Monday, February 15, 2010

Back to VMware

It's been a while since I updated this page. Work and social life has been taking over also starting school this year. I've been tasked with a few projects for the current job where I'm focusing on VMware issues with ESX server. Currently there are no major problems but I wanted to have a solid plan for the future. The biggest hurdle is the smaller servers in the data center.

Many companies have small older servers in the data center for one application, typically something that needs a database or hosts a web site internally. Here at my current job location there's a few servers and they all use very low powered servers, dating at least a few years old. These are prime cadinets for physical to virtual migration. This would also help out getting away from old unsupported hardware. Who knows where to find a power supply for a off built server chassis which the company no longer is in business.

The problem going with the virtual solution is how can I keep the servers up when the ESX server it's self has a problem? I'm currently reading and studying with vSphere, hoping to have a project plan written up soon. Ideally I would like to a system planned out where I can have redundant fail over ESX servers, allowing us to make sure if one system is down does not take down the entire virtual systems.

It's going to take some planning, and understanding of the environment. I'm really aiming towards the simple method, making it the easiest to understand and manage. Considering the environment, this would save space and energy, not to mention support issues since there's just less hardware to fail.

Having been in the role where I was supporting very old hardware, some non-hot swap drives in a RAID 1 configuration, it was extremely painful to take down the server just to swap the drives. Not to say that that's IF you are alerted to the drive failure before they both went down.

No comments: