Fix regression causing abort in voicemail after opening a mailbox with no mesgs.
In order to be more safe, some error handling code was changed to respect more
error conditions including the potential memory allocation failure for deleted
and heard message tracking introduced in 293004. However, last_message_index
returns -1 for zero messages (perhaps as expected) and was triggering the
stricter error checking. Because last_message_index is only called directly
in one place, just return 0 from open_mailbox (for file based storage) when no
messages are detected unless a real error has occurred.
Russell Bryant [Thu, 11 Nov 2010 22:13:38 +0000 (22:13 +0000)]
Remove most of the contents of the doc dir in favor of the wiki content.
This merge does the following things:
* Removes most of the contents from the doc/ directory in favor
of the wiki - http://wiki.asterisk.org/
* Updates the build_tools/prep_tarball script to know how to export
the contents of the wiki in both PDF and plain text formats so that
the documentation is still included in Asterisk release tarballs.
Fix problem with qualify option packets for realtime peers never stopping.
The option packets not only never stopped, but if a realtime peer was not in
the peer list multiple options dialogs could accumulate over time. This
scenario has the potential to progress to the point of saturating a link just
from options packets. The fix was to ensure that the poke scheduler checks to
see if a peer is in the peer list before continuing to poke. The reason a peer
must be in the peer list to be able to properly manage an options dialog is
because otherwise the call pointer is lost when the peer is regenerated from
the database, which is how existing qualify dialogs are detected.
Copied from some notes from the original author (Russell):
Deadlock scenario:
Thread 1: device state change thread
Holds - rdlock on contexts
Holds - hints lock
Waiting on channels container lock
Thread 2: SIP monitor thread
Holds the "iflock"
Holds a sip_pvt lock
Holds channel container lock
Waiting for a channel lock
Thread 3: A channel thread (chan_local in this case)
Holds 2 channel locks acquired within app_dial
Holds a 3rd channel lock it got inside of chan_local
Holds a local_pvt lock
Waiting on a rdlock of the contexts lock
A bunch of other threads waiting on a wrlock of the contexts lock
To address this deadlock, some locking order rules must be put in place and
enforced. Existing relevant rules:
1) channel lock before a pvt lock
2) contexts lock before hints lock
3) channels container before a channel
What's missing is some enforcement of the order when you involve more than any
two. To fix this problem, I put in some code that ensures that (at least in the
code paths involved in this bug) the locks in (3) come before the locks in (2).
To change the operation of thread 1 to comply, I converted the storage of hints
to an astobj2 container. This allows processing of hints without holding the
hints container lock. So, in the code path that led to thread 1's state, it no
longer holds either the contexts or hints lock while it attempts to lock the
channels container.
Richard Mudgett [Tue, 9 Nov 2010 16:55:32 +0000 (16:55 +0000)]
Analog lines do not transfer CONNECTED LINE or execute the interception macros.
Add connected line update for sig_analog transfers and simplify the
corresponding sig_pri and chan_misdn transfer code.
Note that if you create a three-way call in sig_analog before transferring
the call, the distinction of the caller/callee interception macros make
little sense. The interception macro writer needs to be prepared for
either caller/callee macro to be executed. The current implementation
swaps which caller/callee interception macro is executed after a three-way
call is created.
Fix playback failure when using IAX with the timerfd module.
To fix this issue the alert pipe will now be used when the timerfd module is
in use. There appeared to be a race that was not solved by adding locking in the
timerfd module, but needed to be there anyway. The race was between the timer
being put in non-continuous mode in ast_read on the channel thread and the IAX
frame scheduler queuing a frame which would enable continuous mode before the
non-continuous mode event was read. This race for now is simply avoided.
Richard Mudgett [Mon, 8 Nov 2010 17:16:01 +0000 (17:16 +0000)]
valgrind reported references to freed memory during a mISDN hangup collision.
Bad things have been happening in chan_misdn because the chan_misdn
channel private struct chan_list is not protected from reentrancy. Hangup
collisions have be causing read and write accesses to freed memory.
Converted chan_misdn struct chan_list to an ao2 object for its reference
counting feature.
