Koozali.org: home of the SME Server

Dahdi calls scratchy, will upgrade help, and how to upgrade

Offline jufra

  • *
  • 34
  • +0/-0
Dahdi calls scratchy, will upgrade help, and how to upgrade
« on: March 06, 2013, 06:16:17 AM »
Hi,

we're running an SME server on SME 8 beta7 with a Digium Dahdi card in it (pci:0000:01:0c.0 wctdm24xxp+  d161:8005 Wildcard TDM410P)
Following versions of Dahdi/Asterisk/Sail
Sail version is 3.1.0-128
DAHDI Version: 2.3.0.1 Echo Canceller: MG2
Asterisk 1.8.5.0
We have tried to upgrade the box to SME 8 to no avail as the Dahdi card wouldn't work with it anymore.

The issue we are having is that the calls on the Dahdi lines have a lot of quality issues, they are very scratchy (echo is ok) and dropouts happen.
We contacted the Digium support and they suggested upgrading the Dahdi and Asterisk to the latest versions.
My question is how can we do this on SME 8 b7, I suppoer we are limited to what is in the repos?

Any clues as to how to fix this, also, another question, is it to do with the versions or could there be other reasons?

I can supply any files if needed, so please let me know

Many thanks

Frank

Offline chris burnat

  • ****
  • 1,135
  • +2/-0
    • http://www.burnat.com
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #1 on: March 06, 2013, 10:48:50 AM »
Moving to 7.x VoIP (Asterisk, SAIL etc)
- chris
If it does not work out of the box, please fill in a Bug Report @ Bugzilla (http://bugs.contribs.org)  - check: http://wiki.contribs.org/Bugzilla_Help .  Thanks.

Offline Franco

  • *
  • 1,171
  • +0/-0
    • http://contribs.org
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #2 on: March 07, 2013, 05:58:21 PM »
Hi,
First you need to find out the reason for the bad quality, which could be the lines coming in. Attach a regular phone to the line and test it to see the quality improves.
Also, DSL filters are known to pickup radio frequencies, so lookup if you have one.

If everything is OK, then you'll have to tune the cards to your system.

Best,

Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #3 on: March 08, 2013, 01:10:43 PM »
Hi Franco,

no, it's not the line itself, that's fine, we tested that using the method you suggested. It was the first thing we did. Also, no DSL filters etc..., it is a pure, good old fashioned POTS line which works well with a normal phone on it.
We left all the settings pretty much on default, the only thing I did was increase the taps on the EC, since the echo was quite significant on the default 128 taps. 256 seems ok though.
I can post the Dhsi conf file if that helps?

Cheers
Frank

Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #4 on: March 08, 2013, 01:13:44 PM »
1st part
Code: [Select]
;
; DAHDI Telephony Configuration file
;
; You need to restart Asterisk to re-configure the DAHDI channel
; CLI> module reload chan_dahdi.so
;      will reload the configuration file, but not all configuration options
;      are re-configured during a reload (signalling, as well as PRI and
;      SS7-related settings cannot be changed on a reload).
;
; This file documents many configuration variables.  Normally unless you know
; what a variable means or that it should be changed, there's no reason to
; un-comment those lines.
;
; Examples below that are commented out (those lines that begin with a ';' but
; no space afterwards) typically show a value that is not the default value,
; but would make sense under certain circumstances. The default values are
; usually sane. Thus you should typically not touch them unless you know what
; they mean or you know you should change them.

[trunkgroups]
;
; Trunk groups are used for NFAS connections.
;
; Group: Defines a trunk group.
;        trunkgroup => <trunkgroup>,<dchannel>[,<backup1>...]
;
;        trunkgroup  is the numerical trunk group to create
;        dchannel    is the DAHDI channel which will have the
;                    d-channel for the trunk.
;        backup1     is an optional list of backup d-channels.
;
;trunkgroup => 1,24,48
;trunkgroup => 1,24
;
; Spanmap: Associates a span with a trunk group
;        spanmap => <dahdispan>,<trunkgroup>[,<logicalspan>]
;
;        dahdispan   is the DAHDI span number to associate
;        trunkgroup  is the trunkgroup (specified above) for the mapping
;        logicalspan is the logical span number within the trunk group to use.
;                    if unspecified, no logical span number is used.
;
;spanmap => 1,1,1
;spanmap => 2,1,2
;spanmap => 3,1,3
;spanmap => 4,1,4

