My come to religion moment with backup happened in the late 1990s. I was working in my second IT-related job, as the webmaster and a jack of all tech trades, for a small San Francisco Bay-area start-up. Our one web server, the only source of all of our data, had come up lame. And I needed to fly out immediately and race down to a co-location facility in San Jose to fix the problem.
Long story short, a co-worker and I spent the entire night there troubleshooting the server. We slept, awkwardly, in our boss’s VW Jetta when we finally realized at some awful hour of the night that we needed more parts. A trip to Fry’s Electronics and about $2000 later, we were able to fix the problem and do so, delicately, without losing any data. But we were spooked. And as part of our fix, we finally did what we should have done in the first place. We invested in a tape backup. Hey, it was the 1990s.
But that is only part of the story. The final irony here is that the tape backup we bought never actually worked. A few years later, when we dismantled that web server—a $5500 Dell with a Pentium Pro 166 processor—I decided to use the tape backup elsewhere. In doing so, I discovered that the tapes we had backed up to were unreadable. Restore is the other side of the backup coin, of course. And in discovering this horrible reality, I found restore religion too.
Sign up for our new free newsletter to get three time-saving tips each Friday — and get free copies of Paul Thurrott's Windows 11 and Windows 10 Field Guides (normally $9.99) as a special welcome gift!
"*" indicates required fields
Those experiences have guided my subsequent interactions with data backup across a wide range of local and cloud solutions over time. And while I have strayed from time-to-time, I’ve always believed in and preached the importance of maintaining a solid backup strategy. Naturally, this strategy evolved over time with the technology.
Like many of you, I spent many years trying to duplicate the Microsoft corporate infrastructure we’re familiar with at home. Doing so was pragmatic in a number of ways: For those working in IT, this has always been a good way to keep up with the technologies one was using at work or soon would be. For me specifically, it was a way to stay familiar with Microsoft technologies at a time in which I was no longer implementing them professionally. It was a way to keep my skills up to date so that I could write about the technology more effectively.
So I moved through various versions of Windows Server, and eventually migrated to Windows Home Server and then Windows Server Essentials. Those latter products were vastly simplified, for good and bad, but they offered some level of integration with various services, including, over time, backup services both local in the cloud. Since then, I’ve simplified my personal infrastructure, and I now use a prosumer NAS instead of a Windows Server. But the need for a good backup strategy—the “what”—has been consistent. Only the “how” has evolved.
Today, our backup needs are as numerous as they are diverse. And regardless of where you or your organization is, you have an incredible range of choices, from local redundancy and backup to full-on cloud backup and to an array of hybrid solutions that sit in the middle and may ultimately emerge as a best practice, of sorts, for the current era. And when you survey the available options, you’ll find that there are many good choices in each category. And a few that span categories.
One of the most complete of these, CloudBerry Backup, is currently sponsoring the First Ring Daily podcast here at Thurrott.com. And it is important that I discuss that for a moment.
One of the worries I have is that we’ll be approached by a potential sponsor and I’ll need to explain to my coworkers—or the potential sponsor—that their product simply doesn’t meet my expectations, or that it doesn’t meet the level of quality where I would feel comfortable reviewing it.
Today is not that day. Indeed, I am both happy and relieved to point out that we’re now 2-for-2 in the quality department when it comes to these sponsorships.
I don’t expect you to blindly believe that I could be swayed by the needs of a sponsor, but then you can find out for yourself: CloudBerry Backup is available for free to individuals, and you can try it for yourself. You should do so. You will be impressed, I’m sure of it.
To be clear, the product’s maker, CloudBerry Lab, doesn’t just target individuals. Instead, CloudBerry Backup is a comprehensive backup solution that should meet the needs of both individuals and businesses of any size. It is an astonishingly complete product, and it comes with a key advantage over most of the competition: CloudBerry Lab doesn’t sell or resell storage.
The CloudBerry approach, which I think of as Bring Your Own Storage, or BYOS, eliminates the guessing game that often accompanies picking a backup provider. Now, you are free to choose your own storage, and to shop around and switch storage solutions when it makes sense to do so. You can simply eliminate storage quality or cost from the equation and choose the best backup solution on its own merits.
This is very freeing. But it’s also important for regulatory reasons. In certain regions like the EU, it can be illegal for companies to store personal data outside of their own country. By democratizing your storage choices, CloudBerry Lab is making its superior backup product more accessible to everyone.
So, what does this look like in practice?
For an individual, it means using CloudBerry Backup with local storage, like a USB hard drive or a NAS or other network-based storage, or with consumer-oriented cloud storage from OneDrive, Google Drive, or similar services. It means you can do things like boot a PC with a specially made USB disk and restore that PC from the cloud provider of your choice.
For businesses and service providers, it means using Amazon S3 or Glacier, of course, plus Azure, Backblaze, and literally dozens of other cloud providers. It also means mixing and matching: You can perform local backups for a quick recovery and then archive those backups to the cloud, eliminating the need to store tons of redundant data locally. And you can further take advantage of archival cloud storage solutions like Amazon Glacier to provide a hybrid and stepped approach to data archival so that your oldest, less-frequently-needed data is stored as inexpensively as possible.
