Now if you know me, you’ll know I love linux, well on the server at least, for a webserver that serves customer sites, it’s fantastic. Ok with any linux setup you need to know what you’re doing, but when you do, it’s very easy. What’s not so easy, is high availability, or failover. It’s possible using haproxy and keepalived, but it just wasn’t working that well for me, for some reason it kept keeping both machines as master, but that’s a story for another time.
I run a few servers as virtual machines, how many kids these days want access to a minecraft server, well mine does that’s for sure. So I decided that I’d play with visualization on a server that I have. I’ve tried KVM using RHEL/Centos and it worked really well, I tried Ovirt but just couldn’t get on with it. So picture my server, I have 4 virtual machines running on KVM, ok lets try HA…..er oh hell. I say that as I did my research and found multiple tutorials on how to do it, but each one involved editing this file, adding this, editing another file. While I can do this, I just don’t want to spent umteen hours configuring things via text file. If anyone knows an easier way then please let me know. Currently I only have one server, so HA isn’t possible for me at the moment, but I wanted to set it up ready for when I do. As I usually do, I got frustrated with Linux and thought I’d give Windows 2012 server ago, and quite frankly I’m glad I did.
Install Windows 2012 server – check.
Install hyper-v as a role and configure all my virtual machines – check
Install Active directory and DNS and DHCP on a virtual machine – check
Join the Hyper-v server to the domain – check.
Wait till I can afford my second server, so I can pretty much repeat the process, then setup my replication, which from what I’ve been seeing, is a tick box [ok maybe a little more involved]
No text editing in sight.
Now don’t get me wrong, I still love linux, and my webservers will always use it, but for other things, and in this case it’s high availability and clustering and failover, imo MS systems are a lot easier to setup.
No comments on how it’s not free or opensource as that’s totally irrelevant to me. I don’t care, if it works, I’m prepared to spend money.
So what’s the aim of doing all this? Well in my case it’s two fold, I want to learn it as it’s cool. and secondly I want my webservers to have HA and failover support, if one goes down I want the other to take over, I still have customer sites I need to keep up as much as possible.
Probably next year, the lightbulb will switch on and I’ll get a HA Linux book and the aha moment will come and I’ll switch back, but for now, I’m happy with the way things are.