[channels]
;
; Default language
;
;language=en
;
; Context for calls. Defaults to 'default'
;
;context=incoming
;
; Switchtype:  Only used for PRI.
;
; national:    National ISDN 2 (default)
; dms100:      Nortel DMS100
; 4ess:        AT&T 4ESS
; 5ess:        Lucent 5ESS
; euroisdn:    EuroISDN (common in Europe)
; ni1:         Old National ISDN 1
; qsig:        Q.SIG
;
;switchtype=euroisdn
;
; MSNs for ISDN spans.  Asterisk will listen for the listed numbers on
; incoming calls and ignore any calls not listed.
; Here you can give a comma separated list of numbers or dialplan extension
; patterns.  An empty list disables MSN matching to allow any incoming call.
; Only set on PTMP CPE side of ISDN span if needed.
; The default is an empty list.
;msn=
;
; Some switches (AT&T especially) require network specific facility IE.
; Supported values are currently 'none', 'sdn', 'megacom', 'tollfreemegacom', 'accunet'
;
; nsf cannot be changed on a reload.
;
;nsf=none
;
;service_message_support=yes
; Enable service message support for channel. Must be set after switchtype.
;
; Dialing options for ISDN (i.e., Dial(DAHDI/g1/exten/options)):
; R      Reverse Charge Indication
;          Indicate to the called party that the call will be reverse charged.
; K(n)   Keypad digits n
;          Send out the specified digits as keypad digits.
;
; PRI Dialplan: The ISDN-level Type Of Number (TON) or numbering plan, used for
; the dialed number.  For most installations, leaving this as 'unknown' (the
; default) works in the most cases.  In some very unusual circumstances, you
; may need to set this to 'dynamic' or 'redundant'.  Note that if you set one
; of the others, you will be unable to dial another class of numbers.  For
; example, if you set 'national', you will be unable to dial local or
; international numbers.
;
; PRI Local Dialplan:  Only RARELY used for PRI (sets the calling number's
; numbering plan).  In North America, the typical use is sending the 10 digit
; callerID number and setting the prilocaldialplan to 'national' (the default).
; Only VERY rarely will you need to change this.
;
; Neither pridialplan nor prilocaldialplan can be changed on reload.
;
; unknown:        Unknown
; private:        Private ISDN
; local:          Local ISDN
; national:       National ISDN
; international:  International ISDN
; dynamic:        Dynamically selects the appropriate dialplan
; redundant:      Same as dynamic, except that the underlying number is not
;                 changed (not common)
;
;pridialplan=unknown
;prilocaldialplan=national
;
; pridialplan may be also set at dialtime, by prefixing the dialled number with
; one of the following letters:
; U - Unknown
; I - International
; N - National
; L - Local (Net Specific)
; S - Subscriber
; V - Abbreviated
; R - Reserved (should probably never be used but is included for completeness)
;
; Additionally, you may also set the following NPI bits (also by prefixing the
; dialled string with one of the following letters):
; u - Unknown
; e - E.163/E.164 (ISDN/telephony)
; x - X.121 (Data)
; f - F.69 (Telex)
; n - National
; p - Private
; r - Reserved (should probably never be used but is included for completeness)
;
; You may also set the prilocaldialplan in the same way, but by prefixing the
; Caller*ID Number, rather than the dialled number.  Please note that telcos
; which require this kind of additional manipulation of the TON/NPI are *rare*.
; Most telco PRIs will work fine simply by setting pridialplan to unknown or
; dynamic.
;
;
; PRI caller ID prefixes based on the given TON/NPI (dialplan)
; This is especially needed for EuroISDN E1-PRIs
;
; None of the prefix settings can be changed on reload.
;
; sample 1 for Germany
;internationalprefix = 00
;nationalprefix = 0
;localprefix = 0711
;privateprefix = 07115678
;unknownprefix =
;
; sample 2 for Germany
;internationalprefix = +
;nationalprefix = +49
;localprefix = +49711
;privateprefix = +497115678
;unknownprefix =
;
; PRI resetinterval: sets the time in seconds between restart of unused
; B channels; defaults to 'never'.
;
;resetinterval = 3600
;
; Overlap dialing mode (sending overlap digits)
; Cannot be changed on a reload.
;
; incoming: incoming direction only
; outgoing: outgoing direction only
; no: neither direction
; yes or both: both directions
;
;overlapdial=yes
;
; Allow inband audio (progress) when a call is DISCONNECTed by the far end of a PRI
;
;inbanddisconnect=yes
;
; Allow a held call to be transferred to the active call on disconnect.
; This is useful on BRI PTMP NT lines where an ISDN phone can simulate the
; transfer feature of an analog phone.
; The default is no.
;hold_disconnect_transfer=yes
;
; PRI Out of band indications.
; Enable this to report Busy and Congestion on a PRI using out-of-band
; notification. Inband indication, as used by Asterisk doesn't seem to work
; with all telcos.
;
; outofband:      Signal Busy/Congestion out of band with RELEASE/DISCONNECT
; inband:         Signal Busy/Congestion using in-band tones (default)
;
; priindication cannot be changed on a reload.
;
;priindication = outofband
;
; If you need to override the existing channels selection routine and force all
; PRI channels to be marked as exclusively selected, set this to yes.
;
; priexclusive cannot be changed on a reload.
;
;priexclusive = yes
;
;
; If you need to use the logical channel mapping with your Q.SIG PRI instead
; of the physical mapping you must use the qsigchannelmapping option.
;
; logical:  Use the logical channel mapping
; physical: Use physical channel mapping (default)
;
;qsigchannelmapping=logical
;
; If you wish to ignore remote hold indications (and use MOH that is supplied over
; the B channel) enable this option.
;
;discardremoteholdretrieval=yes
;
; ISDN Timers
; All of the ISDN timers and counters that are used are configurable.  Specify
; the timer name, and its value (in ms for timers).
; K:    Layer 2 max number of outstanding unacknowledged I frames (default 7)
; N200: Layer 2 max number of retransmissions of a frame (default 3)
; T200: Layer 2 max time before retransmission of a frame (default 1000 ms)
; T203: Layer 2 max time without frames being exchanged (default 10000 ms)
; T305: Wait for DISCONNECT acknowledge (default 30000 ms)
; T308: Wait for RELEASE acknowledge (default 4000 ms)
; T309: Maintain active calls on Layer 2 disconnection (default 6000 ms)
;       EuroISDN: 6000 to 12000 ms, according to (N200 + 1) x T200 + 2s
;       May vary in other ISDN standards (Q.931 1993 : 90000 ms)
; T313: Wait for CONNECT acknowledge, CPE side only (default 3000 ms)
;
; T-RESPONSE:   Maximum time to wait for a typical APDU response. (default 4000 ms)
;               This is an implementation timer when the standard does not specify one.
; T-ACTIVATE:   Request supervision timeout. (default 10000 ms)
; T-RETENTION:  Maximum  time to wait for user A to activate call-completion. (default 30000 ms)
;               Used by ETSI PTP, ETSI PTMP, and Q.SIG as the cc_offer_timer.
; T-CCBS1:      T-STATUS timer equivalent for CC user A status. (default 4000 ms)
; T-CCBS2:      Maximum  time the CCBS service will be active (default 45 min in ms)
; T-CCBS3:      Maximum  time to wait for user A to respond to user B availability. (default 20000 ms)
; T-CCBS5:      Network B CCBS supervision timeout. (default 60 min in ms)
; T-CCBS6:      Network A CCBS supervision timeout. (default 60 min in ms)
; T-CCNR2:      Maximum  time the CCNR service will be active (default 180 min in ms)
; T-CCNR5:      Network B CCNR supervision timeout. (default 195 min in ms)
; T-CCNR6:      Network A CCNR supervision timeout. (default 195 min in ms)
; CC-T1:        Q.SIG CC request supervision timeout. (default 30000 ms)
; CCBS-T2:      Q.SIG CCBS supervision timeout. (default 60 min in ms)
; CCNR-T2:      Q.SIG CCNR supervision timeout. (default 195 min in ms)
; CC-T3:        Q.SIG CC Maximum time to wait for user A to respond to user B availability. (default 30000 ms)
;
;pritimer => t200,1000
;pritimer => t313,4000
;
; CC PTMP recall mode:
; specific - Only the CC original party A can participate in the CC callback
; global - Other compatible endpoints on the PTMP line can be party A in the CC callback
;
; cc_ptmp_recall_mode cannot be changed on a reload.
;
;cc_ptmp_recall_mode = specific
;
; CC Q.SIG Party A (requester) retain signaling link option
; retain       Require that the signaling link be retained.
; release      Request that the signaling link be released.
; do_not_care  The responder is free to choose if the signaling link will be retained.
;
;cc_qsig_signaling_link_req = retain
;
; CC Q.SIG Party B (responder) retain signaling link option
; retain       Prefer that the signaling link be retained.
; release      Prefer that the signaling link be released.
;
;cc_qsig_signaling_link_rsp = retain
;
; See ccss.conf.sample for more options.  The timers described by ccss.conf.sample
; are not used by ISDN for the native protocol since they are defined by the
; standards and set by pritimer above.
;
; To enable transmission of facility-based ISDN supplementary services (such
; as caller name from CPE over facility), enable this option.
; Cannot be changed on a reload.
;
;facilityenable = yes
;

; This option enables Advice of Charge pass-through between the ISDN PRI and
; Asterisk.  This option can be set to any combination of 's', 'd', and 'e' which
; represent the different variants of Advice of Charge, AOC-S, AOC-D, and AOC-E.
; Advice of Charge pass-through is currently only supported for ETSI.  Since most
; AOC messages are sent on facility messages, the 'facilityenable' option must
; also be enabled to fully support AOC pass-through.
;
;aoc_enable=s,d,e
;
; When this option is enabled, a hangup initiated by the ISDN PRI side of the
; asterisk channel will result in the channel delaying its hangup in an
; attempt to receive the final AOC-E message from its bridge.  The delay
; period is configured as one half the T305 timer length. If the channel
; is not bridged the hangup will occur immediatly without delay.
;
;aoce_delayhangup=yes

