Use peerhangup flag variable to track whether a hangup has been
initiated by the peer or libpri itself (e.g. Layer 2 timeouts).
These changes fix a couple of problems with hangup events not being
handled properly in some situations:
- Call abort caused by incoming RESTART on a channel in use
- T309 timeout after L2 loss
- Improved hangup handling in libpri-side on_hangup() event handler
and state_advance() (FreeTDM side)
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
PROGRESS without PROGRESS INDICATOR IE is invalid according to Q.931,
so always call pri_progress() with info flag set, even if we do not
have media yet.
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
"Safer" version of ftdm_clamp(), that swaps min/max parameters if
vmin happens to be larger than vmax, making sure the output will
always satisfy vmin <= x <= vmax.
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
Some people, it seems, were having sleepless nights from the
log messages this feature produces every 15 minutes.
Default disable the feature, because i hate repeating myself
over and over, explaining them this is a non-issue and intended
behaviour and how to disable it in the config.
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
Channels can be in DOWN state but reserved by an incoming call
(FTDM_CHANNEL_INUSE flag). Additionally check the flag in on_timeout_t3xx()
and skip those channels.
Reported-by: privi #freetdm @ irc.freenode.net
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
Do not try to send RESTART on BRI PTMP spans, libpri will just print
an error and do nothing, causing the channels to be stuck in RESTART
forever.
Add T316 (RESTART ACK timeout), which is not implemented by libpri.
The default timeout is 30 seconds (Q.931 recommends 2 minutes, but
that is a little long and libpri layer 2 is being stupid) and the
restart attempt limit to 3 (instead of 2).
Periodically send RESTART on idle b-channels in PTP NT mode.
Default interval is 15 minutes, feature can be disabled by setting
"idle_restart_interval" to 0 in the span configuration.
Allow timeout / interval values to have an additional time unit
suffix for convenience reasons, the default (no unit specified)
is milliseconds, valid units include:
w (week), d (day), h (hour), m (minute), s (second)
Only full integers are accepted, no fractional numbers, valid examples:
2w = two weeks
37m = 37 minutes
1h = 1 hour
Combinations of multiple numbers and units (e.g. "1w5d") are not supported.
New span configuration parameters:
idle_restart_interval (milliseconds / time unit suffix /
0 = disabled)
t316 / restart_timeout (milliseconds / time unit suffix)
t316_limit / restart_attempts (number of max. attempts)
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
Add per b-channel private data (ftdm_libpri_b_data_t), move the q.931 libpri
call pointer into it and use channel->call_data to make it accessible.
Use ftdm_channel_wait() instead of select() to wait for I/O events on
the D-Channel and rewrite the timeout handling, adding support for custom
timers that can be added to the b-channel private data and started/stopped
with lpwrap_timer_start()/_stop().
The (sorted) list of active lpwrap timers is protected by the timer_mutex,
which is initialized in the now unified lpwrap_init_pri() (see below) and
destroyed in the new lpwrap_destroy_pri() function.
Unify lpwrap_init_pri() and lpwrap_init_bri(). Use the span type to select
BRI/PRI and Point-to-Point vs. Multipoint settings.
Use the new custom lpwrap timer API to implement T302 (aka overlap receiving
digit timeout). The default T302 timeout is 3000 miliseconds and can be set
via the "digit_timeout" or "t302" span configuration parameter
(0 = disable timer, valid range = 3000 ... 30000 miliseconds).
The lpwrap_pri_init() changes made it neccessarry to rewrite the span runtime
thread event loop. The d-channel initialization and libpri event handler
registration have been moved out of the event loop. The b-channel restart
handling in the event loop has been left as-is, pending a further clean up in
the future.
Rename on_info() to on_information() and add a new on_keypad_digit() event
handler to convert incoming KEYPAD DIGIT messages to channel DTMF events
(only useful on BRI PTMP NT-mode).
The channel restart handling has been improved to distinguish between locally
triggered and restarts caused by the remote end (= incoming RESTART messages).
Restarts on the D-Channel now use pri_restart(), causing a Q.921 reset on
the span.
Large patch, i'd have loved to split this into smaller pieces and actually
test some of the bits more toroughly (restart handling), but some people
are f*cking annoying, so here it is.
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
Move the B-channel message handling into a per-span I/O thread,
to solve most of the problems caused by the intermixed data + control
socket interface of mISDN, missing write poll() support on
mISDN B-channels and the FreeTDM I/O model. This eliminates most of
the audio problems (except for a few minor glitches).
A unix stream socket pair is used as a bi-directional pipe replacement
(the pipe code is still included in this commit, but will be removed later),
with the RX and TX buffer sizes carefully tuned to avoid excessive buffering
(= latency) and a deadlock situation between the write() call in ftdm_write()
and the code in misdn_span_run() that needs a minimum amount of data in the
TX buffer, before sending out a PH_DATA_REQ to the mISDN socket
(see misdn_span_run() comments for more details).
The minimum size for pipes is PAGE_SIZE (4k), which is ~500 ms worth of
audio. A socket pair RX/TX buffer size of 3k, seems to hold a maximum
amount of around 500 bytes data in practice, giving us a much lower
maximum latency than a unix pipe. (The socket pair might be replaced by a
custom ring buffer / fifo data structure to get even more fine grained
control of the maximum latency.)
The newly introduced span_start / span_stop callbacks in
ftdm_io_interface_t are used to start / stop the I/O thread. The callback
functions will wait up to 10 seconds for the thread to successfully
start up or shut down (using a mutex + condition var).
NOTE: Using any of the locking ftdm_span_() functions in the I/O will cause
a deadlock between the I/O thread (trying to lock span->mutex) and the
thread calling ftdm_start()/_stop() (holding the span->mutex).
(The I/O thread currently uses direct span member access to avoid this.)
The I/O thread uses the epoll(7) family of functions for event handling.
An epoll context is created on startup and all B-channel sockets are
registered (READ, PRI and ERR). Before entering the event loop,
the I/O thread will send a signal on the condition variable, to
indicate it has completed the startup procedure.
Incoming b-channel and command pipe events are handled by the event loop.
Payload of incoming PH_DATA_IND frames (= audio data) is sent to the
rx_audio_pipe_in end of the b-channel's socket pair and, if enough data is
available, a PH_DATA_REQ of the same size is sent to the b-channel mISDN socket
to transmit audio.
A MISDN_CMD_STOP command on the event pipe will wake up the I/O thread and
cause it to shut down. All b-channels will be unregistered from the epoll context
and the epoll fd closed. The I/O thread terminates itself after signalling the
successfull shutdown on the condition variable.
TODOs:
- Move D-Channel into I/O thread too
- Custom FIFO/ring buffer for data (even lower latency)
- Improve epoll() code (per-channel struct w/ callback, for epfd.data.ptr)
- Use mISDN DSP for audio (e.g. tone generator, dtmf detector, echo cancel)
- Use a per-port / span control socket to execute channel commands
synchronously, or add misdn_commands (queue?) that can be used that way
- Name I/O threads 'mISDN-%SPAN_NAME%', e.g. 'mISDN-M_BRI1'
(= add ftdm_thread_set_namef(thread, fmt, ...) / ftdm_thread_set_name(thread, name))
TL;DR: "tweak", solves "booboo" with audio
Signed-off-by: Stefan Knoblich <stkn@openisdn.net>
