I'm pretty sure you can disable sync requests on startup using the registry. I can't
remember how off-hand - I lost my Notes From the Field book. There's a great section
in it specifically on tweaking bridgeheads. I think some of your problem is the mesh
configuration, though. It makes it harder to understand and control your replication
scheme. Where I worked previously, we had 210 sites that all needed complete
directory replication every 24 hours (that was our OLA). Because we had a multiple
level hub-and-spoke, we could easily collapse and expand the replication once per day
during off-hours and still ensure that everyone got all the updates. It took lots of
resources and we found that the theoretical 64-server per bridgehead limit wasn't
feasible on the hardware we had so we limited to 35 connections per server.
Enterprise-wide directory replication is not something to be configured as an
afterthought. As you're finding out, larger organizations can very quickly
see the bulk of their resources go to replicating instead of actual message delivery.
Pare down as much as you can - if you can find the Notes from the Field book I
recommend it. Consider hub-and-spoke.
I know there isn't any real solid technical information here, just advice. But I
learned the hard way - most of my job had to do with managing replication, and it was
much more challenging than the other aspects.
Eric
On Mon, 17 Sep 2001 14:10:17 +0200, "Exchange Discussions" <[EMAIL PROTECTED]>
wrote:
> I have finished the "roll-back" during weekend and the central hubs returned
> on normal performance.
> But after "midnight off-line backup" (when services are restarted), each
> server generate PubIS sync. messages to all their neighbours (in sum cca
> 20MB per server)
> Does exist some tuning method to decreasing (removal) this traffic after
> restarting MS Exchange IS?
> Has somebody similar experience with so large organization?
> mk
>
> -----P�vodn� zpr�va-----
> Od: Kotas Martin [mailto:[EMAIL PROTECTED]]
> Odesl�no: 15. z��� 2001 12:06
> Komu: Exchange Discussions
> P�edm�t: RE: I have big trouble with site replication - Please Help
>
>
> Yes - all DirRep schedulers on all spokes are set to Sunday night.
> I think that EXS 5.5 distribute routing information by RIP algoritmus and
> this is not good for "Mesch" architecture. (all sites are conected directly
> - without any hops)
> GWARTs are swelling and the number of routing information increase.
> So I must roll-back the configuration to Hub-&-Spoke architekture. (DAMN...)
> mk
>
>
> -----Puvodn� zpr�va-----
> Od: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Odesl�no: 14. z�r� 2001 18:14
> Komu: Exchange Discussions
> Predmet: RE: I have big trouble with site replication - Please Help
>
>
> Have you scheduled the directory replication on BOTH SIDES of the DirRep
> connectors? Perhaps your bridgehead is only requesting updates at night,
> but your spokes are requesting 8 times per day (default IIRC).
>
> Eric
>
> On Fri, 14 Sep 2001 14:28:46 +0200, "Exchange Discussions"
> <[EMAIL PROTECTED]> wrote:
>
> > Our organization contains cca 200 EXS 5.5 connected by IMC thru WAN.
> > All servers replicate directory against one central bridghead server (star
> > repl. topology).
> >
> > After reconfiguring IMC (I add cca 200 connected sites to each server) -
> the
> > central server has overloaded.
> >
> > On all servers is replication scheduler set to night, anyhow central
> server
> > is generating and sending cca 300MB DSA msgs. per hour!!!!! during all
> 24h.
> > This situation persist 8 days!!!!!
> >
> > Can somebody help me how can I solve this (very very big) problem?
> > -How can I EXSs say: "stop this DSA replication"???
> > -How long this process will take?
> > -Why has EXS so big overhead?
> > -Why MS produce this shit????
> >
> > Thanks for all helps and hints.
> >
> > mk, MCSE2000
> >
> > _________________________________________________________________
> > List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
> > Archives: http://www.swynk.com/sitesearch/search.asp
> > To unsubscribe: mailto:[EMAIL PROTECTED]
> > Exchange List admin: [EMAIL PROTECTED]
>
> _________________________________________________________________
> List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
> Archives: http://www.swynk.com/sitesearch/search.asp
> To unsubscribe: mailto:[EMAIL PROTECTED]
> Exchange List admin: [EMAIL PROTECTED]
>
> _________________________________________________________________
> List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
> Archives: http://www.swynk.com/sitesearch/search.asp
> To unsubscribe: mailto:[EMAIL PROTECTED]
> Exchange List admin: [EMAIL PROTECTED]
>
> _________________________________________________________________
> List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
> Archives: http://www.swynk.com/sitesearch/search.asp
> To unsubscribe: mailto:[EMAIL PROTECTED]
> Exchange List admin: [EMAIL PROTECTED]
_________________________________________________________________
List posting FAQ: http://www.swinc.com/resource/exch_faq.htm
Archives: http://www.swynk.com/sitesearch/search.asp
To unsubscribe: mailto:[EMAIL PROTECTED]
Exchange List admin: [EMAIL PROTECTED]
