normally indented in the code.
- rmudgett
On June 19, 2014, 6:04 p.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3593
it should only have
corresponding differences between v1.8 and v11 that were there before.
- rmudgett
On June 19, 2014, 2:29 p.m., rnewton wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3622/#review12244
---
Ship it!
Ship It!
- rmudgett
On June 19, 2014, 2:30 p.m
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3593/#review12245
---
Ship it!
Ship It!
- rmudgett
On June 19, 2014, 6:47 p.m
.
- rmudgett
On June 19, 2014, 10:36 a.m., opticron wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3649/
---
(Updated June 19
On June 16, 2014, 2:04 p.m., rmudgett wrote:
/trunk/channels/chan_dahdi.c, lines 1137-1139
https://reviewboard.asterisk.org/r/3548/diff/2/?file=59691#file59691line1137
You should not be directly accessing the .first and .last list members
directly. This is why I gave you the way
Recheck the default range I gave earlier. It is not 701-720.
Please double check that the stated option defaults are as I put in the
earlier review since I got them out of the code.
- rmudgett
On June 17, 2014, 6:46 p.m., rnewton wrote
.
- rmudgett
On June 16, 2014, 11:07 a.m., Tzafrir Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3548
/#comment22189
Example non-default parkinglot definition.
/branches/1.8/configs/features.conf.sample
https://reviewboard.asterisk.org/r/3621/#comment22188
The information in this note for parkext and parkpos should be put with the
descriptions instead of here.
- rmudgett
On June 13, 2014, 3
are able to go
out.
With the patch, HTTP is able to timeout connections that don't complete.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
with another
channel, the peeraccount codes need to be updated so they match what's accross
the bridge.
I could have replaced the ast_bridge_channel_update_accountcodes with a joining
and leaving version. However, that would change the API unnecessarily.
- rmudgett
. Start a new review for any
subsequent changes.
I'm also going to commit the corresponding changes for
https://reviewboard.asterisk.org/r/2170/ to the ss7_27_knk branch.
- rmudgett
On May 23, 2014, 4:27 a.m., KNK wrote
branch.
I'm going to commit the corresponding changes in
https://reviewboard.asterisk.org/r/2150/ to libss7 trunk.
Start new reviews for any subsequent changes.
- rmudgett
On Aug. 1, 2013, 11:41 a.m., KNK wrote:
---
This is an automatically
/#comment22101
You can actually put this integrity check back into the common code without
any RTTI conditions. The exception for the hash container is no longer true
because of the -r381427 fix by seanbright.
- rmudgett
On June 6, 2014, 11:16 a.m., George Joseph wrote
to the buffer, and grow + add until
we've read a full line.
Might I ask why menuselect would be run on an embedded system and not by the
development machine that compiles Asterisk?
- rmudgett
---
This is an automatically generated e-mail
than two
parties.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo
to determine that the old include files were no longer checked
on subsequent reload attempts.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE
the bridge system.
- rmudgett
On June 4, 2014, 11:43 a.m., Matt Jordan wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3585
/channel.h
https://reviewboard.asterisk.org/r/3585/#comment22053
12.4.0
/branches/12/main/pbx.c
https://reviewboard.asterisk.org/r/3585/#comment22052
Maybe a comment saying we are defensively clearing an escaped UNBRIDGE flag
would be nice.
- rmudgett
On June 5, 2014, 4:27 p.m., Matt
to be updated to include the
new astobj2 files.
- rmudgett
On June 4, 2014, 8:33 p.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3576
ast_check_hangup_locked().
- rmudgett
On June 1, 2014, 4:16 p.m., Matt Jordan wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3568
/astobj2_container.c
https://reviewboard.asterisk.org/r/3576/#comment22061
Move the endif condition to after ast_register_atexit() as it has nothing
to do outside of AST_DEVMODE.
- rmudgett
On June 5, 2014, 5:38 p.m., George Joseph wrote
.
- rmudgett
On June 5, 2014, 6:43 p.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3576/
---
(Updated
://reviewboard.asterisk.org/r/3584/#comment21983
This test is always true. You return at the beginning if neither is valid.
branches/12/channels/chan_pjsip.c
https://reviewboard.asterisk.org/r/3584/#comment21986
These could be grouped with the other connected struct initialization.
- rmudgett
On June 4
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3584/#review12045
---
Ship it!
Ship It!
- rmudgett
On June 4, 2014, 11:06 a.m
On June 4, 2014, 4:47 p.m., rmudgett wrote:
This split is what I had in mind except for the unnecessary pushing of ao2
primitive private declarations into astobj2_private.h.
- rmudgett
---
This is an automatically generated e-mail
the #close. However, this
might cause the release scripts to not recognize the issue is closed on those
branchs.