**********
Removed an impediment to converting chan_list to an ao2 object.
The use of the other_ch member in chan_list is shaky at best. It is set
if the incoming and outgoing call legs are mISDN. The use of the other_ch
member goes against the Asterisk architecture and can even cause problems.
1) It is used to disable echo cancellation. This could be bad if the call
is forked and the winning call leg is not mISDN or the winning call leg is
not the last mISDN channel called by the fork. The other_ch would become
a dangling pointer.
2) It is used when the far end is alerting to hear the far end's inband
audio instead of Asterisk's generated ringback tone. This is bad if the
call is forked. You would only hear the last forked mISDN channel and it
may not be ringing yet.
The other_ch would become a dangling pointer if the call is later
transferred.
**********
codecs/codec_dahdi: Prevent "choppy" audio when receiving unexpected frame sizes.
dahdi-linux 2.4.0 (specifically commit 9034) added the capability for
the wctc4xxp to return more than a single packet of data in response to
a read. However, when decoding packets, codec_dahdi was still assuming
that the default number of samples was in each read.
In other words, each packet your provider sent you, regardless of size,
would result in 20 ms of decoded data (30 ms if decoding G723). If your
provider was sending 60 ms packets then codec_dahdi would end up
stripping 40 ms of data from each transcoded frame resulting in "choppy"
audio.
This would only affect systems where G729 packets are arriving in sizes
greater than 20ms or G723 packets arriving in sizes greater than 30ms.
Party A in an analog 3-way call would continue to hear ringback after party C answers.
All parties are analog FXS ports.
1) A calls B.
2) A flash hooks to call C.
3) A flash hooks to bring C into 3-way call before C answers. (A and B hear ringback)
4) C answers
5) A continues to hear ringback during the 3-way call. (All parties can hear each other.)
* Fixed use of wrong variable in dahdi_bridge() that stopped ringback on
the wrong subchannel.
* Made several debug messages have more information.
A similar issue happens if B and C are SIP channels. B continues to hear
ringback. For some reason this only affects v1.8 and trunk.
* Don't start ringback on the real and 3-way subchannels when creating the
3-way conference. Removing this code is benign on v1.6.2 and earlier.
........
................
Terry Wilson [Wed, 3 Nov 2010 18:05:14 +0000 (18:05 +0000)]
Avoid valgrind warnings for ast_rtp_instance_get_xxx_address
The documentation for ast_rtp_instance_get_(local/remote)_address stated that
they returned 0 for success and -1 on failure. Instead, they returned 0 if the
address structure passed in was already equivalent to the address instance
local/remote address or 1 otherwise. 90% of the calls to these functions
completely ignored the return address and passed in an uninitialized struct,
which would make valgrind complain even though the operation was technically
safe.
This patch fixes the documentation and converts the get_xxx_address functions
to void since all they really do is copy the address and cannot fail.
Additionally two new functions
(ast_rtp_instance_get_and_cmp_(local/remote)_address) are created for the 3
times where the return value was actually checked. The
get_and_cmp_local_address function is currently unused, but exists for the sake
of symmetry.
The only functional change as a result of this change is that we will not do an
ast_sockaddr_cmp() on (mostly uninitialized) addresses before doing the
ast_sockaddr_copy() in the get_*_address functions. So, even though it is an
API change, it shouldn't have a noticeable change in behavior.
Make warning message have more useful information in it.
Change "Unable to get index, and nullok is not asserted" to "Unable to get
index for '<channel-name>' on channel <number> (<function>(), line
<number>)".
........
................
Terry Wilson [Mon, 1 Nov 2010 14:58:00 +0000 (14:58 +0000)]
Only offer codecs both sides support for directmedia
When using directmedia, Asterisk needs to limit the codecs offered to just
the ones that both sides recognize, otherwise they may end up sending audio
that the other side doesn't understand.
Jeff Peeler [Fri, 29 Oct 2010 21:48:38 +0000 (21:48 +0000)]
Modify sip_setoption to not complain about unknown options.
This now behaves just like the other setoption callbacks. For the curious the
offending option for the reporter was AST_OPTION_CHANNEL_WRITE which was getting
passed due to a fix for chan_local in 286189.
