the best option for web hosting redundancy

N

nservices

Verified User
Joined
Dec 11, 2008
Messages
302
Hi,
now i stand to add addition server for hosting and i want to know
what is the best option for web hosting up-time and load redundancy
1. install 2 dedicated servers with Multi Server Setup via DirectAdmin Panel
2. work as 2 VPS with Cluster_configuration (Proxmox VE for example)
http://pve.proxmox.com/wiki/Cluster_configuration_(Video)

wait for your answers,
nservices.
 
Of the two, using a cluster is your ONLY option, since DirectAdmin Multi-Server Setup only creates DNS across multiple servers, nothing else.

Jeff
 
there is Another alternative ?

there is additional alternative Except for the above options?
 
Just use rsync if you wanna mirror your stuff over. But there is gonna be no true failover unless you use something like vmware.

Wow thats the worse video ever. Sounds like he is in his cave...haha
 
@scsi:

Please explain how you can get true failover (no single point of failure) with VMware; enquiring minds want to know.

And please explain how rsync is going to solve problems such as database redundancy.

If there's an easy way to do this without spending lots of money on datacenter redundancy and special hardware, I'd really like to know, because obviously even the large companies such as PayPal, Google, etc., haven't figured it out.

Thanks.

Jeff
 
And as I said in my first response that clustering was the only option. There's a hole in the bucket.

Jeff
 
now: Practical recommendation

i have hosting company and i work with directadmin as panel in Linux servers (CentOS 5.3 x64)
and Plesk as panel for Windows Servers (2008 web server x64)
now i don't have any redundancy for server fail
i just buy new "monster server" just for redundancy
i want to work with VPS on this server and create virtual server per Existing server
Gradually transfer the existing servers to the virtual server configuration to implement the possibility of redundancy

I'm deliberate whether to use
VMware ESXi or Proxmox VE (if you have More choices, I'd love to hear them)

final result at the end of the process, in the case of server fall / Overload,
the other server continues to work as usual
and the customer does not feel any problem (Transfer of information between the servers should be bi-directional)

Best Regards,
Star Network.
 
Last edited:
You must log in or register to reply here.
Back
Top