Of course, CloudBerry isn’t just about storage provider integration. It offers a complete suite of functionality that should meet any needs. System image and file backups, file search, restore, and disaster recovery across desktops and servers on Windows, Mac, and Linux, plus specific workloads like SQL Server and Exchange. You can restore physical systems to virtual machines hosted locally or in the cloud, and across a variety of providers and services. It supports compression and encryption, of course. The firm even provides a managed backup service, essentially Disaster Recovery as a Service (DRaaS), that provides comprehensive reporting and rebranding capabilities, plus integrated with ConenctWise/LabTech and Autotask. It’s a full suite of functionality.
I’m in no position to test all of that, of course. But I made sure to examine CloudBerry Backup’s core functionality: System image backups and file backups, and to do so across a variety of machines, local storage locations, and online services. Over three weeks or so, I worked with local storage, local network (NAS) storage, and a variety of online services. Looking forward, I am also interested in CloudBerry’s archival capabilities, since I have some NAS-based data that is still not backed up remotely (though it is replicated in the box). I am specifically looking at low-cost online storage, including Backblaze B2 or even Glacier.
I will get to my archiving needs at a later time. In fact, this may end up being my primary use of CloudBerry Backup going forward. But let’s start with some basics.
My first step was to create a system image backup of a single machine, a use case that spans both individuals (PCs) and businesses (servers). As noted, this capability is free to individuals, which is great for testing the product. But you can also purchase a Pro perpetual license for $29.99 and obtain more flexible scheduling and retention policies. A perpetual server license is $119.99.
The free and Pro versions of CloudBerry Backup work with consumer cloud services like OneDrive, Amazon Cloud Drive, and Google Drive, but most would likely opt to backup locally instead. (Dropbox is curiously missing from this list, however.) So I tested backing up to both external, USB-based storage and a NAS file share. Most businesses would choose between various cloud services, like Amazon S3, Backblaze, or similar.
The process is straightforward, though those not familiar with IT pro dashboards may be overwhelmed at first by the sheer number of options. I cannot stress enough how comprehensive is.
Using a wizard that is accessed from the Image Based button in CloudBerry Backup’s ribbon, I first choose to create a local or cloud backup. The other option is a hybrid backup choice that backs up to local storage first and then to cloud storage, and will likely be popular with many organizations.
Next, you choose or create a new “account,” or what I think of as the backup location. This can be one of several dozen cloud storage vendors or File System, meaning a hard drive attached to the current PC or a network share. The list of available cloud services is particularly notable, as you can see.
For the system image backups, I stuck with the File System option due to the expected size of those backups. But I did test both attached disks and network storage. In either case, you need to provide a display name, the path to the storage you will use, and, when appropriate, login credentials.
Next, you configure the plan name and specify which partitions you wish to backup. I chose to backup the entire PC each time—using “Backup All Drives”—but you could also back up only the required system partitions or choose the exact partitions you want.
From here, you have a few advanced options to consider. The first is whether to use block level backup, a differential backup feature that is more efficient after the initial full backup because it will only transfer those parts of files that have been modified since that backup. Plus compression and encryption; if you choose the latter, you are responsible for managing the password, as CloudBerry has no access to your data or keys. You can also specify a retention policy, which determines how many backup versions to keep for each file, and configure a schedule. I went with every night for testing purposes, mostly because I wanted to build up a respectable number of backups during the testing period.
The wizard also lets you configure any number of pre- and post-backup actions that can occur under various conditions—backup fail or success, for example—and even a backup chain, which lets you specify the order in which multiple backups will occur. This is a very impressive set of capabilities, as you can see.
After allowing multiple system image backups to run over several days on a few different PCs, I decided to try the restore process.
There are two primary ways to trigger this activity. You can build a bootable USB disk using a tool in CloudBerry Backup, and then reboot the system using that disk. Or you can trigger some restore types directly from the CloudBerry Backup application in Windows; what’s missing in the latter scenario is a complete system restore, or what some call a bare metal restore: For that, you will need to boot from the USB disk. Whatever your plans, you should make a bootable USB disk for disaster recovery purposes, so I did so.
If you do boot from this USB disk—which can be protected with a master password but otherwise appears to be similar to a standard Windows recovery drive, but with the Cloudberry tools—you’re given the option to do a bare-metal recovery or access some basic system tools like Command Prompt and Registry Editor. The first option loads a basic version of the CloudBerry Backup application that lets you access your restore plans and configure backup storage.
In either system image restore scenario, you access available restore options using the Restore Wizard, which is available in the Restore Plans tab. This wizard works much like the system image wizard, and includes many of the same choices: You select the backup location, the plan name (you can run a restore immediately or create a new restore plan), the type of restore (which will be based on what the wizard finds in the backup location (in my case, “Restore Image Based Backup”), and then the restore point, which defaults to the latest version.
(I did briefly run into a “path not found” issue when trying to locate my NAS-based backups. The trick here is to make sure that your backups are running under a specific account and then to specify that same account in the Restore Wizard.)
