In order to facilitate rapid recovery in disaster scenarios, SpinUp highly recommends setting up a backup solution for your environment. A backup solution allows you to rest easy, knowing your critical data is secure even in the event of Cloud Server failure. Continuing reading to learn SpinUp’s recommendations on backup strategy, as well as a few options for backup systems.
There are two primary schools of thought when it comes to backing up your environment. These are known as file-level backup and block-level backups. These both have pros and cons, but SpinUp does recommend a file-level backup for a variety of reasons. This article explains both options.
Regardless of which style of backup solution you employ, there are some universal recommendations you should follow. First, you should not store a finished backup on the same device that is being backed up. This is somewhat self-explanatory, because if there is an issue with the device, the backup may be affected or unavailable, which defeats its purpose. For this reason SpinUp recommends storing your backups on other devices.
Second, an untested backup should not be considered a true backup. For this reason, regardless of backup strategy, SpinUp recommends testing your backups in order to ensure their efficacy. No one wants to be stuck in a situation where the data has been lost on the originating device, only to find out the backup is also not viable to restore from. Testing your backups enables you to identify such situations before they happen and take appropriate action to avoid them.
A file-level backup is a backup solution that backs up individual files from the originating device. This solution has a few key advantages, which is why SpinUp recommends this style of backup strategy. File-level backup enables you to pick and choose the items to be backed up, which reduces the size of the backup and ensures that only what you need is backed up. This is an advantage when you need to minimize storage costs and space.
File-level backups are also more flexible than block-level. Because each file is backed up individually, they can be restored individually, as well. This means you can preform restores on specific files that may be affected by an issue, without having to roll the entire device back to a previous state. This can accelerate recovery times in the unfortunate case that a backup becomes necessary, as well as limit lost data or records.
This flexibility also extends beyond selective backups and restores. Many file-level solutions allow you to configure different backup intervals for different files. This means you can backup critical and rapidly changing files often, ensuring minimal loss of data should a restore be needed, while simultaneously limiting the frequency and space taken up by less critical or less dynamic items.
Block-level backups can also be referred to as system-level backups. This style of backup makes a copy of the entire disk. This can make setup of the backup solution simpler as well as simplify the process of restoring, but comes with a few drawbacks.
The first is pretty self explanatory, which is that all files on the disk are part of the backup. This means that in the case of something like a virus, some other form of malware, or even something as benign as a configuration mistake are copied into the backup. Any restores done from that backup will also contain the problem. This is known as the all-or-nothing problem with block-level strategy. This also means that even if only a few files need to be restored, the entire server must be rolled back to the time the backup was taken.
The second drawback is that all-or-nothing also applies to the backup’s size. A block-level backup is a copy of the entire disk. This means the OS, applications, and all other files are along for the ride, which bloats the size of the backup. This can mean increased costs for storage, as well as longer times to perform backups or restore from them. This can mean longer downtime when you need to do a restore.
The all-or-nothing nature of block-level backups is why SpinUp recommends a file-level solution over a block-level solution. However, any backup strategy is better than none, and both methods can be used in tandem if desired.
Backup considerations for file-level solutions
When using a file-level solution, you need to know what you should and should not back up. The general mentality here should be to select files that are unique to your solution or environment and to not bother with generic files. This means backing up items such as your application or site content, configuration files, user generated content, and a dump from any applicable databases. It also means avoiding easily replaceable things like the OS, the applications themselves (such as Apache or MySQL), and system files. Following these recommendations minimizes the size of the backup, reduces storage costs and system overhead to perform backups, and speeds up restore time.
SpinUp recommends a couple of different file-level backups, which you can find links to below. If you want to use a block-level backup, you can use the snapshot system built into SpinUp Cloud Servers. However, we also recommend reading the following snapshot best practices before using this system.