4 Backup Tips to Prep Your Clients for Disasters

Winter is coming, and not just to Westeros. The Atlantic hurricane season is approaching its last days. So far we have managed to avoid a major hurricane striking the U.S. mainland, a record-breaking nine-year stretch going back to October 2005. (2012’s Superstorm Sandy had been a hurricane earlier when it passed over Cuba, but was no longer one when it hit the New York area on Oct. 29. That is little consolation, though, to those residents still cleaning up the mess.)

But this year is looking to be another rough one for those areas where cold weather is a problem. Last winter a record amount of ice on the Great Lakes lasted until mid-June, and this year we saw 20 inches of snow in Wyoming before the end of summer.

Such events can be bad news for MSPs. So, whether you need to protect your and your clients’ data and systems from tropical storms or Arctic blizzards, here are four best practices to follow.

1. Have a good disaster preparedness plan

When disaster looms, that isn’t the time to start preparations. You may have a few hours’ or days’ notice to prepare for a flood, hurricane or blizzard. But a fire, tornado, computer virus, lightning strike, earthquake or blackout can hit without notice. The plan should be in place well ahead of time, tested and drilled, so it just needs to be executed.

2. Prep for offsite data backup

Have all data ready for backing up offsite automatically. Your customers and employees should save their data on the drives or directories that are being backed up offsite. If off-siting is not automatic and you have some advance notice of something such as an approaching hurricane, make a full backup and transport it to the disaster recovery location.

Part of this preparation includes making sure that all your IT staff are fully briefed and drilled on how to restore the data. After all, high winds, floods or other problems may keep some of your employees from being available to perform the restorations. Staff will also need to be ready to answer any customer questions as to the status of the backup and restoration activities–not just to keep the customers happy, but to reduce the number of customer phone calls that will distract the staff from doing their jobs at that critical point in time.

3. Plan for each possible type of disaster and how to recover from it

Not all disasters are the same. A tornado wreaks complete destruction in a few minutes but has a very narrow footprint. In such a case, a disaster recovery location needs to be only a few miles away. But unless the data center has a hardened shell or is underground, you may need to rebuild the primary location. Similarly, blizzards produce a lower level of concentrated damage–the data center equipment itself will probably survive, but roads within 100 miles may be impassible for days. The October 1989 Loma Prieta earthquake south of San Francisco, on the other hand, lasted less than 15 seconds but collapsed buildings and highways 50 miles away. And a fire or broken water main may affect only a single building, but require a complete rebuild. Each of these situations requires a different type of disaster recovery plan.

4. Test your backup and recovery process

Finally, having a well-thought out DR plan, no matter how good it looks on paper, is not enough. You need to test your backups and your recoveries on a routine basis–not just the backup servers themselves, but the entire process, including the staff. For example, if you have a failover system, now and then deliberately fail over from your primary to your secondary system during the weekend to see if the systems stay online. Also do a practice recovery from the backup sources to the primary storage, or a new server, to verify that the data can be restored.

Visit us to find out more information about How to Build a Disaster Recovery Plan.