Serial number zero and sequence space arithmetic

[Joe Rhett]

No, I believe it just invokes it without supplying a serial number which is
equivalent. (not actually looking at the source)

Regardless, named-xfer's input != named's internal processing of the serial.
Named-xfer doesn't need to think as much about the serial number.
 
On Tue, Jul 11, 2000 at 01:56:20PM -0400, William A. Gianopoulos wrote:
> 
> I did not think serial number zero ever went away, and do not understand 
> how things are working if it has.  I thought the mechanism the slave 
> servers used when they had no backup file for a zone when they did not 
> currently have one was to invoke named-xfer saying the current serial 
> number is zero and this was supposed to do an unconditional zone transfer.
> 
> I have not looked at the source recently however.
> 
> At 01:42 PM 7/11/00, Cricket Liu wrote:
> 
> > > > It looks to me from the 4.9.3 source like sequence space
> > > > arithmetic came in at about 4.9.  (I don't have any 4.9 source
> > > > older than 4.9.3, and 4.8.3 doesn't appear to use sequence
> > > > space.)  Can anyone confirm this?
> > >
> > > my oldest available copy is a snapshot of 4.9 dated 1993-02-02.
> > > Excerpt from that day's named-xfer.c:
> > >
> > > /* compare using sequence space arithmetic */
> > > if (SEQ_GT(zp_start.z_serial, serial_no) || serial_no == 0) {
> >
> >Excellent!  Thank you, Peter.  So sequence space did come in
> >at 4.9, because 4.8.3 doesn't have it.  I show serial 0 being
> >introduced at 4.8.1.  Now if I can just find out when serial 0
> >went away....
> >
> >cricket
> 
> 
> --
> William A. Gianopoulos
> Sr. Information Technologist II
> Raytheon Company
> Computer Systems Engineering
> 
> 

-- 
Joe Rhett                                         Chief Technology Officer
JRhett at ISite.Net                                      ISite Services, Inc.

PGP keys and contact information:          http://www.noc.isite.net/Staff/