Server Maintenance Part 1: Replacing a Drive

I acquired an old tower server from a source that upgraded hardware. Instead of disposing it, I requested to bring it home since it was still running. It’s a great device to use for home lab use. I also wasn’t going to pass a free server (saving money anyway you can). There was some maintenance needed to be done on this device so i figured I document it.

Dealing with Failure

I ran into the first issue which is a hard drive was failing. Luckily, the server, much like many today, implement RAID-like systems so the server doesn’t fail completely. Though it is a warning to deal with the issue immediately since there are only so many failures a device can handle.

The problem was I needed to find the right components. In the enterprise world, you would typically pay for support or have drives in stock to replace failed drives. As a normal consumer, those services weren’t at my disposal. Luckily, Google is pretty handy. It showed me a few sources to look at. It’s important to note, that you need to be precise in what you’re buying. Take a look at the device and obtain as much information as you can. This will make it easier in buying the correct components. I know this sounds self-explanatory, you would be surprised at how easily you can mess this up. There are so many pieces of hardware for many types of servers.

Truth be told, I messed this up. I bought a drive from eBay but the drive was for a rack mount server of a different model. I had to return it which was disappointing given the time it took. After some time, I did eventually find the drive. To cover my bases, I bought an additional drive from the eBay vendor. i’m not sure how often I will have to replace drives but I will at least be prepared for the next one.

Implementation

the green lights on the drives are all green and flashing. They are working on adding the new drive to implement into the RAID.

This part is quite simple. The server allows hot-swapping drives (within the RAID array). This means I can pull out the failed drive and put in the new one without having to turn off the server. Once the drive is in place, it will be working with the other drives to “fill in the gap” of the data to store on it.

Takeaway

It’s important to be patient with this process. It took me a month of time in fulfilling this. I could’ve accomplished this sooner, but I kept doing research on this to experience the least amount of failure possible. It was also relieving to have this resolved for now.

Two reasons for buying an additional drive, are the following. My first reasoning stemmed from advice I learned on the job. That was whatever you’re buying, do n + 1. This is applicable when buying equipment for inventory to out-weight for random scenarios. The second reason