|
I've used lots of implementations of LVS with Win2K no one has ever
mentioned any bandwidth problems.
I don't think the loopback is even used it just lets IIS accept the
packet when it comes in on the normal NIC.
It should be very easy to test.
Regards,
Malcolm Turnbull.
Loadbalancer.org Limited
+44 (0)7715 770523
http://www.loadbalancer.org/
" When a single point of failure is not an option"
Why not try our online demonstration
<http://www.loadbalancer.org/demo.html> ? Or get answers to common
questions <http://www.loadbalancer.org/fud.html> ?
Monty Ree wrote:
> Hello, all.
>
> Dear Joseph answered like below about my question.
>
>> the traffic which the each real server services is over 10M bps.
>> So, 10 Mbps can cause any problem or not?
>
> --> well the clients will have to wait.
> I understood this answer that "Windows RealServer can't service over
> 10Mbps at LVS-DR"
> right or not?
>
> But something curious problem...
> I guess that all traffic goes throught lo:0 interface of the linux
> realserver at LVS-DR.
> But when I execute ifconfig at the realserver, TX/RX of the
> eth0(external interface) is more higher than that of the lo(loopback
> interface).
> and when i execute tcpdump, at linux realserver(LVS-DR),
> I can't see any traffic regarding to lo interface in spite of the many
> connections exist to the realserver.
>
> # tcpdump -i lo
> So, I concluded that bandwidth or speed of the loopback interface has
> nothing to do with the service of the realserver. Am I wrong or not?
>
> Anyone who use Windows(NT or 2000) for realserver and LVS-DR?
>
>
> Thanks in advance.
>
> _________________________________________________________________
> 행운의 주인공이 이번엔 나일꺼야, 진짜루... 인터넷 복권
> http://www.msn.co.kr/money/interlotto/
> _______________________________________________
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users
|
