[Asterisk-Dev] Potential debuggin patches for chan_sip.c

[Bill Moran]

Derek Samford wrote:
> I second this, and I think that I may even be capable of using my meager skills to help
 > along with this. I think it would be useful to have a debug info for just a certain peer
 > in channels, as well as more straightforward debug information as far as PRI's and T1's
 > are concerned. Anyone else have any input? Also, just as a sidenote, some minor changes
 > in how some info is outputted would be useful. For instance, for parsing purposes, it
 > would be easier if at the end of show channels, rather than outputting X active channels,
 > it outputted
> Active Channels: X
> 
> Simple things like that tend to make parsing easier if your playing with managing Asterisk.
 > Comments? Flames? Additional Input?

I don't disagree.

The specific problem we were having was that the phone was unable to register, yet we
were unable to determine _why_ the phone was unable to register (asterisks just says
"phone failed to register")  With full debugging on, I feel it would be good (and
appropriate) to say "phone failed to register with bad password" or "phone failed
to register: could not find matchine in config" (which was basically the problem we
were hitting ... difference in terminology, we didn't understand that what ZyXEL
calls the "phone number" is what it will try to register with ...)

The additional debugging is quick and dirty, and I'd make an effort to clean it up
before submitting ... which is why I asked first. :)

> 
> Derek Samford
> 
> -----Original Message-----
> From:	Bill Moran [mailto:wmoran at potentialtech.com]
> Sent:	Fri 5/28/2004 1:33 PM
> To:	asterisk-dev at lists.digium.com
> Cc:	
> Subject:	[Asterisk-Dev] Potential debuggin patches for chan_sip.c
> We had a lot of trouble getting a ZyXEL phone to work with Asterisks ... it turned
> out to be a configuration problem on our part, but using the debugging information
> to track it down was beyond difficult ... it was impossible.  The only way I finally
> figured it out was to add a bunch of additional debugging statements to chan_sip.c
> 
> Unfortunately, chan_sip.c has already been updated in CVS, so it's gonna be a PITA
> for me to post a patch.
> 
> Thus my first question: I assume I'm not the only one who thinks there's not enough
> debugging info ... is the team interested in my patches?  If so, I'll get the latest
> from CVS and figure out how to get good patches made.
> 
> My second question is related:  I added a few ast_log() statements to find_peer(),
> yet they never show up when asterisk is started with -vvv.  Any hints as to why
> this is?  Is find_peer() somehow unable to log?
> 


-- 
Bill Moran
Potential Technologies
http://www.potentialtech.com