I'd say this change isn't getting us anything now that the additional #close's
are accepted and the other #comment and friends are not useful to the project.
- rmudgett
something.inc
something.inc:
#include another.inc
CLI !touch another.inc
CLI sip reload
I just tested that scenario to be sure and it does work as well.
- rmudgett
---
This is an automatically generated e-mail. To reply, visit
[] didn't get ast_config_AST_CONFIG_DIR
prepended if the include didn't have a '/' as the first character. As a result
the glob would not expand any wildcards unless the include explicitly had the
full path.
- rmudgett
files were no longer checked
on subsequent reload attempts.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http
specific code
5) astobj2 miscellaneous code (unit test, CLI debug commands,
initialization)
Reference https://reviewboard.asterisk.org/r/2110/ for mmichelson's comment
about breaking astobj2.c up. I've covered most of the contents of the exchange
here.
- rmudgett
On May 29, 2014, 7
On May 30, 2014, 3:38 p.m., rmudgett wrote:
branches/12/main/astobj2_private.h, lines 156-159
https://reviewboard.asterisk.org/r/3576/diff/1/?file=59035#file59035line156
There is too much in this file that should not be here.
1) Almost everything from this enum
of the glob statement?
If so, a comment at least indicating that would be nice.
rmudgett wrote:
Yes it is intentional and the do/while loop is actually in the body of
the for loop within the else clause. It has been this way from before
revision -r6000. We're talking about ancient
that the old include files were no longer checked
on subsequent reload attempts.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update
returns to the dialplan with AGENT_STATUS set to NOT_CONNECTED.
* Acks the call correctly. Before the conversation bridge would be leaked and
remain in the stasis cache for bridge show all to display. Now it doesn't.
Thanks,
rmudgett
Diff: https://reviewboard.asterisk.org/r/3561/diff/
Testing
---
Tests still pass with and without the patch for
https://reviewboard.asterisk.org/r/3551/ applied.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided
/
Testing
---
With the patch the video RTP ports are no longer leaked.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update
is equivalent to AST_CAUSE_NORMAL_CLEARING.
That is already documented there.
* \param cause Cause of bridge being destroyed. (If cause = 0 then use
AST_CAUSE_NORMAL_CLEARING)
- rmudgett
---
This is an automatically generated e-mail. To reply
Put this in its own wrapper function like all the others and test if the
function pointer is NULL before calling.
- rmudgett
On May 19, 2014, 9:32 a.m., Tzafrir Cohen wrote:
---
This is an automatically generated e-mail. To reply
/branches/12/res/res_pjsip_session.c 414555
/branches/12/include/asterisk/res_pjsip_session.h 414555
Diff: https://reviewboard.asterisk.org/r/3571/diff/
Testing
---
With the patch the video RTP ports are no longer leaked.
Thanks,
rmudgett
that the waitmerked user would not be suspended when the
marked user leaves the conference.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE
scripts. Two,
the JIRA issue alaready has that information. Three, you get the comment
repeated for every merge.
- rmudgett
On May 22, 2014, 5:36 a.m., wdoekes wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
/agent_acknowledge/agent_acknowledge_error/configs/ast1/agents.conf
5053
Diff: https://reviewboard.asterisk.org/r/3561/diff/
Testing
---
Tests still pass with and without the patch for
https://reviewboard.asterisk.org/r/3551/ applied.
Thanks,
rmudgett
the patch for
https://reviewboard.asterisk.org/r/3551/ applied.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit
the conversation bridge would be leaked and
remain in the stasis cache for bridge show all to display. Now it doesn't.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing
with AGENT_STATUS set to NOT_CONNECTED.
* Acks the call correctly. Before the conversation bridge would be leaked and
remain in the stasis cache for bridge show all to display. Now it doesn't.
Thanks,
rmudgett
--
_
-- Bandwidth
,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3534/#review11909
---
Ship it!
Ship It!
- rmudgett
On May 13, 2014, 6:11 a.m
/#comment21791
This comment should be deleted as it will no longer do this with the patch.
- rmudgett
On May 16, 2014, 8:56 a.m., Olle E Johansson wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
== unbridge || flags == (both)
- rmudgett
On May 14, 2014, 9:14 p.m., Matt Jordan wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3535
/r/3543/diff/
Testing
---
Added debugging statements to indicate when the MOH would have been suspended
on the waitmarked user.
Got confirmation that the waitmerked user would not be suspended when the
marked user leaves the conference.