Next, you can restore to various virtual disk types (Hyper-V, VMWare, Azure VM, and several others), or to a physical disk. The latter choice lets you restore the entire PC—all of the partitions are selected by default—and it works exactly as you’d expect: I restore two PCs to a previous point in time successfully. Again, this only works if you boot the system with CloudBerry’s USB disk: If you try a full system image restore from within Windows, the wizard tells you that it cannot restore to the boot volume. (I assume you could simply de-select that partition, but I did not try this.)
Businesses that are interested in system image backups can also perform full server restores to the same or different hardware using this bare metal recovery option. Or, you can use the product’s Disaster Recovery as a Service functionality to restore Amazon EC2, Azure VM, or Google Cloud VM in the cloud. It’s also possible to perform file-level restores from image backups.
Next, I tested CloudBerry Backup’s files backup capabilities. This functionality will also benefit both individuals and businesses, and it lends itself well to our data-driven world. For either type of user, modern Windows versions already include excellent built-in recovery tools, of course, and backing up your documents and other important files is quicker, more efficient, and less costly, especially if you want to use cloud storage.
A files backup also allows you to choose a source other than the current machine. For example, on my home network, I was able to choose a folder share on my NAS and back that up to an online service, in this case, Backblaze B2, which works a lot like AWS S3. (I also tested against OneDrive for Business, but that isn’t recommended for real-world use.)
To get started, I launched the Create Backup Plan Wizard using the Files button in the CloudBerry Backup wizard. Like the other wizards in this solution, it is both easy to use and functionally complete, with options to meet virtually every need. Here, you can again choose a “Local or Cloud Backup” or a “Hybrid Backup.” But you can optionally enable a ransomware protection feature, too.
The way it works is that CloudBerry runs a few algorithms that look for suspicious changes to the files you’ve backed up. If a suspected ransomware event is occurring—if, for example, an encrypted backup file is unencrypted—it will alert you so that you can take action. I did turn on this feature, but I didn’t experience any alerts or issues so I wasn’t able to test how it works.
Next, you choose the storage service you’ll be using, authenticate as required—Backblaze requires an account ID, an application key, and then a bucket—configure a backup plan name, and choose between various backup modes.
This is the first place where this wizard really diverges from that for a system image backup. Here, you can enable features like encryption and file versioning, and whether to use CloudBerry’s own compression, which will save some amount of storage space, but will make the backups unreadable by anything other than this company’s tools. (The good news: Even if you are no longer a CloudBerry customer, you can always use the tools for free to access your data.)
There are other advanced options to consider: Volume Shadow Copy Service (VSS) for increased compatibility with third-party tools and uninterrupted backups of files in-use, NTFS permissions backup (which will speed things up when you need to restore the files), and a few others. Then, you need to choose which folders to back up. By default, the wizard presents the partitions on the current PC. But you can also arbitrarily add folder shares from other PCs, too (or instead). As noted, I used this wizard to test backups from a share on my NAS as well, and that worked just fine.
From there, you can optionally specify advanced filters rather than backing up all of the content in the selected locations, configure compression and encryption (if available on your storage choice; both are supported with the Backblaze B2 service I tested), a retention policy, a schedule, and pre- and post-backup actions.
At a high level, restoring from these backups works much like a system image restore: You run the Restore Wizard from the Restore Plans tab, choose the backup storage, select which backup you want (the most recent, perhaps, or a specific point in time or another period), and then the source. Here, you navigate through the folder structures of the backup locations, and then select which folders and/or files you wish to restore.
Do I have any complaints? A few, but none major.
The biggest issue I faced was that “path not found” problem, but a quick Google Search helped immensely. (Local restores from USB were never an issue.)
Also, the CloudBerry Backup application itself is a bit old-fashioned looking: It visually resembles Office 2010, with its blue ribbons and other UIs. Worse, it isn’t particularly well-suited for modern high DPI displays, and in certain cases, that can lead to overlapping interface elements that make it hard to select what you want.
And to be truly picky, I’d like to see the backup wizard display the storage savings you’d see by choosing CloudBerry’s compression: This will vary wildly according to the files you’re backing up, and seeing an estimated comparison of the compressed and non-compressed storage required would be helpful, I think.
But that’s about it. From a functional perspective, CloudBerry Backup is incredibly complete, and it is dense with useful features, only some of which I was able to test. And given the low-bar for testing this functionality for yourself—you can use it for free as an individual or test it for free for commercial use—it’s a no-brainer. CloudBerry Backup is one of those products that is so good, I’ll be using it myself. And my next step is to see how I might use this to finally archive my NAS data to a low-cost archival cloud solution. I’ll be examining that and writing it up soon.
Disclosure: As noted in the review, CloudBerry Labs is sponsoring the First Ring Daily podcast for a three-month period that ends March 30. The company asked me to write an honest review of its flagship product, but my opinions about CloudBerry Backup are my own and the company’s sponsorship played no role in this review. I have no qualms—moral or otherwise—recommending this solution to readers of this site. —Paul