The external test suite stops Asterisk using the "core stop gracefully" command.
The logs from the tests show that there are a number of problems with Asterisk
trying to cleanly shut down. This patch addresses the following type of error
that comes from chan_iax2:
[Jun 22 16:58:11] ERROR[29884]: lock.c:129 __ast_pthread_mutex_destroy:
chan_iax2.c line 11371 (iax2_process_thread_cleanup):
Error destroying mutex &thread->lock: Device or resource busy
For an example in the context of a build, see:
http://bamboo.asterisk.org/browse/AST-TRUNK-739/log
The primary purpose of this patch is to change the thread pool shutdown
procedure to be more explicit to ensure that the thread exits from a point
where it is not holding a lock. While testing that, I encountered various
crashes due to the order of operations in unload_module() being problematic.
I reordered some things there, as well.
Review: https://reviewboard.asterisk.org/r/736/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@272370 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Testing proved that if Asterisk sent a connected line reinvite, and
the endpoint to which the reinvite were being sent sent a reinvite, Asterisk
would not properly respond with a 491 response.
The reason is that on connected line reinvites, we set the dialog's invitestate
to INV_CALLING to prevent Asterisk from sending a rapid flurry of connected line
reinvites. For other reinvites we do not do this. Because of the current invitestate,
when Asterisk received the reinvite, we interpreted this as a spiraled INVITE, and thus
did not behave properly.
The fix for this is to not enter the loop detection or spiral logic in handle_request_invite
if the channel state is currently up. This way, no mid-call reinvites will be misinterpreted,
no matter what the nature of the reinvite may have been.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@272090 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This small changes prevents destroy_all_channels() from accessing a lock on an
unused dahdi_pri struct, resolving a ton of ERRORs that get spewed out when
shutting Asterisk down gracefully.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@272052 65c4cc65-6c06-0410-ace0-fbb531ad65f3
RFC 3261 section 9 states that a CANCEL has no effect on a
request to a UAS that has already given a final response. This
patch checks to make sure there is a pending invite before
allowing a CANCEL request to be processed, otherwise it responds
to the CANCEL with a "481 Call/Transaction Does Not Exist".
Review: https://reviewboard.asterisk.org/r/697/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@271977 65c4cc65-6c06-0410-ace0-fbb531ad65f3
https://origsvn.digium.com/svn/asterisk/branches/1.4
........
r271689 | mnicholson | 2010-06-22 07:52:27 -0500 (Tue, 22 Jun 2010) | 8 lines
Modify chan_sip's packet generation api to automatically calculate the Content-Length. This is done by storing packet content in a buffer until it is actually time to send the packet, at which time the size of the packet is calculated. This change was made to ensure that the Content-Length is always correct.
(closes issue #17326)
Reported by: kenner
Tested by: mnicholson, kenner
Review: https://reviewboard.asterisk.org/r/693/
........
This change also adds an ast_str_copy_string() function (similar to ast_copy_string), that copies one ast_str into another, properly handling embedded nulls.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@271690 65c4cc65-6c06-0410-ace0-fbb531ad65f3
According to RFC 3261 section 17.2.2, which describes non-INVITE server
transaction, when a dialog enters the Completed state it must destroy
the dialog after Timer J (T1*64) fires. For a BYE transaction Asterisk
terminates the dialog immediately during sip_hangup() when it should be
waiting T1*64 ms. This results in some odd behavior. For instance if
Asterisk receives a BYE and transmits a 200ok in response, if the endpoint
never receives the 200ok it will retransmit the BYE to which Asterisk
responds with a "481 Call leg/transaction does not exist" because the
dialog is already gone.
To resolve this I made a function called sip_scheddestroy_final(). This
differs slightly from sip_schedestroy() in that it enables a flag that
will prevent the destruction from ever being rescheduled or canceled
afterwards. It also prevents the pvt's needdestroy flag from being set
which triggers the destruction of the dialog within the do_monitor thread().