"!00" evaluated as false, which is incorrect. Fixing.
Reported (though the reporter did not understand he was reporting a bug) on the asterisk-users list:
http://lists.digium.com/pipermail/asterisk-users/2010-October/255505.html
........
................
r293196 | tilghman | 2010-10-28 14:54:34 -0500 (Thu, 28 Oct 2010) | 12 lines
Merged revisions 293194 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4
"!00" evaluated as false, which is incorrect. Fixing.
Reported (though the reporter did not understand he was reporting a bug) on the asterisk-users list:
http://lists.digium.com/pipermail/asterisk-users/2010-October/255505.html
........
................
Fix inprocess_container in voicemail to correctly restrict max messages.
The comparison function logic was off, so the number of sessions for a given
mailbox were not being incremented properly. This problem caused the maximum
number of messages per folder to not be respected when simultaneously leaving
multiple voicemails just below the threshold.
These problems should be fixed by the above, but just in case:
Fixed resequence_mailbox to rely on the actual number of detected number of
files in a directory rather than just assuming only 10 messages more than the
maximum had been left. Also if more messages than the maximum are deleted they
are actually removed now.
The second purpose of this commit should have been separated out probably, but
is related to the above. Again, if the number of messages in a given voicemail
folder exceeds the maximum set limit make sure to allocate enough space for the
deleted and heard index tracking array.
A few random fixes:
There was a forgotten decrement of the inprocess count in imap_store_file.
When using IMAP storage, do not look in the directory where file based storage
messages may still reside and influence the message count.
This patch turns chan_local pvts into astobj2 objects.
chan_local does some dangerous things involving deadlock avoidance.
tech_pvt functions like hangup and queue_frame are provided with a
locked channel upon entry. Those functions are completely safe as
long as you don't attempt to give up that channel lock, but that is
impossible to guarantee due to the required deadlock avoidance necessary
to lock both the tech_pvt and both channels involved.
In the past, we have tried to account for this by doing things like
setting a "glare" flag that indicates what function should destroy the
pvt. This was used in local_hangup and local_queue_frame to decided
who should destroy the pvt if they collided in separate threads. I
have removed the need to do this by converting all chan_local tech_pvts
to astobj2. This means we can ref a pvt before deadlock avoidance
and not have to worry about that pvt possibly getting destroyed under
us. It also cleans up where we destroy the tech_pvt. The only unlink
from the tech_pvt container occurs in local_hangup now, which is where
it should occur.
Since there still may be thread collisions on some functions like
local_hangup after deadlock avoidance, I have added some checks to detect
those collisions and exit appropriately. I think this patch is going to
solve quite a bit of weirdness we have had with local channels in the past.
........
................
Update the LDIF file for LDAP.
The LDIF file asterisk.ldif was quite a bit out of date from the asterisk.ldap-schema file, so I've
now updated that to be in sync. The asterisk.ldif file being out of sync was a problem on my systems
where I was doing an ldapadd to import the schema into the LDAP database, and the existing file
would cause problems and ERROR messages when registering.
Additional documention has been added based on feedback in the issue I'm closing.
Mark Michelson [Fri, 22 Oct 2010 17:09:52 +0000 (17:09 +0000)]
Prevent multiple runs of event_sub_test from producing false failure results.
The array of test subscriptions was declared "static," meaning that the
data.count field would retain its value between runs of the test. After the
first test run, this would result in false reports of test failures.
I chose to just remove the "static" keyword from the structure since it's not
a huge deal to construct this structure during each run of the test. Another
alternative would have been to zero out the data.count fields of each test
subscription instead.
Terry Wilson [Fri, 22 Oct 2010 16:49:34 +0000 (16:49 +0000)]
Add TLS cert helper script
This script is useful for quickly generating self-signed CA, server, and client
certificates for use with Asterisk. It is still recommended to obtain
certificates from a recognized Certificate Authority and to develop an
understanding how SSL certificates work. Real security is hard work.
OPTIONS:
-h Show this message
-m Type of cert "client" or "server". Defaults to server.
