On Fri, 2009-12-18 at 07:03 -0500, Scott Lawrence wrote:
> I don't understand why that shouldn't just be part of the messages in
> the file like the other header data is now (indeed, it is in the raw
> data).  Why would those be at the end of the file?  The extract process
> now doesn't do any of the analysis on anything else - why should it do
> this?

The problem is in siptrace-merge.  What I *want* is to be able to select
for a specified call-id, and all referenced dialogs.  In order for
siptrace-merge to do this selection in one pass through the (sometimes
huge) siptrace files, it has to know at the beginning all of the
references information -- a message early in the file may be part of a
referenced dialog, but the References header that makes it so may be
toward the end of the file.  So siptrace-merge must either make two
passes of the entire siptrace file, or the references information needs
to be put in a specific place in the siptrace file.  If syslog2siptrace
is going to work in one pass, the "specific place" can only be the end
of the siptrace file.

Dale


_______________________________________________
sipx-dev mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to