By using this function we are guaranteed destruction will not occur
until the scheduled time. This allows Asterisk to respond to any possible
retransmits for a dialog after we process the initial BYE request for T1*64 ms.
Other changes: I removed two instances where sip_cancel_destroy is used
right before calling sip_scheddestroy. sip_scheddestroy always calls
sip_cancel_destroy before scheduling the new destruction so it is completely
unnecessary.
Review: https://reviewboard.asterisk.org/r/694/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@271262 65c4cc65-6c06-0410-ace0-fbb531ad65f3
https://origsvn.digium.com/svn/asterisk/branches/1.4
........
r270866 | dvossel | 2010-06-16 12:35:29 -0500 (Wed, 16 Jun 2010) | 22 lines
fixes chan_iax2 race condition
There is code in chan_iax2.c that attempts to guarantee that only a single
active thread will handle a call number at a time. This code works once
the thread is added to an active_list of threads, but we are not currently
guaranteed that a newly activated thread will enter the active_list immediately
because it is left up to the thread to add itself after frames have been
queued to it. This means that if two frames come in for the same call number
at the same time, it is possible for them to grab two separate threads because
the first thread did not add itself to the active_list fast enough. This
causes some pretty complex problems.
This patch resolves this race condition by immediately adding an activated
thread to the active_list within the network thread and only depending on
the thread to remove itself once it is done processing the frames queued to
it. By doing this we are guaranteed that if another frame for the same call
number comes in at the same time, that this thread will immediately be found
in the active_list of threads.
Review: https://reviewboard.asterisk.org/r/720/
........
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@270867 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Clearing the callwaitcas flag in analog_call was causing the incoming D digit
to be ignored which triggers sending the caller ID.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@270836 65c4cc65-6c06-0410-ace0-fbb531ad65f3
chan_sip's "contactdeny" feature screens the "to be registered contact".
In case of nat=yes it should not use the address information from the
Contact header (which is not used at all for routing), but the source
IP address of the request.
Thus, if nat=yes and a client sends a request from a denied IP address
(e.g. by spoofing the src-IP address) it can bypass the screening.
This commit makes contactdeny apply to the src ip when nat=yes instead.
(closes issue #17276)
Reported by: klaus3000
Patches:
patch-asterisk-trunk-contactdeny.txt uploaded by klaus3000 (license 65)
Tested by: klaus3000
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@270658 65c4cc65-6c06-0410-ace0-fbb531ad65f3
The "channel" option would chop the channel name at the last '-', which made
it useless for something like a channel transfer from the dialplan. The
"fullchannel" option will return the channel name as-is.
ABE-2218
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@270260 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Calling dahdi_indicate() within dahdi_fixup() while the owner pointers are
in a potentially inconsistent state is a potentially bad thing in
principle.
However, calling dahdi_indicate() when the channel private lock is already
held can cause a deadlock if the PRI lock is needed because
dahdi_indicate() will also get the channel private lock. The pri_grab()
function assumes that the channel private lock is held once to avoid
deadlock.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@269307 65c4cc65-6c06-0410-ace0-fbb531ad65f3
* chan_dahdi supports dialing configuring and dialing by device file name.
DAHDI/span-name!local!1 will use /dev/dahdi/span-name/local/1 . Likewise
it may appear in chan_dahdi.conf as 'channel => span-name!local!1'.
* A new options for chan_dahdi.conf: 'ignore_failed_channels'. Boolean.
False by default. If set, chan_dahdi will ignore failed 'channel' entries.
Handy for the above name-based syntax as it does not depend on
initialization order.
* have my_pri_make_cc_dialstring() only manupulate dial-strings of group
(gGrR) dialing, which make it lsightly more complicated.
https://reviewboard.asterisk.org/r/535/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@269238 65c4cc65-6c06-0410-ace0-fbb531ad65f3
After 5 years in mantis and over a year on reviewboard, SRTP support is finally
being comitted. This includes generic CHANNEL dialplan functions that work for
getting the status of whether a call has secure media or signaling as defined
by the underlying channel technology and for setting whether or not a new
channel being bridged to a calling channel should have secure signaling or
media. See doc/tex/secure-calls.tex for examples.