-f Config filename (openssl config file format)
-c CA cert filename (creates new CA cert/key as ca.crt/ca.key if not passed)
-k CA key filename
-C Common name (cert field)
For a server cert, this should be the same address that clients
attempt to connect to. Usually this will be the Fully Qualified
Domain Name, but might be the IP of the server. For a CA or client
cert, it is merely informational. Make sure your certs have unique
common names.
-O Org name (cert field)
An informational string (company name)
-o Output filename base (defaults to asterisk)
-d Output directory (defaults to the current directory)
Example:
To create a CA and a server (pbx.mycompany.com) cert with output in /tmp:
ast_tls_cert -C pbx.mycompany.com -O "My Company" -d /tmp
This will create a CA cert and key as well as asterisk.pem and the the two
files that it is made from: asterisk.crt and asterisk.key. Copy asterisk.pem
and ca.crt somewhere (like /etc/asterisk) and set tlscertfile=/etc/asterisk.pem
and tlscafile=/etc/ca.crt. Since this is a self-signed key, many devices will
require you to import the ca.crt file as a trusted cert.
To create a client cert using the CA cert created by the example above:
ast_tls_cert -m client -c /tmp/ca.crt -k /tmp/ca.key -C "Joe User" -O \
"My Company" -d /tmp -o joe_user
This will create client.crt/key/pem in /tmp. Use this if your device supports
a client certificate. Make sure that you have the ca.crt file set up as
a tlscafile in the necessary Asterisk configs. Make backups of all .key files
in case you need them later.
Richard Mudgett [Fri, 22 Oct 2010 15:47:08 +0000 (15:47 +0000)]
Connected line is not updated when chan_dahdi/sig_pri or chan_misdn transfers a call.
When a call is transfered by ECT or implicitly by disconnect in sig_pri or
implicitly by disconnect in chan_misdn, the connected line information is
not exchanged. The connected line interception macros also need to be
executed if defined.
The CALLER interception macro is executed for the held call.
The CALLEE interception macro is executed for the active/ringing call.
JIRA ABE-2589
JIRA SWP-2296
Patches:
abe_2589_c3bier.patch uploaded by rmudgett (license 664)
abe_2589_v1.8_v2.patch uploaded by rmudgett (license 664)
David Vossel [Thu, 21 Oct 2010 16:14:33 +0000 (16:14 +0000)]
Fixes recursive lock problem in manager.c
It is possible for a AMI session to freeze because of invalid
use of recursive locks during the EVENT processing. This
patch removes the unnecessary locks.
(closes issue #18167)
Reported by: sustav
Patches:
manager_locking_v1.diff uploaded by dvossel (license 671)
Tested by: sustav
Richard Mudgett [Thu, 21 Oct 2010 01:02:50 +0000 (01:02 +0000)]
Send CONNECT_ACKNOWLEDGE for CIS calls too.
The originator of the Q.SIG call completion signaling link was not changed
to the active state when the CONNECT message came in. The T309 processing
would immediately kill the signaling link because it was not in the active
state.
Terry Wilson [Tue, 19 Oct 2010 19:27:32 +0000 (19:27 +0000)]
Add sip show peer info about crypto and remove dated comment
This patch adds information about the encryption setting to 'sip show
peers' and removes an out-of-date comment from res_srtp.c and instead
directs users to the proper documentation.
Fix improper operator key acceptance and clean up temp recording files.
This is a fix for when pressing the operator key after recording an unavailable,
busy, name, or temporary message in mailbox options. The operator key should not
be accepted here, but should be allowed during the message recording. If the
operator key is pressed during ensure the file is saved or deleted as
apporopriate. Also, ensure removal of temporary recorded files after an early
hang up or when message acceptance confirmation times out.
If the directive 'directory' is relative, make it relative to the
datadir, rather than to the varlibdir. In the sample configuration
it is relative ('moh').
This has no effect unless you have actively set the datadir explicitly
(at build time or at run time).
(closes issue #16906)
Patches:
moh_datadir uploaded by tzafrir (license 46)
We can't hold the lock while pthread_join is called since aji_log_hook will
attempt to lock from the other therad. We reorder the pthread_join and
ast_aji_disconnect so that we don't do an SSL_read() while SSL_shutdown is
running, causing a crash.
