Can one mirror a web server ?

[Kevin Darcy]

Warron French wrote:

>Well, if the servers are all on the same network, why not create a =
>Virtual IP and bind DNS together for all those IPs that host the copy of =
>the website and configure all the webservers in apache to use the local =
>IP address?
>
>Then point DNS for the website to the Virtual IP address?  Can't that be =
>done?
>
It doesn't usually work so well to configure multiple devices on the 
same subnet with the same IP address (ARP issues, etc.). You need 
something frontending those webservers with a VIP. Cisco calls their 
device a "local director", but they're not the only ones peddling the 
technology...

                                                                         
                                             - Kevin

>Warron French
>Sr. Network Engineer
>Xtria, LLC
>8045 Leesburg Pike #400
>Vienna, VA 22182
>Desk: 703-821-6110
>Main: 703-821-6000
>Fax:  703-827-0374
>
>
>-----Original Message-----
>From: bind-users-bounce at isc.org [mailto:bind-users-bounce at isc.org]On
>Behalf Of Jim Popovitch
>Sent: Wednesday, February 02, 2005 4:17 PM
>To: barrett bonden
>Cc: comp-protocols-dns-bind at isc.org
>Subject: Re: Can one mirror a web server ?
>
>
>On Wed, 2005-02-02 at 15:53 -0500, barrett bonden wrote:
>  
>
>>Can one mirror a web server ?
>>=20
>>The idea is to make a  failsafe  and transparent dual computer system =
>>    
>>
>at two
>  
>
>>sites.
>>=20
>>This would involve (can it be done ?) one dns setting point to two ip
>>addresses ?
>>    
>>
>
>Yes/No.
>
>Yes: It can be done such that www.yourdomain.com resolves to multiple IP
>addresses.  Do a `host www.google.com` for an example.
>
>No: What you will get with multiple A records is... well... multiple A
>records.  One user will get the first, the next visitor will get the
>second, the third visitor will get the third.  This achieves
>load-distribution, but not failover.  In the event that one IP goes
>dead, that A record will still be distributed and the "lucky" visitor
>will get a page timeout.
>
>Question for Bind developers:  Wny couldn't a feature (defaulted to off)
>be added that allows for a simple ICMP test before a request is
>returned. Yes it would slow things down, but it could allow for
>rudimentary failover.  Just a thought.
>
>-Jim P.
>
>
>
>
>
>
>
>
>
>
>  
>