Backup and Recovery

In this video, we’re talking about backup and recovery. So unfortunately, the worst happened. You had the virus outbreak on your network, a Denial-of-Service attack that ends up corrupting data, someone unleashes ransomware on your system, a USB drop delete certain files out of your system, or some other disaster on your network. Well, data backups are one of the easiest and fastest way to recover from a disaster. Say your data server catches on fire. Your MDF catches fire, your servers are lost. Again, you get ransomware. Someone goes through and starts wiping out important data. Again, having a data backup, you can restore that data provided you have an offsite backup or a backup in a safe place.

You can quickly and easily recover your data. So part of what you need to do for recovery is you need to identify, track, and plan. So say, if our network falls victim to ransomware, file deletion, hardware failure or other catastrophes, our data will need to be recovered quickly.

Now in data loss, we’re talking days, weeks, or years worth of data in some cases. Some of this data is going to be irreplaceable information. We could be talking about photos, documents, database information, client records, financial records, et cetera.

So there’s a few basic rules for backups. So incremental. Incremental means - it’s a type of backup. It saves a lot of time because we’re backing up certain things. Say if we run an incremental backup - say on Monday, we’re only backing up people’s data shares. Tuesday, we’re backing up, say, our important servers. Say server one, two, and three. Wednesday, we’re backing up server three, four, and five. And so on, and so on. So we’re backing up certain portions of our network. Or we can be just doing incremental backup where we’re backing up changes that the network sees from day one to day two, and so on.

Now these type of backups are efficient in the fact that they’re a lot quicker than doing a full backup. The downside is, if something happens, well, we could potentially recover that data because it’s, again, incremental backup. The problem is if something major happened, like a ransomware attack, everything gets locked or all the files get deleted. Well, then an incremental backup will probably not be able to give us all our files back. It would give us back whatever backed up that day. This is where a full backup comes in handy. So a full backup will back up everything, but this is a time-consuming process.

And depending on which data you have, this could take all day or night, or it could take several days or weeks to back up.

The next rule is the rule of three. So it states that you should have at least three copies of your backup on three different types of medium. Meaning, say, one of them could be on a data server. A bunch of hard drives in your server room. A second backup could be - well, offsite backup. Again, in a physical location. A NAS server, for example. And a third type could be cloud storage. Backing up to AWS, or Microsoft Azure servers, or even Google or whatnot. These are three different types of backups. And again, the idea behind three types as well, if one gets lost or corrupted, well, you have two more backups to recover from.

Well, backup two ended up being corrupt, or the media is no longer available. It’s obsolete, you can’t recover from it. Well, that’s why there’s a third different type of backup. Now the other rule is offsite. So you always want to make sure you have at least one full backup that is kept offsite in a secure location. And again, the reason for this is - again, say our building catches fire, a tornado hits it, there’s a fire or flood and whatnot. Well, that building gets wiped out, and if we have a backup in, say, the bottom floor, and then on the third floor and then the whole building gets wiped out, well, there goes our backups. It’s kind of pointless.

That’s why you want a offsite backup. Something happens to the building, you still have offsite backup. Again, say with Microsoft or Amazon, for example. And then once you get your computers back, replaced, you could restore your backup. Likewise, if something happens to, say, Amazon or Microsoft data server that happens to have your data, well, you should have your own backup at your place, so everything is still potentially fine. Now the last rule for Basic Rules of Backup is testing. And this is really important. And it seems like a no brainer, but it’s worth mentioning that you do want to test your backups.

So unless you test your backups, and you know for a fact that these backups are good, then it’s going to be worthless. Because when the time comes - you’re backing up, you’re backing up for years, you think you’re good, ransomware attack hits. You go restore your backup and you find out, wow, this did not back up what I thought it did. Or this backup does not work at all, because we didn’t back it up properly. Well, that’s when you really need to know that your backup is working. So you do want to try to restore periodically or validate that information to make sure that it is backing up properly.

Another good thing about testing your backups periodically, other than verifying the data integrity, is that also that you’re trained to actually go through the motions and restore the data. So when the time comes, even if you have a good backup you never restored it before, you don’t want to be sitting there for days, hours, weeks trying to figure out, well, how do I restore this backup now? And potentially damaging the data or wasting more time. Time is money when you’re down, and you do want to get things up and running very quickly and efficiently.