So right now, we’re in the course introduction. We will be going through Backup Exec Fundamentals. That is really a very high level overview of the product. It’s the 30,000 foot overview. And then we’re going to talk about installing Backup Exec. Getting it installed is pretty easy. In this chapter, we will also talk about upgrades, and I know several folks are upgrading from 2010. So we’ll talk about the things that you have to consider before you do that upgrade. Then we will talk about storage devices. Where do you want your backups to be written to? So storage devices can be hard drive space, they can be tape, they can be removable disk cartridge, so we’ll talk about the differences in how all of those work and how do I set them up.
Then we’ll talk about data management. In older versions of Backup Exec this used to be called media management, and everything in older versions, Backup Exec thought everything was tape. That has changed in this version. So everything that’s being written to disk now has what’s called data lifecycle management. So we’ll go through the differences between what’s written on disk and what’s written on tape and how Backup Exec manages that and how you determine how long you would like to be able to restore the data that you have backed up. So once I have my media sets set up, then we can talk about backing up data and I have to backup data to a destination, a storage device, and say how long I want to keep it as part of that backup. But there are several other questions you need to answer during the backup. So we’ll go through all the questions you need to answer and how to get your backups to work properly. And then we’ll talk about restoring data. Really, this class is all about restores. Backups are just a way to get to restores. So restoring is the important part. This is why you do backups everyday. This lesson is really very short, but it’s probably one of the more important ones. Then we’re going to talk about simplified disaster recovery, so how do I run my backups in such a way that I can recover a server from the dead, from bare metal, as easily as possible. This is new to this version of the product. In older versions it was called Intelligent Disaster Recovery, but it has changed significantly with this version of the product. So there are certain things you have to do in the backup in order to be able to recover it, and then we’ll take a look at what those are, and then how do I recover, so how do I bring the machine back from the dead. The next lesson is virtual machine conversion. This is also new to this version of the product. Virtual machine conversion allows me to run a backup and either at the same time or when you schedule it, convert that backup to a virtual machine. So I can do it as the backup runs. I’m backing up the data to my hard drive somewhere and converting to a virtual machine at the same time. So you can convert to a VMDK or a VHD. Or you can do this immediately after the backup, or you can schedule it. So I want to run my backups at night and do my conversions during the day. So we’ll go through the different rules for doing virtual machine conversions. There are two types of virtual machine conversions. They are physical to virtual, which I think most people are familiar with, and then backup to virtual, which is you can do it at any given point. So if I’m running backups on a regular basis and my server goes down, then I can right click on any of my backups and say convert that to a virtual machine. So I can have that machine up and running very quickly as a virtual machine strictly from backups. The last lesson in this section is database maintenance. This is backing up the Backup Exec stuff. This is arguably the most important lesson in this class. If your Backup Exec server dies and probably because you whole data center, something happened in flood or fire, before you can recover anything else you have to be able to recover the Backup Exec server. So this is what do I need to backup so that I can recover the Backup Exec server. And then you can begin to do your other recoveries. An interesting question, if you’re looking at these lessons, we’ve talked about simplified disaster recovery and bringing a server back from the dead. The question becomes couldn’t I just do a simplified disaster recovery of my Backup Exec server, and the answer is which came first, the chicken or the egg. You get yourself into a catch 22. I have to have Backup Exec in order to do a simplified disaster recovery, but if my Backup Exec server is dead then I can’t do a simplified disaster recovery of it. So the answer is no. So what do you have to backup in order to be able to recover this server from the dead? That’s the end of the first section of the course. This will probably take us through Wednesday, late Wednesday, possibly Thursday morning.