; pritimer cannot be changed on a reload.
;
; Signalling method. The default is "auto". Valid values:
; auto:           Use the current value from DAHDI.
; em:             E & M
; em_e1:          E & M E1
; em_w:           E & M Wink
; featd:          Feature Group D (The fake, Adtran style, DTMF)
; featdmf:        Feature Group D (The real thing, MF (domestic, US))
; featdmf_ta:     Feature Group D (The real thing, MF (domestic, US)) through
;                 a Tandem Access point
; featb:          Feature Group B (MF (domestic, US))
; fgccama:        Feature Group C-CAMA (DP DNIS, MF ANI)
; fgccamamf:      Feature Group C-CAMA MF (MF DNIS, MF ANI)
; fxs_ls:         FXS (Loop Start)
; fxs_gs:         FXS (Ground Start)
; fxs_ks:         FXS (Kewl Start)
; fxo_ls:         FXO (Loop Start)
; fxo_gs:         FXO (Ground Start)
; fxo_ks:         FXO (Kewl Start)
; pri_cpe:        PRI signalling, CPE side
; pri_net:        PRI signalling, Network side
; bri_cpe:        BRI PTP signalling, CPE side
; bri_net:        BRI PTP signalling, Network side
; bri_cpe_ptmp:   BRI PTMP signalling, CPE side
; bri_net_ptmp:   BRI PTMP signalling, Network side
; sf:             SF (Inband Tone) Signalling
; sf_w:           SF Wink
; sf_featd:       SF Feature Group D (The fake, Adtran style, DTMF)
; sf_featdmf:     SF Feature Group D (The real thing, MF (domestic, US))
; sf_featb:       SF Feature Group B (MF (domestic, US))
; e911:           E911 (MF) style signalling
; ss7:            Signalling System 7
; mfcr2:          MFC/R2 Signalling. To specify the country variant see 'mfcr2_variant'
;
; The following are used for Radio interfaces:
; fxs_rx:         Receive audio/COR on an FXS kewlstart interface (FXO at the
;                 channel bank)
; fxs_tx:         Transmit audio/PTT on an FXS loopstart interface (FXO at the
;                 channel bank)
; fxo_rx:         Receive audio/COR on an FXO loopstart interface (FXS at the
;                 channel bank)
; fxo_tx:         Transmit audio/PTT on an FXO groundstart interface (FXS at
;                 the channel bank)
; em_rx:          Receive audio/COR on an E&M interface (1-way)
; em_tx:          Transmit audio/PTT on an E&M interface (1-way)
; em_txrx:        Receive audio/COR AND Transmit audio/PTT on an E&M interface
;                 (2-way)
; em_rxtx:        Same as em_txrx (for our dyslexic friends)
; sf_rx:          Receive audio/COR on an SF interface (1-way)
; sf_tx:          Transmit audio/PTT on an SF interface (1-way)
; sf_txrx:        Receive audio/COR AND Transmit audio/PTT on an SF interface
;                 (2-way)
; sf_rxtx:        Same as sf_txrx (for our dyslexic friends)
; ss7:            Signalling System 7
;
; signalling of a channel can not be changed on a reload.
;
;signalling=fxo_ls
;
; If you have an outbound signalling format that is different from format
; specified above (but compatible), you can specify outbound signalling format,
; (see below). The 'signalling' format specified will be the inbound signalling
; format. If you only specify 'signalling', then it will be the format for
; both inbound and outbound.
;
; outsignalling can only be one of:
;   em, em_e1, em_w, sf, sf_w, sf_featd, sf_featdmf, sf_featb, featd,
;   featdmf, featdmf_ta, e911, fgccama, fgccamamf
;
; outsignalling cannot be changed on a reload.
;
;signalling=featdmf
;
;outsignalling=featb
;
; For Feature Group D Tandem access, to set the default CIC and OZZ use these
; parameters (Will not be updated on reload):
;
;defaultozz=0000
;defaultcic=303
;
; A variety of timing parameters can be specified as well
; The default values for those are "-1", which is to use the
; compile-time defaults of the DAHDI kernel modules. The timing
; parameters, (with the standard default from DAHDI):
;
;    prewink:     Pre-wink time (default 50ms)
;    preflash:    Pre-flash time (default 50ms)
;    wink:        Wink time (default 150ms)
;    flash:       Flash time (default 750ms)
;    start:       Start time (default 1500ms)
;    rxwink:      Receiver wink time (default 300ms)
;    rxflash:     Receiver flashtime (default 1250ms)
;    debounce:    Debounce timing (default 600ms)
;
; None of them will update on a reload.
;
; How long generated tones (DTMF and MF) will be played on the channel
; (in milliseconds).
;
; This is a global, rather than a per-channel setting. It will not be
; updated on a reload.
;
;toneduration=100
;
; Whether or not to do distinctive ring detection on FXO lines:
;
;usedistinctiveringdetection=yes
;
; enable dring detection after caller ID for those countries like Australia
; where the ring cadence is changed *after* the caller ID spill:
;
;distinctiveringaftercid=yes
;
; Whether or not to use caller ID:
;
usecallerid=no
;
; Type of caller ID signalling in use
;     bell     = bell202 as used in US (default)
;     v23      = v23 as used in the UK
;     v23_jp   = v23 as used in Japan
;     dtmf     = DTMF as used in Denmark, Sweden and Netherlands
;     smdi     = Use SMDI for caller ID.  Requires SMDI to be enabled (usesmdi).
;
;cidsignalling=v23
;
; What signals the start of caller ID
;     ring        = a ring signals the start (default)
;     polarity    = polarity reversal signals the start
;     polarity_IN = polarity reversal signals the start, for India,
;                   for dtmf dialtone detection; using DTMF.
;     (see https://wiki.asterisk.org/wiki/display/AST/Caller+ID+in+India)
;     dtmf        = causes monitor loop to look for dtmf energy on the
;                   incoming channel to initate cid acquisition
;
;cidstart=polarity
;
; When cidstart=dtmf, the energy level on the line used to trigger dtmf cid
; acquisition. This number is compared to the average over a packet of audio
; of the absolute values of 16 bit signed linear samples. The default is set
; to 256. The choice of 256 is arbitrary. The value you should select should
; be high enough to prevent false detections while low enough to insure that
; no dtmf spills are missed.
;
;dtmfcidlevel=256
;
; Whether or not to hide outgoing caller ID (Override with *67 or *82)
; (If your dialplan doesn't catch it)
;
;hidecallerid=yes
;
; Enable if you need to hide just the name and not the number for legacy PBX use.
; Only applies to PRI channels.
;hidecalleridname=yes
;
; On UK analog lines, the caller hanging up determines the end of calls.  So
; Asterisk hanging up the line may or may not end a call (DAHDI could just as
; easily be re-attaching to a prior incoming call that was not yet hung up).
; This option changes the hangup to wait for a dialtone on the line, before
; marking the line as once again available for use with outgoing calls.
;waitfordialtone=yes
;
; The following option enables receiving MWI on FXO lines.  The default
; value is no.
; The mwimonitor can take the following values
; no - No mwimonitoring occurs. (default)
; yes - The same as specifying fsk
; fsk - the FXO line is monitored for MWI FSK spills
; fsk,rpas - the FXO line is monitored for MWI FSK spills preceded
; by a ring pulse alert signal.
; neon - The fxo line is monitored for the presence of NEON pulses
; indicating MWI.
; When detected, an internal Asterisk MWI event is generated so that any other
; part of Asterisk that cares about MWI state changes is notified, just as if
; the state change came from app_voicemail.
; For FSK MWI Spills, the energy level that must be seen before starting the
; MWI detection process can be set with 'mwilevel'.
;

Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #5 on: March 08, 2013, 01:15:49 PM »
Part 2
Code: [Select]