Thanks,
rmudgett
://reviewboard.asterisk.org/r/3535/#comment21768
Soft hangup flags are bit flags. You need to be using the operator to
test not the == operator.
if (flags (ASYNCGOTO | UNBRIDGE))
at least one of the flags is set
- rmudgett
On May 13, 2014, 7:11 a.m., Matt Jordan wrote
.
Is there is any way the detection of match_more / need_dialtone can be
put into small functions, that would be... nice.
match_more = can_pri_match_more(e, pri);
need_dialtone = do_i_need_dialtone(match_more, e, pri);
Something like that.
rmudgett wrote
for more digits.
Observed that the inband audio available progress indication ie was sent when
needed on the SETUP_ACKNOWLEDGE message.
Added a debug message that indicated when Asterisk did and did not receive the
inband audio indication for the SETUP_ACKNOWLEDGE message.
Thanks,
rmudgett
on the SETUP_ACKNOWLEDGE message.
Added a debug message that indicated when Asterisk did and did not receive the
inband audio indication for the SETUP_ACKNOWLEDGE message.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3522/#review11846
---
Ship it!
Ship It!
- rmudgett
On May 8, 2014, 9:59 a.m
that is the main problem?
- rmudgett
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3521/#review11837
---
On May 1, 2014, 9:35 p.m
another function for this change.
- rmudgett
On May 7, 2014, 5:10 p.m., Joshua Colp wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3522
: 'SIPREFERTOHDR'
+Variable: '_{0,2}SIPREFERTOHDR'
Value: 'sip:charlie@127.0.0.1'
count: 2
-
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http
is not
safe. You would need to unlink the old node and link the new node in its place.
You've done this swap in several places.
- rmudgett
On April 30, 2014, 5:13 p.m., George Joseph wrote:
---
This is an automatically generated e-mail
a debug message that indicated when Asterisk did and did not receive the
inband audio indication for the SETUP_ACKNOWLEDGE message.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com
for the SETUP_ACKNOWLEDGE message.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo
-
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
On April 25, 2014, 1:03 p.m., rmudgett wrote:
/branches/11/channels/chan_sip.c, lines 21287-21295
https://reviewboard.asterisk.org/r/3474/diff/3/?file=57909#file57909line21287
These are supposed to be AST_TRANSPORT_xxx declarations.
SIP_TRANSPORT_xxx declarations don't exist
() that returns the transport string.
/branches/11/channels/chan_sip.c
https://reviewboard.asterisk.org/r/3474/#comment21508
Add curlies.
/branches/11/channels/chan_sip.c
https://reviewboard.asterisk.org/r/3474/#comment21507
red blob
- rmudgett
On April 24, 2014, 12:33 p.m., Patrick
/#comment21523
Merge marker left in.
/branches/12/res/parking/parking_manager.c
https://reviewboard.asterisk.org/r/3446/#comment21525
Useless return.
- rmudgett
On April 24, 2014, 11:37 a.m., Jonathan Rose wrote
://reviewboard.asterisk.org/r/3446/#comment21527
Moving this to where you test bridge_channel for NULL only leaves two
places where bridge_channel needs to be cleaned up instead of the current five.
RAII_VAR usage could then be easily eliminated.
- rmudgett
On April 24, 2014, 11:37 a.m., Jonathan
these prototypes have is to ensure that the various
implementations above provide the same API.
- rmudgett
On April 19, 2014, 11:57 a.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
://reviewboard.asterisk.org/r/3405/#comment21488
Why did this get moved?
- rmudgett
On April 22, 2014, 1:35 p.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3405
On April 22, 2014, 11:58 a.m., rmudgett wrote:
branches/12/include/asterisk/spinlock.h, lines 403-405
https://reviewboard.asterisk.org/r/3405/diff/5/?file=57747#file57747line403
This statement isn't quite true since these prototypes are declared
after the functions themselves
/r/3451/diff/
Testing
---
Dialed a number that redirected the call to a new number. Call still went were
expected.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk
that is busy and got the
AST_CONTROL_BUSY code. (chan_dahdi ast_request() does not create the channel
like other channel drivers in this case)
Did an AMI originate to a DAHDI group that has no members available and got the
AST_CONTROL_CONGESTED code.
Thanks,
rmudgett
on start up.
Saw that the TE-PTP side responded with the permanently assigned TEI 0.
Saw that the NT-PTMP side handled the unknown TEI by removing it. (TEI 0 is
unknown by that side.)
Thanks,
rmudgett
--
_
-- Bandwidth
are to ensure that the
spinlock implementations above provice the same API.
- rmudgett
On April 17, 2014, 12:26 p.m., George Joseph wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3405
with the priv_data.cached_in container.
Unreffing the object until it drops to one ref while it is cached is ok.
Manually unlinking the cached object from the cache container has an
unbalanced unref to the container.
