[VoIP] Firewall logs & status check results

[Steph Kerman]

Shane Young wrote:
> It works correctly if it's configured.
>
> Today, the CNET tandems (for the most part) are un-aware of each other 
> except during a call.
I don't see why they would need to be aware at any other time.  
Electromechanical exchanges with metallic trunks weren't aware of each 
other except during a call setup attempt either.
> Asterisk has the ability to send maintainence messages back and forth 
> to verify the distant peer is still alive (this is how I made the 
> status list).
>
> Because (normally, on CNET) we don't define peers for *all* of the 
> distant asterisk systems, we can't be aware of their state.
>
> If everyone built peers for each of the other tandems, there could 
> start to be a lot of maintainence messages going back and forth.
That would seem pointless as long as they can determine status at call 
setup attempt time.

Steph
> Quoting Steph Kerman <stfkerman at jps.net>:
>
>> John,
>>
>> Strictly speaking, the system was not actually hung as I understand the
>> term.   Its connection to the outside world was cut off by a failure of
>> the ISP's equipment and hence the broadband connection.  Perhaps that's
>> a distinction without a difference as far as other * boxes attempting to
>> connect to it are concerned.
>>
>> I suppose the lack of separate timeouts could be corrected if someone
>> modified the * source code but perhaps the benefit is not worth the
>> trouble of propagating the modifications forward with each new release
>> of *.  Call completion is pretty slow by present day PSTN standards and
>> I am used to that so if I perceived it as a no-connect and started
>> stabbing around to see whether it would respond to additional digits it
>> must have been a pretty long time.  So perhaps the effort of a separate
>> timeout would be justified.
>>
>> Steph
>>
>> John R. Covert wrote:
>>     
>>> Steph,
>>>
>>>
>>> You were calling through my portal to a system that was hung. It was
>>> not returning anything at all.
>>>
>>> So you needed to wait for the timeout, which unfortunately is the same
>>> as the timeout to allow someone to actually answer a
>>> phone. I wish there were two separate timeouts, but when systems
>>> connect autonomously rather than by registering with each other, each
>>> connection has to make a full attempt to connect.
>>>
>>> /john
>>>       
>> _______________________________________________
>> VoIP mailing list
>> VoIP at ckts.info
>> http://lists.ckts.info/mailman/listinfo/voip
>> Project Web Page: http://www.ckts.info/
>>
>>     
>
> --Shane
> +1-821-7311 CNET
>
> ----------------------------------------------------------------
>
> _______________________________________________
> VoIP mailing list
> VoIP at ckts.info
> http://lists.ckts.info/mailman/listinfo/voip
> Project Web Page: http://www.ckts.info/
>
>