;mwimonitor=no
;mwilevel=256
;
; This option is used in conjunction with mwimonitor.  This will get executed
; when incoming MWI state changes.  The script is passed 2 arguments.  The
; first is the corresponding mailbox, and the second is 1 or 0, indicating if
; there are messages waiting or not.
;
;mwimonitornotify=/usr/local/bin/dahdinotify.sh
;
; The following keyword 'mwisendtype' enables various VMWI methods on FXS lines (if supported).
; The default is to send FSK only.
; The following options are available;
; 'rpas' Ring Pulse Alert Signal, alerts intelligent phones that a FSK message is about to be sent.
; 'lrev' Line reversed to indicate messages waiting.
; 'hvdc' 90Vdc OnHook DC voltage to indicate messages waiting.
; 'hvac' or 'neon' 90Vac OnHook AC voltage to light Neon bulb.
; 'nofsk' Disables FSK MWI spills from being sent out.
; It is feasible that multiple options can be enabled.
;mwisendtype=rpas,lrev
;
; Whether or not to enable call waiting on internal extensions
; With this set to 'yes', busy extensions will hear the call-waiting
; tone, and can use hook-flash to switch between callers. The Dial()
; app will not return the "BUSY" result for extensions.
;
callwaiting=yes
;
; Configure the number of outstanding call waiting calls for internal ISDN
; endpoints before bouncing the calls as busy.  This option is equivalent to
; the callwaiting option for analog ports.
; A call waiting call is a SETUP message with no B channel selected.
; The default is zero to disable call waiting for ISDN endpoints.
;max_call_waiting_calls=0
;
; Allow incoming ISDN call waiting calls.
; A call waiting call is a SETUP message with no B channel selected.
;allow_call_waiting_calls=no
;
; Configure the ISDN span to indicate MWI for the list of mailboxes.
; You can give a comma separated list of up to 8 mailboxes per span.
; An empty list disables MWI.
; The default is an empty list.
;mwi_mailboxes=mailbox_number[@context]{,mailbox_number[@context]}
;
; Whether or not restrict outgoing caller ID (will be sent as ANI only, not
; available for the user)
; Mostly use with FXS ports
; Does nothing.  Use hidecallerid instead.
;
;restrictcid=no
;
; Whether or not to use the caller ID presentation from the Asterisk channel
; for outgoing calls.
; See dialplan function CALLERID(pres) for more information.
; Only applies to PRI and SS7 channels.
;
usecallingpres=yes
;
; Some countries (UK) have ring tones with different ring tones (ring-ring),
; which means the caller ID needs to be set later on, and not just after
; the first ring, as per the default (1).
;
;sendcalleridafter = 2
;
;
; Support caller ID on Call Waiting
;
callwaitingcallerid=yes
;
; Support three-way calling
;
threewaycalling=yes
;
; For FXS ports (either direct analog or over T1/E1):
;   Support flash-hook call transfer (requires three way calling)
;   Also enables call parking (overrides the 'canpark' parameter)
;
; For digital ports using ISDN PRI protocols:
;   Support switch-side transfer (called 2BCT, RLT or other names)
;   This setting must be enabled on both ports involved, and the
;   'facilityenable' setting must also be enabled to allow sending
;   the transfer to the ISDN switch, since it sent in a FACILITY
;   message.
;   NOTE:  This should be disabled for NT PTMP mode.  Phones cannot
;   have tromboned calls pushed down to them.
;
transfer=yes
;
; Allow call parking
; ('canpark=no' is overridden by 'transfer=yes')
;
canpark=yes
;
; Support call forward variable
;
cancallforward=yes
;
; Whether or not to support Call Return (*69, if your dialplan doesn't
; catch this first)
;
callreturn=yes
;
; Stutter dialtone support: If a mailbox is specified without a voicemail
; context, then when voicemail is received in a mailbox in the default
; voicemail context in voicemail.conf, taking the phone off hook will cause a
; stutter dialtone instead of a normal one.
;
; If a mailbox is specified *with* a voicemail context, the same will result
; if voicemail received in mailbox in the specified voicemail context.
;
; for default voicemail context, the example below is fine:
;
;mailbox=1234
;
; for any other voicemail context, the following will produce the stutter tone:
;
;mailbox=1234@context
;
; Enable echo cancellation
; Use either "yes", "no", or a power of two from 32 to 256 if you wish to
; actually set the number of taps of cancellation.
;
; Note that when setting the number of taps, the number 256 does not translate
; to 256 ms of echo cancellation.  echocancel=256 means 256 / 8 = 32 ms.
;
; Note that if any of your DAHDI cards have hardware echo cancellers,
; then this setting only turns them on and off; numeric settings will
; be treated as "yes". There are no special settings required for
; hardware echo cancellers; when present and enabled in their kernel
; modules, they take precedence over the software echo canceller compiled
; into DAHDI automatically.
;
;
echocancel=yes
;
; Some DAHDI echo cancellers (software and hardware) support adjustable
; parameters; these parameters can be supplied as additional options to
; the 'echocancel' setting. Note that Asterisk does not attempt to
; validate the parameters or their values, so if you supply an invalid
; parameter you will not know the specific reason it failed without
; checking the kernel message log for the error(s) put there by DAHDI.
;
echocancel=512
;echocancel=128,param1=32,param2=0,param3=14
;
; Generally, it is not necessary (and in fact undesirable) to echo cancel when
; the circuit path is entirely TDM.  You may, however, change this behavior
; by enabling the echo canceller during pure TDM bridging below.
;
;echocancelwhenbridged=yes
;
; In some cases, the echo canceller doesn't train quickly enough and there
; is echo at the beginning of the call.  Enabling echo training will cause
; DAHDI to briefly mute the channel, send an impulse, and use the impulse
; response to pre-train the echo canceller so it can start out with a much
; closer idea of the actual echo.  Value may be "yes", "no", or a number of
; milliseconds to delay before training (default = 400)
;
; WARNING:  In some cases this option can make echo worse!  If you are
; trying to debug an echo problem, it is worth checking to see if your echo
; is better with the option set to yes or no.  Use whatever setting gives
; the best results.
;
; Note that these parameters do not apply to hardware echo cancellers.
;
echotraining=yes
echotraining=800
;
; If you are having trouble with DTMF detection, you can relax the DTMF
; detection parameters.  Relaxing them may make the DTMF detector more likely
; to have "talkoff" where DTMF is detected when it shouldn't be.
;
;relaxdtmf=yes
;
; You may also set the default receive and transmit gains (in dB)
;
; Gain Settings: increasing / decreasing the volume level on a channel.
;                The values are in db (decibells). A positive number
;                increases the volume level on a channel, and a
;                negavive value decreases volume level.
;
; Dynamic Range Compression: you can also enable dynamic range compression
;                on a channel. This will amplify quiet sounds while leaving
;                louder sounds untouched. This is useful in situations where
;                a linear gain setting would cause clipping. Acceptable values
;                are in the range of 0.0 to around 6.0 with higher values
;                causing more compression to be done.
;
;                There are several independent gain settings:
;   rxgain: gain for the rx (receive - into Asterisk) channel. Default: 0.0
;   txgain: gain for the tx (transmit - out of Asterisk Asterisk) channel.
;           Default: 0.0
;   cid_rxgain: set the gain just for the caller ID sounds Asterisk
;               emits. Default: 5.0 .
;   rxdrc: dynamic range compression for the rx channel. Default: 0.0
;   txdrc: dynamic range compression for the tx channel. Default: 0.0

rxgain=1.0
txgain=1.0
;
rxdrc=0.0
txdrc=0.0
;
; Logical groups can be assigned to allow outgoing roll-over.  Groups range
; from 0 to 63, and multiple groups can be specified. By default the
; channel is not a member of any group.
;
; Note that an explicit empty value for 'group' is invalid, and will not
; override a previous non-empty one. The same applies to callgroup and
; pickupgroup as well.
;
group=1
;
; Ring groups (a.k.a. call groups) and pickup groups.  If a phone is ringing
; and it is a member of a group which is one of your pickup groups, then
; you can answer it by picking up and dialing *8#.  For simple offices, just
; make these both the same.  Groups range from 0 to 63.
;
callgroup=1
pickupgroup=1

; Channel variable to be set for all calls from this channel
;setvar=CHANNEL=42
;setvar=ATTENDED_TRANSFER_COMPLETE_SOUND=beep   ; This channel variable will
                                                ; cause the given audio file to
                                                ; be played upon completion of
                                                ; an attended transfer.

;
; Specify whether the channel should be answered immediately or if the simple
; switch should provide dialtone, read digits, etc.
; Note: If immediate=yes the dialplan execution will always start at extension
; 's' priority 1 regardless of the dialed number!
;
;immediate=yes
;
; Specify whether flash-hook transfers to 'busy' channels should complete or
; return to the caller performing the transfer (default is yes).
;
;transfertobusy=no

; Calls will have the party id user tag set to this string value.
;
;cid_tag=

; With this set, you can automatically append the MSN of a party
; to the cid_tag.  An '_' is used to separate the tag from the MSN.
; Applies to ISDN spans.
; Default is no.
;
; Table of what number is appended:
;      outgoing  incoming
; net  dialed    caller
; cpe  caller    dialed
;
;append_msn_to_cid_tag=no