Manage and Administer
The second section of the course is known as Manage and Administer, and in that section of the course what we will cover is remote agents. Remote agents, there are two major ones. They are the remote agent for Windows, and the remote agent for Linux. There is also a remote agent for Mac, but it is really covered under Linux. Macintosh is really a version of Linux under the hood. So when we talk about the remote agent for Linux, all those rules also apply to the remote agent for Mac. The next lesson is an agent for applications and databases. This is really databases 101. I don’t intend to make anyone a DBA, that’s not the intent here. The intent is that you have to understand how databases work in order to back them up appropriately. Why do I have to run the backups that I do. So we’ll talk about how databases are structured, and how you back them up appropriately. And then we will also talk about the agent for Active Directory recovery. And the agent for Active Directory allows me to recover individual items out of Active Directory with their original GUID, so that’s very important. The next lesson is how to backup SQL. For every backup that you do for a SQL Server you should have at least two backups that touch that SQL Server. So we’ll talk about what those are and how they work. In the next lesson, the lesson about Exchange, what we’re going to talk about is how to backup Exchange 2003 and Exchange 2007 but then Microsoft changed all the rules for how Exchange 2010 works. So we’re going to have to take a very short overview of how Exchange 2010 works and how to back it up. So that one’s a fairly in-depth lesson. The next lesson is SharePoint. SharePoint is really a glorified SQL Server with an IIS front end, so we’ll talk about the different pieces. One of the challenges with SharePoint is that SharePoint is a distributed installation meaning that I can have part of it installed on a Web Server, part of it installed on a SQL Server and part of it so that the documents and things that you’re sharing are on a user server somewhere. So SharePoint can be installed in many different places. Backup Exec is smart enough to just go out and find all the pieces and parts of SharePoint and back it up. So we’ll talk about how to set it up so that you can get that automated — just go find my pieces of SharePoint and back it up appropriately. The next lesson is virtual environment agents. There are two of these. This is really a three part lesson. It is an overview of virtualization and how does virtualization work. And then we’ll talk about the agent for VMware, and how does the agent for VMware back up. It backs up everything in one pass. You get the VMDK, you get everything that’s inside the VMDK, and you can still do granular recovery, so I can restore individual files and folders, or if it is a virtualized Exchange server I can restore individual e-mail or mailboxes. The last part of that lesson, the third part of that lesson, is the virtual agent for Hyper-V. Hyper-V works slightly different than VMware, so we’ll take a look at some of those differences and the differences in the rules of how they work. The last lesson that we’re going to cover is deduplication. Data deduplication is new to many people, so if you haven’t heard of what it is, in a nutshell data deduplication is a fancy form of compression. Let’s take this PowerPoint presentation just as an example. If I were to e-mail this PowerPoint presentation to everyone in the room, then we would all have a copy of it, and we would all have that in our user directory. How many times do we need to back it up in order to be able restore it? Do I need to back it up four or five or six times? The answer is no, I only need to back it up once and I can restore it. But data deduplication is more than that. It is looking at the individual pieces inside of a file. So if I e-mail this PowerPoint out to everyone and only one person changed a couple slides, we would backup the whole PowerPoint once, but then we would only backup those chunks that change, the deltas of that slide presentation, so just the slides that changed. So data deduplication is a way of backing up the data so that it occupies the least amount of space possible on the destination. It’s extremely efficient. And the idea is that if I’m using hard drive space for backups, rather than only keeping three or four weeks worth of backups in that drive space Because data deduplication uses this space so efficiently, I can now keep two or three months worth of backups in that same amount of drive space. We’re not costing you any more drive space. So what are the rules for data deduplication? How does it work? And how do you set up your backups to use it? And there are several different rules to that. So that gets us through all the lessons we have for this week. You can tell it’s going to be a very busy week, and we have a lot of new material to cover.
So after completing this course, you will be able to describe the basic functions of Backup Exec from a very high level overview. The components of Backup Exec; you’ll be able to manage your devices and more importantly manage how long your backups are kept for you to be able to restore, and there’s two parts to that. There’s the media set and there’s data lifecycle management. You’ll be able to run basic backup jobs, these are known as backup definitions, and you’ll be able to create new ones and edit the ones that you have. You’ll be able to explain how backup and restore workflows work. You’ll be able to restore — the most important part of the class — you’ll be able to restore your data. You’ll be able to list the different virtual conversions that Backup Exec is able to do, that’s P2V and B2V. You’ll be able to create a simplified disaster recovery disk. This is a disk that you boot to to recover your server, and you can put your own drivers on it if you need to. Most people don’t, but you can if you need to. So we’ll talk about how to do that. And then you’ll be able to use that disk to recover servers. We’ll also go through the new features and enhancements that have been made to Backup Exec, and we’ll list a few of the end of life items, not all of them, but we will go through a few of them. So those are all things you’ll be able to do when you’re finished with this course.
Data protection: Is backup or restore the priority?
