3 Critical Oversights with Network Backup and How to Avoid Them

At NovaStor we have a huge open office, and my desk is close to some of our support engineers. As I’m naturally curious, I listen to their calls sometimes. If you have ever worked in support, you probably know that some calls make you laugh and some calls make you doubt in your own products or even the viability of the human race. Anyway, during the last few days I have realized that there is a lack of information in the backup world about some of what we take for granted as basic things. Don’t get me wrong, that’s totally normal, most of our customers don’t have a dedicated backup administrator. They have to deal with literally everything (databases, storage, PCs for the colleagues, wiring …) and no one can ever know everything in detail. Much rather it’s our fault that we didn’t inform you earlier. That’s why I wanted to share some insights regarding network backup that hopefully help you with your backup software. The following are my three favorite critical oversights when dealing with network backup from the last couple of weeks.

Running more than one backup software on one machine

Here we have an ‘either’ – ‘or’ situation. If you run two (or more) backup solutions to backup files from one server, usually you don’t need to worry about compromising your data or your backups. A couple years ago that was an issue with most data protection solutions, because they requested information from the system’s archive bit. Nowadays most vendors have implemented their own monitoring bit to determine which data has to be saved to the backup device (depending on the backup type).

But there are other situations in which one solution could negatively affect the other. For example, a big yellow vendor for security and backup solutions uses their own system driver to access tape drives. That is not a problem, if you use just their software, but if you need to use a second backup solution (or you want to test a new one), you will have trouble getting the tape drive working correctly. You need to manually remove the driver and install the regular Windows drivers in order to grant access to the other backup software.

Another situation would be the backup of Exchange. Most data protection solutions truncate the log files after successfully transmitting the data (again, depending on the type of Exchange backup). Thus, a second backup solution might not be able to receive the complete data for the same backup type anymore. Most of the time that happens when customers are using a backup software for local backups and a second one for remote backups.

To solve this problem you have two options:

• Use just one backup software for everything or
• If that is not possible, figure out a backup strategy that includes all solutions.

I will write a more detailed article about backup strategies for more than one backup solution later, but to keep things simple for now:

--> If you have a situation similar to the above mentioned Exchange debacle, backup the database locally to e.g. a NAS with software A and then backup the NAS to the remote site and don’t touch the database at all with software B.

Email reports for failed backup jobs

Setting up a report to get a message in case the backup job didn’t run correctly makes sense, and I think that every system administrator typically sets up reports automatically without even thinking about it. Could you tell me right now in which intervals the reports are sent or if you have just not gotten a report when you were supposed to?

If it’s in problematic situations only, you should rethink your cycle. Let’s say the mail server stops working and a day later your backup told you that the storage is now full and no further backups are able to run. Without the mailing service, you wouldn’t get any notice. How long would it take for you to realize that something is going terribly wrong?

Setup your email reports to send emails daily. Not only will you have a reporting tool to show legal entities that you really take care about business critical data, you will also have a guarantee that your service is running correctly.

Remote/ offsite backup

I thought my colleagues are joking around with me the other day when mentioning that they just had another call with one of our customers who lost his data due to an electrical surge (we had a huge thunderstorm during the weekend in Southern California, and the customer is located near our office). The storage that kept all the backups from the last 3 years, just didn’t want to come up this morning. As he didn’t copy his data to an offsite storage, my colleagues couldn’t do anything except trying to cheer him up (which obviously didn’t really help him in his recovery).

I know, it’s someone else and who knows how this guy set up his environment. That will never happen to us! But let me compare it to my car insurance. I pay nearly $150 per month (yes, I know…) and I really hope that I will never have an accident and need to call the insurance. But at least I pay every month and have something just in case, the same goes with the backup. We all hope we never have to deal with a burned down data center, theft, or Godzilla, but we are happy that we have something safe, just in case. And to be very clear: an offsite backup really is not a luxury option, it is a necessity. Depending on the amount of data you have, get a single tape, activate Amazon S3, or hire an MSP with a remote backup option. No matter what you do, just do it!

If you haven’t already, sign up for our email to receive information about the technology behind NovaStor DataCenter, NovaStor’s technology partners, Webinar invitations, and general network backup and restore knowledge.