Want to implement HA and loab balancer cluster and connect to data server ( SVN repo)

[visu4582]

Hi All,
I am planing to implement clustering system and for that i was
referring a lots of site. I got good knowledge from the site and based
on that i decided my requirement of clustering. I need HA and load
balancing combine clustering.
I got one good link of Ultramonkey and one report from VGLUG group. I
read that report. Now i have a couple of questions:
Before i ask i am giving you brief idea on what i want to implemnent.
Clinet
|
|
----------------------------------------
| |
| HA with LB |--> route client request on any of cluster server
|-----------------------------------------
|
|
----------------------------------------
| |
| cluster Servers | ----> SVN application or SVN web service
|-----------------------------------------
|
|
----------------------------------------
| |
|RAID server | -> Data storage of SVN
|-----------------------------------------


Above is my planning to implement and all i need from open source
community.
a) Shall i do this way?
b) Can you guys help me fast?
c) Any suggestion or good link for implementation to this type of
repository system requirement
Any quick response would be appreciated.
Regards,
Vishal

[Hlingler]

Hello visu4582, and Welcome to the forum.

This thread has been moved to "General Support" because it is neither a Guide nor a How-To.

FedoraForum.org Guidelines

V

[sej7278]

high availability for a subversion repo seems a bit over the top, how many commits you expecting?! if its opensource i'd just go with sourceforge or google code.

ha+lb can be done easily with just nic bonding.

clustering could be done using round robin dns with the servers pulling the subversion repo from an nfs share on the raid array.