Original patch by mikma, updated for trunk and revised by me.
(closes issue #5413)
Reported by: mikma
Tested by: twilson, notthematrix, hemanshurpatel
Review: https://reviewboard.asterisk.org/r/191/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@268894 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Extract the SS7 specific code out of chan_dahdi like what was done to
ISDN/PRI and analog signaling. The new SS7 structures were modeled on
sig_pri.
The changes to sig_pri are an enhancement and a bug fix made possible
because SS7 was extracted.
1) The sig_pri TRANSFERCAPABILITY channel variable should have been set
unconditionally in sig_pri_new_ast_channel().
2) SS7/PRI transfer capability interaction in dahdi_new() fixed because of
SS7 extraction.
3) Module ref count error in dahdi_new() if startpbx failed to start the
PBX for some reason.
Review: https://reviewboard.asterisk.org/r/661/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@268774 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Changes.
1. RFC 3261 states in section 17.1.2.2 and 17.1.1.2 that retransmission
timers should initially be set to timer T1. T1 by default is 500ms.
Asterisk was starting the retransmission timers at T1*2 which shouldn't
cause any problems, but is not RFC compliant.
2. RFC 3261 states in section 17.1.2.2 that for a non-INVITE client transaction,
if the retransmit timer fires while in the proceeding state that
the request must be retransmitted. Asterisk currently ack's
requests for both INVITE and non-INVITE transactions when a
1XX response is received, this patch changes this for non-INVITE requests.
3. The 'registerattempts' option in sip.conf is supposed to set
how many registry attempts will be made before giving up. When
this option is set to 1, I would expect only one registry attempt
to be made before stopping because of a failure, but instead two are
made. In my opinion this is not expected behavior. This option does
not indicate that these are re-attempts. The logic behind this option
has been changed to only attempt registers the exact number of times
this option is set to. If this option is 0, it still continues to
re-attempt the registration forever.
Review: https://reviewboard.asterisk.org/r/687/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@268205 65c4cc65-6c06-0410-ace0-fbb531ad65f3
The problem would manifest itself if your dialplan matching could accept
more digits to match than were actually dialed. The time out waiting for
overlap digits disconnected the call instead of matching any accumulated
digits to the dialplan.
Accidental conversion of a break out of loop as a break out of switch.
(closes issue #17401)
Reported by: avalentin
Patches:
issue17401_digit_timeout.patch uploaded by rmudgett (license 664)
Tested by: avalentin, rmudgett
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@267928 65c4cc65-6c06-0410-ace0-fbb531ad65f3
After some debugging, the random chan_h323 build failures appear to be due
to complications introduced by some chan_h323 specific build stuff getting
triggered during a clean. Simplify this by moving the h323 clean commands
down into channels/makefile.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@267352 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Add the ability to announce a call to an endpoint when there are no B
channels available. A call waiting call is a SETUP message with no B
channel selected.
Relevant specification: EN 300 056, EN 300 057, EN 300 058
For DAHDI/ISDN channels, the CHANNEL() dialplan function now supports the
"no_media_path" option.
* Returns "0" if there is a B channel associated with the call.
* Returns "1" if no B channel is associated with the call. The call is
either on hold or is a call waiting call.
If you are going to allow incoming call waiting calls then you need to use
CHANNEL(no_media_path) do determine if you must drop a call to accept the
new call.
Review: https://reviewboard.asterisk.org/r/568/
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@267261 65c4cc65-6c06-0410-ace0-fbb531ad65f3
Two struct sockaddr_ins are created when applying directmedia
host access rules. The addresses of these are passed to the RTP
engine to be filled in. However, the RTP engine inspects the fields
of the structs before actually taking action. This inspection caused
valgrind to be a bit unhappy.
git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@267097 65c4cc65-6c06-0410-ace0-fbb531ad65f3