; caller ID can be set to "asreceived" or a specific number if you want to
; override it.  Note that "asreceived" only applies to trunk interfaces.
; fullname sets just the
;
; fullname: sets just the name part.
; cid_number: sets just the number part:
;
;callerid = 123456
;
;callerid = My Name <2564286000>
; Which can also be written as:
;cid_number = 2564286000
;fullname = My Name
;
;callerid = asreceived
;
; should we use the caller ID from incoming call on DAHDI transfer?
;
;useincomingcalleridondahditransfer = yes
;
; AMA flags affects the recording of Call Detail Records.  If specified
; it may be 'default', 'omit', 'billing', or 'documentation'.
;
;amaflags=default
;
; Channels may be associated with an account code to ease
; billing
;
;accountcode=lss0101
;
; ADSI (Analog Display Services Interface) can be enabled on a per-channel
; basis if you have (or may have) ADSI compatible CPE equipment
;
;adsi=yes
;
; SMDI (Simplified Message Desk Interface) can be enabled on a per-channel
; basis if you would like that channel to behave like an SMDI message desk.
; The SMDI port specified should have already been defined in smdi.conf.  The
; default port is /dev/ttyS0.
;
;usesmdi=yes
;smdiport=/dev/ttyS0
;
; On trunk interfaces (FXS) and E&M interfaces (E&M, Wink, Feature Group D
; etc, it can be useful to perform busy detection either in an effort to
; detect hangup or for detecting busies.  This enables listening for
; the beep-beep busy pattern.
;
;busydetect=yes
;
; If busydetect is enabled, it is also possible to specify how many busy tones
; to wait for before hanging up.  The default is 3, but it might be
; safer to set to 6 or even 8.  Mind that the higher the number, the more
; time that will be needed to hangup a channel, but lowers the probability
; that you will get random hangups.
;
;busycount=6
;
; If busydetect is enabled, it is also possible to specify the cadence of your
; busy signal.  In many countries, it is 500msec on, 500msec off.  Without
; busypattern specified, we'll accept any regular sound-silence pattern that
; repeats <busycount> times as a busy signal.  If you specify busypattern,
; then we'll further check the length of the sound (tone) and silence, which
; will further reduce the chance of a false positive.
;
;busypattern=500,500
;
; NOTE: In make menuselect, you'll find further options to tweak the busy
; detector.  If your country has a busy tone with the same length tone and
; silence (as many countries do), consider enabling the
; BUSYDETECT_COMPARE_TONE_AND_SILENCE option.
;
; To further detect which hangup tone your telco provider is sending, it is
; useful to use the dahdi_monitor utility to record the audio that main/dsp.c
; is receiving after the caller hangs up.
;
; For FXS (FXO signalled) ports
;   switch the line polarity to signal the connected PBX that an outgoing
;   call was answered by the remote party.
; For FXO (FXS signalled) ports
;   watch for a polarity reversal to mark when a outgoing call is
;   answered by the remote party.
;
;answeronpolarityswitch=yes
;
; For FXS (FXO signalled) ports
;   switch the line polarity to signal the connected PBX that the current
;   call was "hung up" by the remote party
; For FXO (FXS signalled) ports
;   In some countries, a polarity reversal is used to signal the disconnect of a
;   phone line.  If the hanguponpolarityswitch option is selected, the call will
;   be considered "hung up" on a polarity reversal.
;
;hanguponpolarityswitch=yes
;
; polarityonanswerdelay: minimal time period (ms) between the answer
;                        polarity switch and hangup polarity switch.
;                        (default: 600ms)
;
; On trunk interfaces (FXS) it can be useful to attempt to follow the progress
; of a call through RINGING, BUSY, and ANSWERING.   If turned on, call
; progress attempts to determine answer, busy, and ringing on phone lines.
; This feature is HIGHLY EXPERIMENTAL and can easily detect false answers,
; so don't count on it being very accurate.
;
; Few zones are supported at the time of this writing, but may be selected
; with "progzone".
;
; progzone also affects the pattern used for buzydetect (unless
; busypattern is set explicitly). The possible values are:
;   us (default)
;   ca (alias for 'us')
;   cr (Costa Rica)
;   br (Brazil, alias for 'cr')
;   uk
;
; This feature can also easily detect false hangups. The symptoms of this is
; being disconnected in the middle of a call for no reason.
;
;callprogress=yes
;progzone=uk
;
; Set the tonezone. Equivalent of the defaultzone settings in
; /etc/dahdi/system.conf. This sets the tone zone by number.
; Note that you'd still need to load tonezones (loadzone in
; /etc/dahdi/system.conf).
; The default is -1: not to set anything.
;tonezone = 0 ; 0 is US
;
; FXO (FXS signalled) devices must have a timeout to determine if there was a
; hangup before the line was answered.  This value can be tweaked to shorten
; how long it takes before DAHDI considers a non-ringing line to have hungup.
;
; ringtimeout will not update on a reload.
;
;ringtimeout=8000
;
; For FXO (FXS signalled) devices, whether to use pulse dial instead of DTMF
; Pulse digits from phones (FXS devices, FXO signalling) are always
; detected.
;
;pulsedial=yes
;
; For fax detection, uncomment one of the following lines.  The default is *OFF*
;
;faxdetect=both
;faxdetect=incoming
;faxdetect=outgoing
;faxdetect=no
;


Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #6 on: March 08, 2013, 01:16:48 PM »
Part 3
Code: [Select]
; When 'faxdetect' is used, one could use 'faxbuffers' to configure the DAHDI
; transmit buffer policy.  The default is *OFF*.  When this configuration
; option is used, the faxbuffer policy will be used for the life of the call
; after a fax tone is detected.  The faxbuffer policy is reverted after the
; call is torn down.  The sample below will result in 6 buffers and a full
; buffer policy.
;
;faxbuffers=>6,full
;
; This option specifies a preference for which music on hold class this channel
; should listen to when put on hold if the music class has not been set on the
; channel with Set(CHANNEL(musicclass)=whatever) in the dialplan, and the peer
; channel putting this one on hold did not suggest a music class.
;
; If this option is set to "passthrough", then the hold message will always be
; passed through as signalling instead of generating hold music locally. This
; setting is only valid when used on a channel that uses digital signalling.
;
; This option may be set globally or on a per-channel basis.
;
;mohinterpret=default
;
; This option specifies which music on hold class to suggest to the peer channel
; when this channel places the peer on hold.  This option may be set globally,
; or on a per-channel basis.
;
;mohsuggest=default
;
; PRI channels can have an idle extension and a minunused number.  So long as
; at least "minunused" channels are idle, chan_dahdi will try to call "idledial"
; on them, and then dump them into the PBX in the "idleext" extension (which
; is of the form exten@context).  When channels are needed the "idle" calls
; are disconnected (so long as there are at least "minidle" calls still
; running, of course) to make more channels available.  The primary use of
; this is to create a dynamic service, where idle channels are bundled through
; multilink PPP, thus more efficiently utilizing combined voice/data services
; than conventional fixed mappings/muxings.
;
; Those settings cannot be changed on reload.
;
;idledial=6999
;idleext=6999@dialout
;minunused=2
;minidle=1
;
;
; ignore_failed_channels: Continue even if some channels failed to configure.
; False by default, as if even a single channel failed to configure, it might
; mean other channels are misplaced and having them work may not be a good
; idea. If enabled (set to true), chan_dahdi will nevertheless attempt to
; configure other channels rather than giving up. This normally makes sense
; only if you use names (<subdir>!<number>) for DAHDI channels.
;ignore_failed_channels = true
;
; Configure jitter buffers in DAHDI (each one is 20ms, default is 4)
; This is set globally, rather than per-channel.
;
;jitterbuffers=4
;
;------------------------------ JITTER BUFFER CONFIGURATION --------------------------
; jbenable = yes              ; Enables the use of a jitterbuffer on the receiving side of a
                              ; DAHDI channel. Defaults to "no". An enabled jitterbuffer will
                              ; be used only if the sending side can create and the receiving
                              ; side can not accept jitter. The DAHDI channel can't accept jitter,
                              ; thus an enabled jitterbuffer on the receive DAHDI side will always
                              ; be used if the sending side can create jitter.

; jbmaxsize = 200             ; Max length of the jitterbuffer in milliseconds.

; jbresyncthreshold = 1000    ; Jump in the frame timestamps over which the jitterbuffer is
                              ; resynchronized. Useful to improve the quality of the voice, with
                              ; big jumps in/broken timestamps, usually sent from exotic devices
                              ; and programs. Defaults to 1000.

; jbimpl = fixed              ; Jitterbuffer implementation, used on the receiving side of a DAHDI
                              ; channel. Two implementations are currently available - "fixed"
                              ; (with size always equals to jbmax-size) and "adaptive" (with
                              ; variable size, actually the new jb of IAX2). Defaults to fixed.

; jbtargetextra = 40          ; This option only affects the jb when 'jbimpl = adaptive' is set.
                              ; The option represents the number of milliseconds by which the new
                              ; jitter buffer will pad its size. the default is 40, so without
                              ; modification, the new jitter buffer will set its size to the jitter
                              ; value plus 40 milliseconds. increasing this value may help if your
                              ; network normally has low jitter, but occasionally has spikes.

