions 410028 from http://svn.asterisk.org/svn/asterisk/branches/12
Mark Michelson [Thu, 6 Mar 2014 15:58:13 +0000 (15:58 +0000)]
Don't attempt to link in an aggregate MWI subscription if an endpoint does not aggregate MWI.
Attempting to link a NULL object into an ao2 container had been benign previously, but since
enabling DO_CRASH in the testsuite, this is now causing a crash. It's better to be right
here anyway.
........
Merged revisions 410011 from http://svn.asterisk.org/svn/asterisk/branches/12
This patch creates the AST_SORCERY dialplan function which allows someone to
retrieve any value from a sorcery-based config file. It's similar to
AST_CONFIG.
The creation of the function itself was fairly straightforward but it required
changes to the underlying sorcery infrastructure that rippled into individual
sorcery objects. The changes stemmed from inconsistencies in how sorcery
created ast_variable objectsets from sorcery objects and the inconsistency
in how individual objects used that feature especially when it came to
parameters that can be specified multiple times like contact in aor and match
in identify. You can read more here...
http://lists.digium.com/pipermail/asterisk-dev/2014-February/065202.html
So, what this patch does, besides actually creating the AST_SORCERY function,
is the following...
* Creates ast_variable_list_append which is a helper to append one ast_variable
list to another.
* Modifies the ast_sorcery_object_field_register functions to accept the
already-defined sorcery_fields_handler callback.
* Modifies ast_sorcery_objectset_create to accept a parameter indicating return
type preference...a single ast_variable with all values concatenated or an
ast_variable list with multiple entries. Also fixed a few bugs.
* Modifies individual sorcery object implementations to use the new function
definition of the ast_sorcery_object_field_register functions.
* Modifies location.c and res_pjsip_endpoint_identifier_ip.c to implement
sorcery_fields_handler handlers so they return multiple occurrences as an
ast_variable_list.
* Added a whole bunch of tests to test_sorcery.
Matthew Jordan [Thu, 6 Mar 2014 02:22:59 +0000 (02:22 +0000)]
res_fax_spandsp: Fix crash when passing ulaw/alaw data to spandsp
When acting as a T.38 fax gateway, res_fax_spandsp would at times cause a crash
in libspandsp. This would occur when, during fax tone detection, a ulaw/alaw
frame would be passed to modem_connect_tones_rx. That particular routine
expects the data to be in slin format. This patch looks at the frame type and,
if the data is ulaw/alaw, converts the format to slin before passing it to
modem_connect_tones_rx.
Review: https://reviewboard.asterisk.org/r/3296
(closes issue ASTERISK-20149)
Reported by: Alexandr Gordeev
Tested by: Michal Rybarik
patches:
spandsp_g711decode.diff uploaded by Michal Rybarik (license 6578)
........
Merged revisions 409990 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 409991 from http://svn.asterisk.org/svn/asterisk/branches/12
Kinsey Moore [Wed, 5 Mar 2014 20:41:37 +0000 (20:41 +0000)]
config: Fix inverted test
The test of the result of the stat() call was inverted such that its
output was only used if the call failed. This inverts the test so that
the output of stat() is used correctly. This was causing full reloads
on unchanged files.
(closes issue ASTERISK-23383)
Reported by: David Woolley
........
Merged revisions 409916 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 409917 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 409918 from http://svn.asterisk.org/svn/asterisk/branches/12
Kinsey Moore [Wed, 5 Mar 2014 19:28:31 +0000 (19:28 +0000)]
bridge_native_rtp: Fix crash involving masquerade
It is possible for a channel to be masqueraded out of a bridge which
means it may no longer have RTP glue to check upon leaving said bridge.
If this situation occurred (it's possible at least during dial and call
pickup) then Asterisk would crash. This change makes sure the glue is
checked before use.
(closes issue AST-1290)
Reported by: John Bigelow
........
Merged revisions 409900 from http://svn.asterisk.org/svn/asterisk/branches/12
Richard Mudgett [Wed, 5 Mar 2014 18:51:37 +0000 (18:51 +0000)]
alembic: Add missing queue and CDR table creation scripts.
* Added the queues and queue_members tables to the config alembic scripts.
* Added the CDR table alembic creation script. The CDR table is more of
an example for new setups since the actual table can be fully customized
in cdr_adaptive_odbc.conf.
David M. Lee [Wed, 5 Mar 2014 16:58:21 +0000 (16:58 +0000)]
Corrected cross-platform stat nanosecond code
When nanosecond time resolution was added for identifying config file
changes, it didn't cover all of the myriad of ways that one might obtain
nanosecond time resolution off of struct stat.
Rather than complicate the #if even further figuring out one system from
the next, this patch directly tests for the three struct members I know
about today, and #ifdef's accordingly.
Moises Silva [Wed, 5 Mar 2014 16:22:44 +0000 (16:22 +0000)]
Fix WebRTC over WSS not working
Several fixes for the WebSockets implementation in res/res_http_websocket.c
* Flush the websocket session FILE* as fwrite() may not actually guarantee sending
the data to the network. If we do not flush, it seems that buffering on the SSL
socket for outbound messages causes issues
* Refactored ast_websocket_read to take into account that SSL file descriptors
may be ready to read via fread() but poll() will not actually say so because
the data was already read from the network buffers and is now in the libc buffers
........
Correct RTP handling in chan_unistim and fix transfer process broken in previous fix:
- Fixed too early RTP setup with phone, that cause no ringback tone on caller side
- Handle call transfer cancel only in STATE_CALL case (related to ASTERISK-23073)
(Reported by: Németh Tamás, niurkin sil)
........
Merged revisions 409761 from http://svn.asterisk.org/svn/asterisk/branches/11
Add update_peer function to unistim_rtp_glue, improve other unistim_rtp_glue functions conforming to other channel drivers. Do not forget auto-detected and user-selected phone settings on 'unistim reload'
........
Merged revisions 409705 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 409745 from http://svn.asterisk.org/svn/asterisk/branches/11
func_audiohookinheritance: Check If A Channel Was Specified
This patch prevents a crash when using the function audiohookinheritance without
setting the channel.
(closes issue ASTERISK-23104)
Reported by: Joel Vandal
Tested by: Joel Vandal
Patches:
asterisk-23104_audiohook_inherit_no_channel-11.diff
uploaded by Michael L. Young (license 5026)
Kinsey Moore [Tue, 4 Mar 2014 16:55:43 +0000 (16:55 +0000)]
AO2: Add an assert for bad objects
This adds an assert that will only be active if Asterisk is compiled
with DO_CRASH and allows the testsuite to fail tests that would
otherwise require log file parsing.
........
Merged revisions 409566 from http://svn.asterisk.org/svn/asterisk/branches/1.8
........
Merged revisions 409567 from http://svn.asterisk.org/svn/asterisk/branches/11
........
Merged revisions 409568 from http://svn.asterisk.org/svn/asterisk/branches/12
projects / thirdparty / asterisk.git / log
