[asterisk-dev] Feedback on hashtab-trunk

[Brian Capouch]

It built for me just fine (modulo a couple of longstanding pieces of 
cruft that require a bit of tweaking when cross-compiling for openWRT).

I am running it now, and made some test calls just to see how things looked.

I have encountered one problem so far, which appears to be in the code 
that interfaces the pbx to the CLI.  The call's uniqueid is reported 
just fine on the CLI when the call is set up, but then is mangled 
garbage as the call is torn down.  Related or not, this then causes an 
error when the 'h' extension is reached.  I might note that the dialplan 
lookup for this call was done via realtime.  I'm still testing, but it 
doesn't look like the problem exists when the call logic is built up 
from an entry in the extensions file.

Here's the CLI dump, thanks.

b.

-- Executing Dial("SIP/spa3k-0066f400", "IAX2/bc2/80100")
        > doing dnsmgr_lookup for '10.1.2.3'
        > doing dnsmgr_lookup for '10.1.2.3'
     -- Called bc2/80100
        > doing dnsmgr_lookup for '10.1.2.3'
     -- Call accepted by 10.1.2.3 (format gsm)
     -- Format for call is gsm
     -- IAX2/bc2-4 answered SIP/spa3k-0066f400
        > doing dnsmgr_lookup for '10.1.2.3'
     -- Hungup 'IAX2/bc2-4'
     -- Auto fallthrough, channel 'SIP/spa3k-0066f400' status is 'ANSWER'
     -- Executing Hangup("SIP/spa3k-0066f400", "IAX2/bc2￿￿[")
[Nov 11 13:57:36] NOTICE[825]: pbx.c:6407 pbx_builtin_hangup: Invalid 
cause given to Hangup(): "IAX2/bc2￿￿["


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.