; jblog = no                  ; Enables jitterbuffer frame logging. Defaults to "no".
;-----------------------------------------------------------------------------------
;
; You can define your own custom ring cadences here.  You can define up to 8
; pairs.  If the silence is negative, it indicates where the caller ID spill is
; to be placed.  Also, if you define any custom cadences, the default cadences
; will be turned off.
;
; This setting is global, rather than per-channel. It will not update on
; a reload.
;
; Syntax is:  cadence=ring,silence[,ring,silence[...]]
;
; These are the default cadences:
;
;cadence=125,125,2000,-4000
;cadence=250,250,500,1000,250,250,500,-4000
;cadence=125,125,125,125,125,-4000
;cadence=1000,500,2500,-5000
;
; Each channel consists of the channel number or range.  It inherits the
; parameters that were specified above its declaration.
;
;
;callerid="Green Phone"<(256) 428-6121>
;channel => 1
;callerid="Black Phone"<(256) 428-6122>
;channel => 2
;callerid="CallerID Phone" <(630) 372-1564>
;channel => 3
;callerid="Pac Tel Phone" <(256) 428-6124>
;channel => 4
;callerid="Uniden Dead" <(256) 428-6125>
;channel => 5
;callerid="Cortelco 2500" <(256) 428-6126>
;channel => 6
;callerid="Main TA 750" <(256) 428-6127>
;channel => 44
;
; For example, maybe we have some other channels which start out in a
; different context and use E & M signalling instead.
;
;context=remote
;signaling=em
;channel => 15
;channel => 16

;signalling=em_w
;
; All those in group 0 I'll use for outgoing calls
;
; Strip most significant digit (9) before sending
;
;stripmsd=1
;callerid=asreceived
;group=0
;signalling=fxs_ls
;channel => 45

;signalling=fxo_ls
;group=1
;callerid="Joe Schmoe" <(256) 428-6131>
;channel => 25
;callerid="Megan May" <(256) 428-6132>
;channel => 26
;callerid="Suzy Queue" <(256) 428-6233>
;channel => 27
;callerid="Larry Moe" <(256) 428-6234>
;channel => 28
;
; Sample PRI (CPE) config:  Specify the switchtype, the signalling as either
; pri_cpe or pri_net for CPE or Network termination, and generally you will
; want to create a single "group" for all channels of the PRI.
;
; switchtype cannot be changed on a reload.
;
; switchtype = national
; signalling = pri_cpe
; group = 2
; channel => 1-23
;
; Alternatively, the number of the channel may be replaced with a relative
; path to a device file under /dev/dahdi . The final element of that file
; must be a number, though. The directory separator is '!', as we can't
; use '/' in a dial string. So if we have
;
;   /dev/dahdi/span-name/pstn/00/1
;   /dev/dahdi/span-name/pstn/00/2
;   /dev/dahdi/span-name/pstn/00/3
;   /dev/dahdi/span-name/pstn/00/4
;
; we could use:
;channel => span-name!pstn!00!1-4
;
; or:
;channel => span-name!pstn!00!1,2,3,4
;
; See also ignore_failed_channels above.

;  Used for distinctive ring support for x100p.
;  You can see the dringX patterns is to set any one of the dringXcontext fields
;  and they will be printed on the console when an inbound call comes in.
;
;  dringXrange is used to change the acceptable ranges for "tone offsets".  Defaults to 10.
;  Note: a range of 0 is NOT what you might expect - it instead forces it to the default.
;  A range of -1 will force it to always match.
;  Anything lower than -1 would presumably cause it to never match.
;
;dring1=95,0,0
;dring1context=internal1
;dring1range=10
;dring2=325,95,0
;dring2context=internal2
;dring2range=10
; If no pattern is matched here is where we go.
;context=default
;channel => 1

; AMI alarm event reporting
;reportalarms=channels
;Possible values are:
;channels - report each channel alarms (current behavior, default for backward compatibility)
;spans - report an "SpanAlarm" event when the span of any configured channel is alarmed
;all - report channel and span alarms (aggregated behavior)
;none - do not report any alarms.

; ---------------- Options for use with signalling=ss7 -----------------
; None of them can be changed by a reload.
;
; Variant of SS7 signalling:
; Options are itu and ansi
;ss7type = itu

; SS7 Called Nature of Address Indicator
;
; unknown:        Unknown
; subscriber:     Subscriber
; national:       National
; international:  International
; dynamic:        Dynamically selects the appropriate dialplan
;
;ss7_called_nai=dynamic
;
; SS7 Calling Nature of Address Indicator
;
; unknown:        Unknown
; subscriber:     Subscriber
; national:       National
; international:  International
; dynamic:        Dynamically selects the appropriate dialplan
;
;ss7_calling_nai=dynamic
;
;
; sample 1 for Germany
;ss7_internationalprefix = 00
;ss7_nationalprefix = 0
;ss7_subscriberprefix =
;ss7_unknownprefix =
;

; This option is used to disable automatic sending of ACM when the call is started
; in the dialplan.  If you do use this option, you will need to use the Proceeding()
; application in the dialplan to send ACM.
;ss7_explictacm=yes

; All settings apply to linkset 1
;linkset = 1

; Point code of the linkset.  For ITU, this is the decimal number
; format of the point code.  For ANSI, this can either be in decimal
; number format or in the xxx-xxx-xxx format
;pointcode = 1

; Point code of node adjacent to this signalling link (Possibly the STP between you and
; your destination).  Point code format follows the same rules as above.
;adjpointcode = 2

; Default point code that you would like to assign to outgoing messages (in case of
; routing through STPs, or using A links).  Point code format follows the same rules
; as above.
;defaultdpc = 3

; Begin CIC (Circuit indication codes) count with this number
;cicbeginswith = 1

; What the MTP3 network indicator bits should be set to.  Choices are
; national, national_spare, international, international_spare
;networkindicator=international

; First signalling channel
;sigchan = 48

; Additional signalling channel for this linkset (So you can have a linkset
; with two signalling links in it).  It seems like a silly way to do it, but
; for linksets with multiple signalling links, you add an additional sigchan
; line for every additional signalling link on the linkset.
;sigchan = 96

; Channels to associate with CICs on this linkset
;channel = 25-47
;
; For more information on setting up SS7, see the README file in libss7 or
; https://wiki.asterisk.org/wiki/display/AST/Signaling+System+Number+7
; ----------------- SS7 Options ----------------------------------------

; ---------------- Options for use with signalling=mfcr2 --------------

; MFC-R2 signaling has lots of variants from country to country and even sometimes
; minor variants inside the same country. The only mandatory parameters here are:
; mfcr2_variant, mfcr2_max_ani and mfcr2_max_dnis.
; IT IS RECOMMENDED that you leave the default values (leaving it commented) for the
; other parameters unless you have problems or you have been instructed to change some
; parameter. OpenR2 library uses the mfcr2_variant parameter to try to determine the
; best defaults for your country, also refer to the OpenR2 package directory
; doc/asterisk/ where you can find sample configurations for some countries. If you
; want to contribute your configs for a particular country send them to the e-mail
; of the primary OpenR2 developer that you can find in the AUTHORS file of the OpenR2 package



Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #7 on: March 08, 2013, 01:17:51 PM »
Part 4:
Code: [Select]
; MFC/R2 variant. This depends on the OpenR2 supported variants
; A list of values can be found by executing the openr2 command r2test -l
; some valid values are:
; ar (Argentina)
; br (Brazil)
; mx (Mexico)
; ph (Philippines)
; itu (per ITU spec)
; mfcr2_variant=mx

; Max amount of ANI to ask for
; mfcr2_max_ani=10

; Max amount of DNIS to ask for
; mfcr2_max_dnis=4

; whether or not to get the ANI before getting DNIS.
; some telcos require ANI first some others do not care
; if this go wrong, change this value
; mfcr2_get_ani_first=no

