I notice during the NGREP session , there is this
field called Call-ID
. It is something like this format :-
9ce97fd3-1a800040(a)172.16.1.123 <mailto:9ce97fd3-1a800040@172.16.1.123>.
According to the RFC , it is said that this field is generated by the
UA who started the request. I was just wondering, how did they get the
IP Address at the back ? Which IP Address did they reference it from ?
Most often
this is the address of the interface the UA is using.
From the RFC, it also says that the all request/response in the dialog
should use the same Call-ID. When they say this, does it mean the
entire field with the ip address (eg. 9ce97fd3-1a800040(a)172.16.1.123
<mailto:9ce97fd3-1a800040@172.16.1.123> ) or just the front part of it
(eg. 9ce97fd3-1a800040 ) ?
Entire.
I notice that during a dialog session from a UA to the
server going to
the PSTN gateway, it was refering to this Call-ID until the point when
the request was forwarded to the PSTN gateway. When the PSTN gateway
replies, it replied with a same Call-ID with a different IP address at
the back (9ce97fd3-1a800040(a)123.234.123.234
<mailto:9ce97fd3-1a800040@123.234.123.234> )
172.16.1.123 - Private address of the UA
123.234.123.234 - Public address of the UA
Strange ? Give your comment. Somehow i feel that this implementation
is wrong.
Yes, a unique dialog is composed of to/from and call-id. Creating a new
call-id creates a new dialog and can only be done by a dialog-creation
SIP message.
g-)