Linux - SoftwareThis forum is for Software issues.
Having a problem installing a new program? Want to know which application is best for the job? Post your question in this forum.
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
I am looking for a reliable, low cost backup / failover solution for my 2 colocated dell webservers.
We have 2 servers, COLO1 is running SUSE9.0, Plesk7.5 and manages the client accounts. COLO2 is running REDHAT, Plesk7.5 and manages our business websites.
Our hosting company wants to make some offsite NAS space available, and for us to backup over to that. In event of failure this would mean re-imaging the server with a current snapshot and taking the most recent plesk dump and associated files. Our coders arent too happy with this, as they have a LOT of custom applications running on the server, and not sure about the best ways of sending these over to the NAS storage.
Their solution is to install a 3rd colocated server with 3 partitions, 1 running a linux distro and used for configuring, the 2nd partition would be a mirror of COLO1 and 3rd a mirror of COLO2, using RSYNC or similar daily to mirror / copy over the drives. In event of failure, boot order can be changed from partition 1 to 2 or 3, and with some IP fiddling, should bring a mirror of the broken server back online.
My question, is this possible or unlikely to work? Obviously reluctant to spend on a brand new server, only to find it does not work.
Plesk Licensing is also an issue, but they are considering moving to Webmin to avoid the issues.
Does this sound any good, or is the NAS storage a better, safer route, OR is there a better way?
First thing that comes to mind would be to rearrange the current two boxes mirrorring eachother and using virtualisation with two guest OSes one running account mgmnt and one for websites. Second thing comes to mind would be to carefully determine what scenario's you need to provide solutions for (data protection, backbone outage, attacks, HW failure, etc, etc.) before talking solutions. This paints a realistic picture of what you need to cover after which you can research to how best spend your money according to priorities.