Using some software to protect your data? Here’s the biggest mistakes you can make, and what you should do if you’re currently making them.
1. Performing Manual Backups
Wanting to control the process is only natural. Whether it’s copy-pasting your data to a disk drive or performing an on-demand backup through some software, being hands-on is a bad idea.
Why? You’re introducing human error into the equation. It’s highly likely that at some point, you’re going to forget. The last words you want to hear are “I forgot to back up” when a virus runs through your server. Not only is it risky, it’s a massive (and unnecessary) time-sink.
If you’re manually backing up, it’s also likely you have no version control. E.g. The ability to roll back a file to a certain point of time.
What you should do: Use automated backup software. You can set a backup schedule that suits your needs. From there, it’s mostly set-and-forget, with the exception of occasionally reading job reports.
2. Only Backing Up To One Place
You’ve made a backup of your valuable data. Great! Have you made two? You might be asking “Why two? I’ve already got one!”
The reason is simple: your data is not really safe unless you’ve got two copies in two different places.
Why? There’s more than one reason, but the simplest to explain is that if the device you back up to – hard drive or external disk – is compromised in any way, your data is done for. Hardware failure. Physical damage, such as fire or flood. Misplacement. Burglary. The list goes on.
What you should do: Make multiple backups on two different types of devices to mitigate hardware failure. Make sure one of these devices is offsite, preferably a public or private cloud destination you can back up to. If you’ve got a copy offsite, then it is a lot harder for things like theft and natural disasters to put you out of business.
3. Only Backing Up Data, Not Systems
It can be a real time saver – not to mention a space-saver – to just back up your most crucial data. It requires less storage space, which is a big deal if you’re paying to store it in the cloud. And if your whole system goes down, it’s not too hard to wipe and reinstall, right? This is a bad move.
What you should do: You should perform a “complete” full system backup every week or so, especially if you’re backing up a server. A complete backup – also known as an “image snapshot” or “bare-metal backup”, doesn’t just copy your data – it takes a snapshot of your whole system. This includes your programs and OS.
There are a number of reasons this is a good idea. You can roll back your system to before a major incident. Your downtime is minimized. You skip the painful reinstallation process. You won’t forget any crucial data. And most importantly, you won’t invite compatibility issues by reinstalling an app and finding your saved data is now out of sync. E.g. Settings are now different.
4. You Only Focus on Backups
So you’ve set up your backups and you’ve got plenty of space left on your destination. You’re even backing up to different locations. You don’t have to lift a finger ever again.
Well, not quite. Sure, backups are a big part of your backup solution – but it’s recovering your data that really matters. What good is storing it if it can’t be properly brought back, right?
What you should do: Even with the best backup and DR solution, you need to occasionally perform a test restore of your data – that’s trying to restore some of it to make sure it was backed up correctly. Why? Because there are a number of things that can go wrong even if everything is set up right.
Even if you’ve got an automated backup verification system, don’t rely on it. The types of data you can back up are practically infinite, and a verification system can only look out for a finite number of issues on its checklist. The best way is to check yourself every so often.
5. Confusing Cloud Backup, Sync, and Storage
So you’re sending your data to the cloud. Great! But a lot of people use the wrong tool for the job. This can cause troubles, because you think that you have a greater level of data safety than you actually do. You can read more about the difference between Cloud Backup, Sync and Storage here.
Using a Cloud Sync solution (E.g. Dropbox, Google Drive, Microsoft OneDrive) is great for protecting home user data and sharing your data with others. The problem? It sucks when it comes to protecting large chunks of data and has zero version control. For almost any business, this is a deal breaker.
What you should do: If you’re protecting home data, nothing. If you’re protecting business data, stop using Cloud Sync or manual Cloud Storage (Copy-and-paste) and start using an automated Cloud Backup solution.
6. No Future Planning
We get it. Your boss wants to know that you’re making the most of that server space, and doesn’t want to spend a dime more. But when your business needs grow – and it will, if things are going well – then you’re suddenly in a lot of trouble. This is especially true of your backup destination.
If you don’t have enough storage space on your backup destination, the backup fails. And if you’re not checking your job reports regularly, this means you can suffer a data disaster only to find out you haven’t had a successful backup for the last few months.
What you should do: Plan for growth. Make sure you have space on your destination not just for your next backup, but all the backups you’ll ever need to store on that destination. Make sure you account for data growth. Adopt a scalable solution like cloud, and cold-store / archive your unused data regularly.
7. Leaving Things Insecure
Backups do not safeguard you against every single scenario, though they do cover a lot of them. If something like ransomware makes it onto your backup device (or gets backed up accidentally), your data is compromised.
What you should do: Give some thought to security when it comes to your backups. This can mean using a dedicated anti-ransomware solution to protect your backups such as CryptoSafeGuard.
8. Disk Rotation Scheme Too Short
Imagine this: you’ve got a two-week disk rotation scheme. It seems to work, but then you have a server fail due to corruption. When you try to restore from your backup, you find out that you’ve backed up the corruption – and because the corruption existed for some time, there’s not a single uncorrupted backup to be found.
What you should do: Use a rotation scheme that includes a monthly backup that is kept for 12 months, or at least keep some of your backup as long-term archives. This is another good reason for performing regular test-restores.
9. Not Having an On-Site Backup
You’re shipping your backups off-site using a courier service, or sending it to the cloud. These are great solutions, but you don’t want to be using this as your sole solution. Why? If you have a server failure, it’s going to be hours (or more) before you can get the whole thing back.
What you should do: Have an On-Site backup so you can restore servers quickly. Alternatively, make sure you don’t ship those backups off-site too fast just in case something happens.