[Home]

Summary:ASTERISK-20049: Double PROGRESS message when overlap dialing is enabled
Reporter:Maciej Krajewski (jamicque)Labels:
Date Opened:2012-06-26 04:33:14Date Closed:2012-06-29 08:23:34
Priority:MajorRegression?
Status:Closed/CompleteComponents:Channels/chan_dahdi
Versions:1.8.13.0 Frequency of
Occurrence
Related
Issues:
duplicatesASTERISK-18420 [regression] Inbound ISDN Overlap dial breaks with Asterisk Version >= 1.8.4.0
Environment:Attachments:( 0) overlap.log
Description:When Asterisk is receiving overlap digits through DAHDI it sends double PROGRESS message.
I thing that it might be connected to ASTERISK-17141
Sending double PROGRESS breaks dialling in some cases.
I've attached the log.
Comments:By: Maciej Krajewski (jamicque) 2012-06-26 04:34:20.045-0500

Here is the dialplan of the context:
{noformat}
 'T' =>            1. Hangup()                                   [pbx_config]
 'h' =>            1. NoOP("po AGI ===================="${EXTEN}" "${ORGDEST}") [pbx_config]
                   2. Hangup()                                   [pbx_config]
 't' =>            1. Hangup()                                   [pbx_config]
 '_.' =>           1. GotoIf($["${ORGDEST}" != ""]?4)            [pbx_config]
                   2. Set(__ORGDEST=${EXTEN})                    [pbx_config]
                   3. Set(TIMEOUT(digit)=5)                      [pbx_config]
                   4. agi(agi://127.0.0.1/script.agi)            [pbx_config]
                   5. NoOP("po AGI ===================="${EXTEN}" "${ORGDEST}") [pbx_config]
                   6. Hangup()                                   [pbx_config]
{noformat}

By: Michael L. Young (elguero) 2012-06-26 11:03:12.038-0500

I can't say I am an expert on this but based on your PRI log that is attached, nothing is jumping out at me.  I see 3 _different_ PROGRESS messages based on that attached log.  1 from the switch side.  2 from Asterisk.  The second PROGRESS message from Asterisk seems to be indicating to the switch that it is ready to accept Overlap Dial.  So, you are saying that this second message is messing up dialing?  This log that you attached, did the call fail?
By: Maciej Krajewski (jamicque) 2012-06-26 14:25:48.510-0500

You should look for those messages:
Message Type: PROGRESS

I guess that it might be because of my dialplan (_.).
However I do not think that there should be a PROGRESS message unless the Dial goes on the other leg (and there should be only one, not two).
By: Michael L. Young (elguero) 2012-06-26 15:39:45.961-0500

Thanks for clarifying... I was looking at the wrong thing... (Progress Indicator != Message Type: PROGRESS).

This report seems to be the same as ASTERISK-18420.  Take a look at the comments.  At that time it was determined that Asterisk did not have a bug but rather the other equipment was not handling PROGRESS correctly.  PROGRESS is supposed to indicate that a B channel should be attached in order to send inband tones/messages to the caller; it does not indicate a call state change.

Also, I see why you are saying double but it appears that it is not sending it twice.  The log from the other issue has the same messages and it was determined to be normal.  I think, if you look through the log, there is a brief description of the message and then what follows it is the details for that message - hence giving the appearance of being their twice.  I am not the expert on this though and shall wait for someone else with better knowledge to confirm or correct this observation of mine.
By: Maciej Krajewski (jamicque) 2012-06-26 16:08:10.686-0500

Thanks Michael, however I'm not exactly sure that it is one PROGRESS message.
By: Richard Mudgett (rmudgett) 2012-06-26 16:10:24.465-0500

{noformat}
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > DL-DATA request
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > Protocol Discriminator: Q.931 (8)  len=9
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > TEI=0 Call Ref: len= 2 (reference 1/0x1) (Sent to originator)
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > Message Type: PROGRESS (3)
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 TEI=0 Transmitting N(S)=6, window is open V(A)=5 K=7
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > Protocol Discriminator: Q.931 (8)  len=9
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > TEI=0 Call Ref: len= 2 (reference 1/0x1) (Sent to originator)
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > Message Type: PROGRESS (3)
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > [1e 02 81 88]
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 > Progress Indicator (len= 4) [ Ext: 1  Coding: CCITT (ITU) standard (0)  0: 0  Location: Private network serving the local user (1)
[2012-06-26 11:16:55] VERBOSE[17364] chan_dahdi.c: PRI Span: 1 >                               Ext: 1  Progress Description: Inband information or appropriate pattern now available. (8) ]
{noformat}

This is not sending the PROGRESS message twice.  The first part is Q.931 passing the PROGRESS message to the Q.921 layer (DL-DATA request).  The second part is Q.921 actually sending the message.

You should also be using something like the example below to overlap dial through your Asterisk box:
{noformat}
; Dial 9 for an outside line.
exten => _9!,1,Dial(DAHDI/g1/${EXTEN:1})
{noformat}

The '!' wildcard was created for dial-through applications because it will not wait for more digits to be dialed once the first part matches like the '.' wildcard will wait.

By: Rusty Newton (rnewton) 2012-06-28 18:34:59.825-0500

jamicque, is there still an issue present ? What do you mean by "breaks dialing"?. Richard couldn't see any issues in the your log file.
By: Maciej Krajewski (jamicque) 2012-06-29 02:18:14.024-0500

yes It was helpful I've misinterpreted the log. If DL-DATA did not send proceeding message - than everything is ok.

Changing the dialplan also seem to help to resolve the issue