Hi, The size of the zone should not be that much of a matter if you use IFXR. Aside from that, did you ever consider using another mechanism for synchronization, like rsync or lzmaing the zone and transferring it via your protocol of choice? Then again, why would one have a TLD coloc on a 256kbps line, that seems very unreasonable.
Regards -Sven On Wed, 2010-07-07 at 14:58 +0200, Niklas Jakobsson wrote: > Hello, > > On ons, 2010-07-07 at 14:41 +0200, Tom Schmitt wrote: > > -------- Original-Nachricht -------- > > > Datum: Wed, 07 Jul 2010 13:13:45 +0200 > > > Von: Niklas Jakobsson <[email protected]> > > > An: [email protected] > > > Betreff: bind says \'clocks are unsynchronized\' but they are not > > > > > Hello, > > > > > > I have some problems with our bind servers complaining that 'clocks are > > > unsynchronized' when doing zone transfers with TSIG. The problem is the > > > clocks are correct, synced with ntp and everything. > > > > Maybe one of the two servers doing the zone transfer is running in a chroot > > where it has another time setting than the server itself? > > > > Not running any chroot. > > > > > > > > > The problems seems to occur mostly on zone transfers that take a long > > > time (ie. hours). > > > > > > > HOURS?? > > There is defnitly something wrong. I cannot imagine a zone so big or a > > connection so slow that a zonetransfer could take hours. Or do you make a > > axfr of the tld com. over a serial connection? ;-) > > > > > > Tom. > > > > Size of a tld that should not be named: 256947194 bytes > Speed of connection to a site very far away: 256 kbit/s > > 256947194/(256*1000/8)/60/60 = 2.23 ~ little over 2 hours... > > /Nico > > > _______________________________________________ > bind-users mailing list > [email protected] > https://lists.isc.org/mailman/listinfo/bind-users _______________________________________________ bind-users mailing list [email protected] https://lists.isc.org/mailman/listinfo/bind-users

