<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<br>
<br>
On 6/16/2012 5:38 PM, Shaun Ruffell wrote:
<blockquote cite="mid:20120616223806.GA19205@digium.com" type="cite">
<pre wrap="">On Sat, Jun 16, 2012 at 04:37:06PM -0500, Vladimir Mikhelson wrote:
</pre>
<blockquote type="cite">
<pre wrap="">
On top of that FXO channels exhibit intermittent problems with incoming
caller ID, FXS -- with DTMF detection. These two problems manifest
themselves with both Digium and generic cards. It looks like it is just
a deficiency in the current DAHDI implementation.
</pre>
</blockquote>
<pre wrap="">
Why do you believe it is a deficiency in DAHDI? The only issue that
I know of related to DTMF detection is DAHLIN-241 [1], which you
and I are currently trying to definitely establish is related to the
time it takes to page Asterisk pages back into memory, which isn't
really a DAHDI issue.
[1] <a class="moz-txt-link-freetext" href="https://issues.asterisk.org/jira/browse/DAHLIN-241">https://issues.asterisk.org/jira/browse/DAHLIN-241</a></pre>
</blockquote>
<br>
Shaun, I respect your opinion, and the swap theory is one of the
valid theories.<br>
<br>
But interestingly enough, yesterday morning I had zero (0) bytes in
the swap file and still experienced missing DTMF detection on an
outgoing call.<br>
<br>
<blockquote cite="mid:20120616223806.GA19205@digium.com" type="cite">
<pre wrap="">
Also, the only recent CID issue I'm aware of which was related to
DAHDI was introduced in 2.6.0 and fixed in 2.6.1 [2]. This issue
was that the ring event was reported late to Asterisk, so Asterisk
would bump the gains in the middle of the spill. If the line quality
was marginal the callerid would be missed.
[2] <a class="moz-txt-link-freetext" href="http://svnview.digium.com/svn/dahdi?view=revision&revision=10473">http://svnview.digium.com/svn/dahdi?view=revision&revision=10473</a>
Are there other issues with CID detection that you are currently
having? Generally CID issues aren't necesarily DAHDI related since
generation and detection is done in Asterisk and not DAHDI, like
DTMF detection (DTMF generation can be done in DAHDI...).</pre>
</blockquote>
<br>
I am not sure whether it is DAHDI or Asterisk specifically, well may
be one or both.<br>
<br>
All I know is I opened support ticket with Digium, tried all their
recommendations, and I am still experiencing the same CID detection
intermittency. Interestingly, in my case 2.6.0 was an improvement,
whereas 2.6.1 returned everything back to where it was before.<br>
<br>
I still did not find time to file on JIRA, but I eventually will.<br>
<br>
Here are the issues (not detailed, from memory):<br>
<ol>
<li>Incoming calls CID intermittency<br>
</li>
<li>Lack of a dropped call detection where 45 seconds long voice
mails with CO generated bipping are left and the line is
captured for up to 3 minutes after a caller dropped a call when
the IVR was active</li>
<li>Intermittent outgoing DTMF drop</li>
<li>CID flags are not being passed into dial plan for analysis, so
we cannot distinguish between private caller, no CID name, etc.<br>
</li>
</ol>
<br>
<br>
<br>
<blockquote cite="mid:20120616223806.GA19205@digium.com" type="cite">
<pre wrap="">
Cheers,
Shaun
</pre>
</blockquote>
</body>
</html>