When a server runs out of disk space bad things happen, usually resulting in downtime and even data loss. For years, to avoid this scenario we have personally reached out to customers asking them to reduce disk usage, and when required manually upgraded disks. This is time consuming, expensive and inefficient, so we have automated the process.
Below we'll explain how the two steps in the system work, so you know when you'll be notified and at what point your server's disk will be upgraded. Note, in many cases disk upgrades are permanent, so it's best to deal with the problem early.
When our system sees your server's disk space drop below a healthy threshold (in most cases this is below 5GB or 10% – whichever the lesser), you will receive an early warning about your servers disk space. The aim of this message is to allow you time to investigate and determine at your own pace, whether or not your server might require an upgrade.
If our system sees your server's disk space drop below our emergency threshold (usually below 5% or 2.5GB – whichever is the lesser) and it has been at least three hours since the warning was emailed the system will automatically add 5GB of disk space to your server.
There's no downtime involved as we only do this on servers that support live upgrades, and we will only automatically do this once every seven days, to avoid cascading upgrades.
For those of you reading closely you have probably realised there are situations where our system will take no automated action:
In these cases your disk health will be escalated to our traditional, and more manual system. As your disk usage may be increasing unexpectedly fast there could be an underlying issue requiring investigation. This sort of problem is better suited to be dealt with by one of our experienced team members.