; Caller Category to send
; national_subscriber
; national_priority_subscriber
; international_subscriber
; international_priority_subscriber
; collect_call
; usually national_subscriber works just fine
; you can change this setting from the dialplan
; by setting the variable MFCR2_CATEGORY
; (remember to set _MFCR2_CATEGORY from originating channels)
; MFCR2_CATEGORY will also be a variable available in your context
; on incoming calls set to the value received from the far end
; mfcr2_category=national_subscriber

; Call logging is stored at the Asterisk
; logging directory specified in asterisk.conf
; plus mfcr2/<whatever you put here>
; if you specify 'span1' here and asterisk.conf has
; as logging directory /var/log/asterisk then the full
; path to your MFC/R2 call logs will be /var/log/asterisk/mfcr2/span1
; (the directory will be automatically created if not present already)
; remember to set mfcr2_call_files=yes
; mfcr2_logdir=span1

; whether or not to drop call files into mfcr2_logdir
; mfcr2_call_files=yes|no

; MFC/R2 valid logging values are: all,error,warning,debug,notice,cas,mf,stack,nothing
; error,warning,debug and notice are self-descriptive
; 'cas' is for logging ABCD CAS tx and rx
; 'mf' is for logging of the Multi Frequency tones
; 'stack' is for very verbose output of the channel and context call stack, only useful
; if you are debugging a crash or want to learn how the library works. The stack logging
; will be only enabled if the openr2 library was compiled with -DOR2_TRACE_STACKS
; You can mix up values, like: loglevel=error,debug,mf to log just error, debug and
; multi frequency messages
; 'all' is a special value to log all the activity
; 'nothing' is a clean-up value, in case you want to not log any activity for
; a channel or group of channels
; BE AWARE that the level of output logged will ALSO depend on
; the value you have in logger.conf, if you disable output in logger.conf
; then it does not matter you specify 'all' here, nothing will be logged
; so logger.conf has the last word on what is going to be logged
; mfcr2_logging=all

; MFC/R2 value in milliseconds for the MF timeout. Any negative value
; means 'default', smaller values than 500ms are not recommended
; and can cause malfunctioning. If you experience protocol error
; due to MF timeout try incrementing this value in 500ms steps
; mfcr2_mfback_timeout=-1

; MFC/R2 value in milliseconds for the metering pulse timeout.
; Metering pulses are sent by some telcos for some R2 variants
; during a call presumably for billing purposes to indicate costs,
; however this pulses use the same signal that is used to indicate
; call hangup, therefore a timeout is sometimes required to distinguish
; between a *real* hangup and a billing pulse that should not
; last more than 500ms, If you experience call drops after some
; minutes of being stablished try setting a value of some ms here,
; values greater than 500ms are not recommended.
; BE AWARE that choosing the proper protocol mfcr2_variant parameter
; implicitly sets a good recommended value for this timer, use this
; parameter only when you *really* want to override the default, otherwise
; just comment out this value or put a -1
; Any negative value means 'default'.
; mfcr2_metering_pulse_timeout=-1

; Brazil uses a special calling party category for collect calls (llamadas por cobrar)
; instead of using the operator (as in Mexico). The R2 spec in Brazil says a special GB tone
; should be used to reject collect calls. If you want to ALLOW collect calls specify 'yes',
; if you want to BLOCK collect calls then say 'no'. Default is to block collect calls.
; (see also 'mfcr2_double_answer')
; mfcr2_allow_collect_calls=no

; This feature is related but independent of mfcr2_allow_collect_calls
; Some PBX's require a double-answer process to block collect calls, if
; you ever have problems blocking collect calls using Group B signals (mfcr2_allow_collect_calls=no)
; then you may want to try with mfcr2_double_answer=yes, this will cause that every answer signal
; is changed by answer->clear back->answer (sort of a flash)
; (see also 'mfcr2_allow_collect_calls')
; mfcr2_double_answer=no

; This feature allows to skip the use of Group B/II signals and go directly
; to the accepted state for incoming calls
; mfcr2_immediate_accept=no

; You most likely dont need this feature. Default is yes.
; When this is set to yes, all calls that are offered (incoming calls) which
; DNIS is valid (exists in extensions.conf) and pass collect call validation
; will be accepted with a Group B tone (either call with charge or not, depending on mfcr2_charge_calls)
; with this set to 'no' then the call will NOT be accepted on offered, and the call will start its
; execution in extensions.conf without being accepted until the channel is answered (either with Answer() or
; any other application resulting in the channel being answered).
; This can be set to 'no' if your telco or PBX needs the hangup cause to be set accurately
; when this option is set to no you must explicitly accept the call with DAHDIAcceptR2Call
; or implicitly through the Answer() application.
; mfcr2_accept_on_offer=yes

; Skip request of calling party category and ANI
; you need openr2 >= 1.2.0 to use this feature
; mfcr2_skip_category=no

; WARNING: advanced users only! I really mean it
; this parameter is commented by default because
; YOU DON'T NEED IT UNLESS YOU REALLY GROK MFC/R2
; READ COMMENTS on doc/r2proto.conf in openr2 package
; for more info
; mfcr2_advanced_protocol_file=/path/to/r2proto.conf

; Brazil use a special signal to force the release of the line (hangup) from the
; backward perspective. When mfcr2_forced_release=no, the normal clear back signal
; will be sent on hangup, which is OK for all mfcr2 variants I know of, except for
; Brazilian variant, where the central will leave the line up for several seconds (30, 60)
; which sometimes is not what people really want. When mfcr2_forced_release=yes, a different
; signal will be sent to hangup the call indicating that the line should be released immediately
; mfcr2_forced_release=no

; Whether or not report to the other end 'accept call with charge'
; This setting has no effect with most telecos, usually is safe
; leave the default (yes), but once in a while when interconnecting with
; old PBXs this may be useful.
; Concretely this affects the Group B signal used to accept calls
; The application DAHDIAcceptR2Call can also be used to decide this
; in the dial plan in a per-call basis instead of doing it here for all calls
; mfcr2_charge_calls=yes

; ---------------- END of options to be used with signalling=mfcr2

; Configuration Sections
; ~~~~~~~~~~~~~~~~~~~~~~
; You can also configure channels in a separate chan_dahdi.conf section. In
; this case the keyword 'channel' is not used. Instead the keyword
; 'dahdichan' is used (as in users.conf) - configuration is only processed
; in a section where the keyword dahdichan is used. It will only be
; processed in the end of the section. Thus the following section:
;
;[phones]
;echocancel = 64
;dahdichan = 1-8
;group = 1
;
; Is somewhat equivalent to the following snippet in the section
; [channels]:
;
;echocancel = 64
;group = 1
;channel => 1-8
;
; When starting a new section almost all of the configuration values are
; copied from their values at the end of the section [channels] in
; chan_dahdi.conf and [general] in users.conf - one section's configuration
; does not affect another one's.
;
; Instead of letting common configuration values "slide through" you can
; use configuration templates to easily keep the common part in one
; place and override where needed.
;
;[phones](!)
;echocancel = yes
;group = 0,4
;callgroup = 3
;pickupgroup = 3
;threewaycalling = yes
;transfer = yes
;context = phones
;faxdetect = incoming
;
;[phone-1](phones)
;dahdichan = 1
;callerid = My Name <501>
;mailbox = 501@mailboxes
;
;
;[fax](phones)
;dahdichan = 2
;faxdetect = no
;context = fax
;
;[phone-3](phones)
;dahdichan = 3
;pickupgroup = 3,4
#include dahdi-channels.conf

Offline SARK devs

  • ****
  • 2,806
  • +1/-0
    • http://sarkpbx.com
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #8 on: March 08, 2013, 08:41:21 PM »
Hi there

You can install version 8 if you use one of our distros which you can find here

http://sailpbx.com/iso/SME8/

They are modified SME iso images with sail, asterisk/dahdi and the correct drivers already on it.   You can only do fresh installs with it, you can't upgrade an existing box but it may get you where you want to go.


Best
S

Offline Franco

  • *
  • 1,171
  • +0/-0
    • http://contribs.org
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #9 on: March 09, 2013, 03:06:25 AM »
Hi Frank,
The echocancel parameter can either be =yes (defaults to 128) or a different number lower or higher than 128 ( ^2).
You need to edit it and either leave as =yes or =512 as per your example.