........
David Vossel [Thu, 14 Oct 2010 21:27:42 +0000 (21:27 +0000)]
Safer xml parsing, treat all clients the same, and better local candidate selection.
The gtalk channel driver was doing several unsafe operations
in regards to how it parsed incoming XML messages. I have cleaned
that code up so it should be much safer now.
We now treat all clients types the same. We have no reason to
distinguish between GMAIL and GOOGLE VOICE clients anymore because
they all work the same way.
I also modified how the local ip is found. If no bindaddress is provided
in the config file, we attempt to determine the local ip we
would use to connect to google.com. If that fails, then
we fall back to the ast_find_ourip() function as a last resort.
Using the new method makes it much less likely that we would ever
advertise a local RTP candidate as a loopback address.
Paul Belanger [Thu, 14 Oct 2010 15:15:12 +0000 (15:15 +0000)]
Add the ability for ast_find_ourip to return IPv4, IPv6 or both.
While testing chan_gtalk I noticed jabber was using my IPv6 address
and not IPv4. When using bindaddr=0.0.0.0 it is possible for ast_find_ourip()
to return both IPv6 and IPv4 results. Adding a family parameter gives you
the ablility to choose.
Since jabber/gtalk/h323 do not support IPv6, we should only return IPv4 results.
Deadlock between dahdi_exception() and dahdi_indicate().
There is a deadlock between dahdi_exception() and dahdi_indicate() for
analog ports. The call-waiting and three-way-calling feature can
experience deadlock if these features are trying to do something and an
event from the bridged channel happens at the same time.
Deadlock avoidance code added to obtain necessary channel locks before
attemting an operation with call-waiting and three-way-calling.
(closes issue #16847)
Reported by: shin-shoryuken
Patches:
issue_16847_v1.4.patch uploaded by rmudgett (license 664)
issue_16847_v1.6.2.patch uploaded by rmudgett (license 664)
issue_16847_v1.8_v2.patch uploaded by rmudgett (license 664)
Tested by: alecdavis, rmudgett
Don't ignore frames that have been queued when softhangup'd
When an outgoing call is answered and hung up by the far end *very* quickly, we
may not read any frames and therefor end up with a call that displays the wrong
disposition/DIALSTATUS. The reason is because ast_queue_hangup() immediately
sets the _softhangup flag on the channel and then queues the HANGUP control
frame, but __ast_read refuses to read any frames if ast_check_hangup() indicates
that a hangup request has been made (which it will if _softhangup is set). So,
we end up losing control frames.
This change makes __ast_read continue to read frames even if a soft hangup has
been requested. It queues a hangup frame to make sure that __ast_read() will
still eventually return NULL.
Much thanks to David Vossel for all of the reviews, discussion, and help!
Terry Wilson [Wed, 13 Oct 2010 22:24:44 +0000 (22:24 +0000)]
Add a simple AMI client web page
This patch uses the XML docs to parse all of the available AMI commands
and allows you to enter the command name and be presented with a form with
the available fields. You can then rapidly tab through the fields and submit
the command and view the response. It is much faster/easier than having to
use telnet for testing purposes.
Richard Mudgett [Wed, 13 Oct 2010 20:21:02 +0000 (20:21 +0000)]
The chan_dahdi faxdetect option only works for the first FAX call.
The chan_dahdi faxdetect option only works for the first call. After that
the option no longer works. The struct dahdi_pvt.callprogress member is
the encoded user config setting for the callprogress and faxdetect config
options. Changing this value alters the configuration for all following
calls until the chan_dahdi.conf file is reloaded.
* Fixed the chan_dahdi ast_channel_setoption callback to not change the
users faxdetect config setting except for the current call.
* Fixed the chan_dahdi ast_channel_queryoption callback to read the active
DSP setting of the faxdetect option.
* Made actually disable the active faxdetect DSP setting for the current
call on the analog port. my_handle_dtmfup() is used for normal analog
ports. dahdi_handle_dtmfup() is the legacy code and is no longer used
unless in a radio mode.
projects / thirdparty / asterisk.git / log
