“Disaster,” when it comes to your data center, can be defined in a few different ways. Maybe it’s really just a minor inconvenience. Maybe it’s a major headache. It could even be a career-ending fiasco.
The difference between those, officials from appliance deployment provider NEI (News - Alert) say in a recent blog post, is probably found in “how well you have planned and implemented a data backup strategy.”
NEI has some suggestions you’ll want to consider.
Different types of backups are crucial to the successful server backup process. There are two classes of backups: physical backups, which give you the ability to restore a system on-site, and soft-backups, which store data off-site and they both address different problems. Physical backups restore data when something goes wrong on-site, such as a hardware failure while soft backups are for when flood, fire or something similar happens.
Instead of the single backup scenario, multiple incremental backups are the way to go, including daily backups for the previous three days, a weekly backup, a monthly backup, and then fully maintaining the past three weekly and monthly backups.
In addition, “bare metal restore” means starting without anything on the hard drive. If this is what you choose for your server backup procedure, be sure you have all bootable media, drivers and COAs, and that they match the system.
Remember, RAID is not backup. RAID, as NEI officials point out, is “redundancy to protect against a hardware failure, not a backup plan.”
Also, put your backup plan in writing. That’s right, write down your backup and recovery strategy. Decide how to handle a problem before it occurs, instead of while the situation is unfolding.
Finally, verify your server data backup and your restore procedures. In essence, test them to prevent a scenario in which you need to backup your system, only to realize your steps are not working correctly. That probably won’t be your finest day with the company, but it could very well be your last one.