Backup Strategy – Do you have one?
Backup is most definitely one of those topics that is always overlooked in a lot of network installations we come across. Not necessarily by the original installers but by the people managing and maintaining that backup – simply from ensuring USB drives or Tapes are changed when they need to be. Many people who get this role to swap USB drives in the morning either swap them without checking if the backup has worked. Likely there has been no easy way of checking setup in the first place or they are simply are too busy and forget to swap them. Common scenarios in most smaller businesses. Getting users on board with the importance of backup is key to success.
The symptom of this attitude is they have likely never needed their backup. They have been lucky in that nobody has deleted a critical file that needed to be restored and their server(s) have remained OK without requiring rolling back ever. Never been hit by any natural disasters and the servers hard drives have been working fine up until now. This can and will change in an instant. Hardware fails – over time hard drives need replacing, equipment gets sluggish and starts to become unreliable – the list goes on.
Any good IT support company will ensure each one of their customers has a way of restoring their data should the worst happen and identify their recovery time objective (RTO) and tailoring a solution around this. The quicker you need to be back up and running, the more expensive the solution and the more hardware you generally require.
I’m going to cover a few typical backup strategies which can apply to most businesses we support;
Minimum – ‘Bronze’
If your core data doesn’t change much as you have a small number of users or you utilize lots of cloud based services to run your day-to-day these usually have backup routines with the provider so you don’t have to worry about it. We always recommend an offsite backup at least weekly but ideally daily to a USB drive. Easily portable in a briefcase of handbag, encrypted if supported.
Max data loss – 1 day or 1 week dependant on snapshot schedule.
Target Restore Time – 1 Day for full server down.
Double Bubble – ‘Silver’
Your data may change a lot all day and you may also have critical databases and system on site which need backing up regularly. The above USB solution each evening coupled with a NAS device. Regular ‘snapshots’ of important volumes to this NAS appliance for quick restoration (anything from 15-60 minutes dependent on data). You can even add a second NAS device and replicate to an offsite location.
Max data loss – 15-60 minutes dependant on snapshot schedule.
Target Restore Time – 2-4 Hours for full server down.
Belt and Braces – ‘Gold’
Snapshots between 5 and 60 minutes of all servers to an on-site NAS, replicated to a secondary NAS. Backups then transformed into Virtual machines (switched off but ready to go) to an on or off premise ‘standby’ server reducing your RTO drastically. Daily offsite backups via USB Drives.
Max data loss – 5 minutes -10 minutes dependant on replication schedule
Target Full server Restore Time – 2 Hours if utilising offsite DR, 15-20 minutes if using Onsite or local standby server (assuming server down)
Obviously the above times dependent entirely on the data to be restored, the physical server to be restored onto and the severity of the issue – if your building sets on fire you need to source replacement hardware & a new office.
All of these solutions should be monitored, email alerting setup and regular backup/DR verification tests (Just because a backup has completed does not necessarily mean the data its backed up is any good, don’t assume!). Champion a user or users on site and show them how to use the backup software and test their restores.