So the question becomes why do we care, and the reality about backups is that most environments really don’t want to put a lot of time or effort into backups, particularly money, into backups until they have had a disaster. And once they lose data, all of a sudden backups become important. So when you look at the slide that you see here about data protection, what’s the priority, the answer is it restores the priority, but in order to restore very quickly you have to back it up in such a way that you can restore it very quickly. I’m going to give you a rule that we’re going to talk about all week, and that is the faster the backup, the slower the restore. The faster the restore, the slower the backup. Everything in life is a trade off. So if I’m backing up to tape, that’s my slowest backup, but it’s also my slowest restore. My tape is sequential access. There’s all kinds of things that are in effect now for very fast recoveries. I can do clustering, I can do a backup to virtual, so I can bring a machine up as a virtual machine, or I can do disaster recoveries of the whole machine. So the trade off here is I have my recovery point that I want to recover from, hopefully that’s as current as it can be, but then how long does it take me to do that recovery. And that’s where the longer the backup, the shorter the restore. If I did a full backup every night, that’s a much faster recovery. So I have my recovery point of my full backup and my time to recovery is very short if I had done full backups. This all takes planning, and that’s really your mission as a backup administrator is to understand these rules and understand which ones are faster and which ones are slower and then apply that to your servers as needed. Because the reality is that you may only have one or two administrators on your network, and if your data center is full of water and you have to recover the data center as a whole, which servers do you recover first? You’re one person. How many can you recover at a time? Who comes first? Who comes last? Which server can you stand to be without for three or four days? Those are all questions that you have to answer, and you’re not going to answer them all today, but they’re questions that you have to think about, and this is really our priority in this class is to get you to think about the different types of backups that you run, why you run them, and then how can you use those to your best benefit when you have to do recoveries.
Data protection plan
So you have to plan this out. Being able to recover is really more a function of how you backed it up. So your backup method becomes the driving force for how you can do recoveries. So you have to plan what kind of a recovery do I need to do and I have to run my backups accordingly. Then the next thing that comes into play is how long are you keeping your data. How many people really restore backups that are a year old or two years old? Most people restore backups that are less than 60 days old, so do you really need to keep your data for long periods of time. Well, for some legal requirements the answer is yes, some data has to be kept forever and ever, amen. Some data you keep for a very short time, maybe two weeks at a maximum. So how do I store my data to make the best use of my resources, and one of those methods is called archiving. And that’s where I can take the data off of the source where the users access it, let’s say for example Exchange, and so the users have their e-mail on their Exchange server, but users keep e-mail forever. And so I’m going to say as an administrator that they can only have things on that Exchange server that’s less than two years old. And once it hits two years old, we’re going to take it off of the Exchange server and put it onto drive space that something else manages, maybe Backup Exec. And so now I’m going to pull it off the Exchange server and put it onto drive space that Backup Exec manages, and if the user needs it back we can get it back, but it frees up space on the Exchange server. That’s known as archiving. So all of these data protection and recovery things require planning, and part of the planning on this is what kind of service level agreements do you have. Service level agreements say how long can a server be down. Some of you have written service level agreements that say this server can only be down for one hour or two hours and it has to be back up and running. Some service level agreements are implied. Take down your Exchange server and tell me how long it’s going to be before users are howling at you that the Exchange server is down, right. They’re going to be calling you on the phone, they’re going to be knocking at your door, I can’t send e-mail, I can’t send e-mail. What’s your service level agreement for that Exchange server? Well, if you didn’t have one written, I will tell you it’s about 30 seconds, if it’s that long. So you’re going to have to plan your backups accordingly so that you can recover your data as quickly as needed for your business, Exchange and SQL being two of the big ones. So this all drives your backup strategy. The last piece of this that’s very important is that you’ve got to test. The only way to find out that what you backed up is truly valid data is to do restores. If you’ve never done a pretend drill where you pretend that a server is dead and try to recover it, it’s a very useful learning experience. Because it’s easy for us to sit here and be armchair quarterbacks with our cup of coffee and say when a disaster happens this is how we’re going to recover, but it’s a very different thing to actually do it, and walk through. And a question is, can you? And I think what you’ll find is that with certain types of data it’s much more difficult to do those recoveries than it is with other types of data, SQL being one of the more difficult. SQL is also one of your more mission critical because this is the one usually that holds all of your billing, all of your money coming in and money going out is in a database. So if you can’t recover that one, what happens to your organization? Often people will say but I don’t have the time or the resources to do recoveries. With this version of Backup Exec you can take that physical box or that virtual machine and back it up and recover it as another virtual machine. So just to make sure that you know how to do your recovery. So it’s a very easy and efficient way to do testing to make sure that you can bring your data back. And you thought this was a backup class. This is really a restore class. Backup is just how you get there.
For our lab environment, what we have is a physical host that’s running five different virtual machines. We have a domain controller that is running Active Directory, SQL, and SharePoint. It is a 64-bit 2008 R2 box. Then we have a backup server that is also 2008, 64-bit that is our backup server and our Outlook Client for Outlook 2007. We have two Exchange servers. One of them is the major Exchange server, the other one is a member of the data availability group, database availability group, or DAG, and that’s the one that we’re going to have to talk about Exchange 2010. Then we also have an ESX server that is also hosting guests so that we can backup virtual machines. So that’s our lab environment, and I’ll go through that more when we break to do labs.