- rmudgett
On April 17, 2014, 6:09 p.m., Corey Farrell wrote
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3440/#review11610
---
Ship it!
Ship It!
- rmudgett
On April 15, 2014, 11:57 a.m
was unsuccessful in reproducing the testsuite channel staging assertion
failure.
However, SIP calls can still setup and teardown with the patch installed.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api
that the TE-PTP side responded with the permanently assigned TEI 0.
Saw that the NT-PTMP side handled the unknown TEI by removing it. (TEI 0 is
unknown by that side.)
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http
.
- rmudgett
On April 14, 2014, 5:08 p.m., Jonathan Rose wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3446
/12/main/stasis_channels.c 412438
/branches/12/apps/app_queue.c 412438
/branches/12/apps/app_dial.c 412438
Diff: https://reviewboard.asterisk.org/r/3451/diff/
Testing
---
Dialed a number that redirected the call to a new number. Call still went were
expected.
Thanks,
rmudgett
://reviewboard.asterisk.org/r/3440/#comment21338
I think you should remove \brief from all of these lines for the same
reason.
- rmudgett
On April 11, 2014, 11:33 a.m., Jonathan Rose wrote:
---
This is an automatically generated e
On April 10, 2014, 6:20 p.m., rmudgett wrote:
I'm not seeing any protection from loss of frames when the channels
optimize out. Losing media frames isn't nice but is tollerable. Losing
control frames is unacceptable.
Joshua Colp wrote:
1. Can you explain the scenario and how I
by removing it. (TEI 0 is
unknown by that side.)
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com
/bridge.c
https://reviewboard.asterisk.org/r/3415/#comment21327
Though locking is not strictly necessary here, for completeness, lock the
bridge before calling here to satisfy the pre condition.
- rmudgett
On April 4, 2014, 9:31 a.m., opticron wrote
the channels optimize
out. Losing media frames isn't nice but is tollerable. Losing control frames
is unacceptable.
- rmudgett
On April 9, 2014, 2:49 p.m., Joshua Colp wrote:
---
This is an automatically generated e-mail. To reply, visit
installed.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
. In
sip_hangup() there cannot be a bridge peer anymore. However, the stats were
already set anyway if the channel was in a bridge because the BYE processing
sets the stats on the channel and its peer.
- rmudgett
On April 8, 2014, 8:45 a.m., wdoekes wrote:
Bugs: ASTERISK-22846
Is that related?
It is what started the series of events.
- rmudgett
---
This is an automatically generated e-mail. To reply, visit:
https
, the messages are output to
stderr.
Diffs
-
/branches/1.8/main/asterisk.c 411882
Diff: https://reviewboard.asterisk.org/r/3423/diff/
Testing
---
Messages are displayed.
Thanks,
rmudgett
--
_
-- Bandwidth
/
Testing
---
Messages are displayed.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com
/12/main/utils.c 411651
/branches/12/main/stasis_cache.c 411651
/branches/12/main/channel.c 411651
Diff: https://reviewboard.asterisk.org/r/3411/diff/
Testing
---
Inserted an assertion that failed to see that the backtrace is generated.
Thanks,
rmudgett
for
this problem showed there was almost no lock contention when grabbing the 7
locks needed for optimization.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
looking for why the masquerade test fails and I have
seen just about all the failure cases reported.
The core show locks code works when I uncomment and comment the right lines.
It showed that the problem was not a deadlock on Asterisk v12.
Thanks,
rmudgett
an AMI originate to a DAHDI group that has no members available and got the
AST_CONTROL_CONGESTED code.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
build agent.
v1.8 takes 21-22 seconds to run on the 64 bit build agent.
As a side note, the counting instrumentation code I added to v12 to look for
this problem showed there was almost no lock contention when grabbing the 7
locks needed for optimization.
Thanks,
rmudgett
between internal timing and loop timing modes. We are always one or the other
when we are generating frames.
- rmudgett
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3414/#review11480
)
- rmudgett
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3412/#review11482
---
On April 1, 2014, 7:09 p.m., rmudgett wrote
on Asterisk v12.
Thanks,
rmudgett
--
_
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
asterisk-dev mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk
() requires the involved bridges to be
locked on entry.
- rmudgett
On April 3, 2014, 2:13 p.m., opticron wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3415
://reviewboard.asterisk.org/r/3362/#comment21229
ran
/trunk/funcs/func_periodic_hook.c
https://reviewboard.asterisk.org/r/3362/#comment21227
ast_strdupa() is not NULL tollerant.
Doing
ast_strdupa(S_OR(data, ))
will then cause the interval message to go out.
- rmudgett
On April 3
601 - 700 of 861 matches
Mail list logo