You should also disable echotraining, as it can make it worse.

To tune your card, see here:
http://kb.digium.com/articles/Troubleshooting/how-to-basic-fxotune
and here:
http://kb.digium.com/articles/Configuration/General-Echo-HOWTO

This is what my system reads and I have perfect results:
Code: [Select]
[root@linux ~]# cat /etc/fxotune.conf
2=5,0,0,0,0,0,0,0,0
4=6,0,0,0,0,0,0,0,0

Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #10 on: March 10, 2013, 02:37:55 PM »
Thanks Franco, did the changes in the dahdi.conf file (echo training etc), RE fxotune it tells me:

Code: [Select]
fxotune -i -n 4 -b 1 -e 4 -p -vvv
/dev/dahdi/1 absent: Device or resource busy
/dev/dahdi/2 absent: Device or resource busy
/dev/dahdi/3 absent: Device or resource busy
/dev/dahdi/4 absent: Device or resource busy

I cannot stop the dahdi service:

Code: [Select]
service dahdi stop
Unloading DAHDI hardware modules: ERROR: Module wctdm24xxp is in use
ERROR: Module wctdm24xxp is in use
ERROR: Module dahdi_voicebus is in use by wctdm24xxp
ERROR: Module dahdi_echocan_mg2 is in use
ERROR: Module dahdi is in use by dahdi_echocan_mg2,wctdm24xxp,dahdi_voicebus
error

How do I get the thing to run? Sorry for the potentially stupid question, but I haven't worked with this before
MAny thanks for your help
FRank

Offline Franco

  • *
  • 1,171
  • +0/-0
    • http://contribs.org
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #11 on: March 10, 2013, 07:49:26 PM »
Hi Frank,

You need to stop asterisk first.
I use the FreePBX contrib, therefore:
Code: [Select]
amportal stop

Let me know how it goes,

Best,

Offline Franco

  • *
  • 1,171
  • +0/-0
    • http://contribs.org
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #12 on: March 10, 2013, 08:36:48 PM »
I just did on a machine, installed a digium card and the echo was terrible.

Code: [Select]
amportal stop
/etc/init.d/dahdi stop

then I edited /etc/dahdi/system.conf and added:
echocanceller=none

edited /etc/asterisk/chan_dahdi.conf and added a ; infont of echotrining
;echotraining=800

/etc/init.d/dahdi start

then:

fxotune -i 5
Tuning module /dev/dahdi/1
Done!
Tuning module /dev/dahdi/2
Done!

fxotune -i 5 (took 2 minutes for each channel)

fxotune -s to record the values

then:
amportal start

and I have:
cat /etc/fxotune.conf
1=0,0,0,0,0,0,0,0,0
2=9,255,1,4,0,0,1,255,0

All good to go and perfect sound.

Best,

Offline apmuthu

  • *
  • 244
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #13 on: March 11, 2013, 05:46:36 AM »
Wiki-ed it.


Offline jufra

  • *
  • 34
  • +0/-0
Re: Dahdi calls scratchy, will upgrade help, and how to upgrade
« Reply #14 on: March 11, 2013, 11:25:43 AM »
Hi Franco, hmm, this thing leaves me scratching my head.
I've had no success yet, still scratchy noise, not echo, but terrible sound quality which keeps getting worse as the call ages...
I've done what you suggested in chan_dahdi.conf

Code: [Select]
; Enable echo cancellation
; Use either "yes", "no", or a power of two from 32 to 256 if you wish to
; actually set the number of taps of cancellation.
;
; Note that when setting the number of taps, the number 256 does not translate
; to 256 ms of echo cancellation.  echocancel=256 means 256 / 8 = 32 ms.
;
; Note that if any of your DAHDI cards have hardware echo cancellers,
; then this setting only turns them on and off; numeric settings will
; be treated as "yes". There are no special settings required for
; hardware echo cancellers; when present and enabled in their kernel
; modules, they take precedence over the software echo canceller compiled
; into DAHDI automatically.
;
;
;echocancel=yes
;
; Some DAHDI echo cancellers (software and hardware) support adjustable
; parameters; these parameters can be supplied as additional options to
; the 'echocancel' setting. Note that Asterisk does not attempt to
; validate the parameters or their values, so if you supply an invalid
; parameter you will not know the specific reason it failed without
; checking the kernel message log for the error(s) put there by DAHDI.
;
echocancel=512
;echocancel=128,param1=32,param2=0,param3=14
;
; Generally, it is not necessary (and in fact undesirable) to echo cancel when
; the circuit path is entirely TDM.  You may, however, change this behavior
; by enabling the echo canceller during pure TDM bridging below.
;
;echocancelwhenbridged=yes
;
; In some cases, the echo canceller doesn't train quickly enough and there
; is echo at the beginning of the call.  Enabling echo training will cause
; DAHDI to briefly mute the channel, send an impulse, and use the impulse
; response to pre-train the echo canceller so it can start out with a much
; closer idea of the actual echo.  Value may be "yes", "no", or a number of
; milliseconds to delay before training (default = 400)
;
; WARNING:  In some cases this option can make echo worse!  If you are
; trying to debug an echo problem, it is worth checking to see if your echo
; is better with the option set to yes or no.  Use whatever setting gives
; the best results.
;
; Note that these parameters do not apply to hardware echo cancellers.
;
;echotraining=yes
;echotraining=800
;
; If you are having trouble with DTMF detection, you can relax the DTMF
; detection parameters.  Relaxing them may make the DTMF detector more likely
; to have "talkoff" where DTMF is detected when it shouldn't be.
;
;relaxdtmf=yes
;
; You may also set the default receive and transmit gains (in dB)
;
; Gain Settings: increasing / decreasing the volume level on a channel.
;                The values are in db (decibells). A positive number
;                increases the volume level on a channel, and a
;                negavive value decreases volume level.
;
; Dynamic Range Compression: you can also enable dynamic range compression
;                on a channel. This will amplify quiet sounds while leaving
;                louder sounds untouched. This is useful in situations where
;                a linear gain setting would cause clipping. Acceptable values
;                are in the range of 0.0 to around 6.0 with higher values
;                causing more compression to be done.
;
;                There are several independent gain settings:
;   rxgain: gain for the rx (receive - into Asterisk) channel. Default: 0.0
;   txgain: gain for the tx (transmit - out of Asterisk Asterisk) channel.
;           Default: 0.0
;   cid_rxgain: set the gain just for the caller ID sounds Asterisk
;               emits. Default: 5.0 .
;   rxdrc: dynamic range compression for the rx channel. Default: 0.0
;   txdrc: dynamic range compression for the tx channel. Default: 0.0

rxgain=1.0
txgain=1.0
;
rxdrc=0.0
txdrc=0.0
;

and my system.conf looks like this:
Code: [Select]
cat /etc/dahdi/system.conf
# Autogenerated by /usr/sbin/dahdi_genconf on Sun Jan 27 16:32:08 2013
# If you edit this file and execute /usr/sbin/dahdi_genconf again,
# your manual changes will be LOST.
# Dahdi Configuration File
#
# This file is parsed by the Dahdi Configurator, dahdi_cfg
#
# Span 1: WCTDM/0 "Wildcard TDM410P Board 1" (MASTER)
fxsks=1
#echocanceller=mg2,1
echocanceller=none

fxsks=2
#echocanceller=mg2,2
echocanceller=none

fxsks=3
#echocanceller=mg2,3
echocanceller=none

fxsks=4
#echocanceller=mg2,4
echocanceller=none

# Global data

loadzone = au
defaultzone = au

Did the fxotune and the fxotune.conf looks like this:
Code: [Select]
cat /etc/fxotune.conf
1=5,255,252,0,2,254,0,255,255
2=5,3,251,250,3,254,255,255,255
3=5,255,252,0,2,254,0,255,255
4=0,0,0,0,0,0,0,0,0


So, is it an echo canceller problem, or is it some sort of transcoding issue? I have no idea where to go with this one. And it also seems I'm stuck with the current versions of asterisk and dahdi unless we did a complete reinstall with the more recent versions... and this is a big problem as the server is a live box which is used in a small business

Cheers
Frank