For many companies, reliance on IT means that they would not be able to operate if their system went down. If this sounds like your company, it’s crucial that you have a disaster recovery plan in place so if the worst happens, you are fully prepared to get back online as swiftly as possible. If you haven’t got around to it yet, here are ten tips to help you put an effective disaster recovery plan together.
Know your most likely risks
Start your disaster recovery planning by identifying the most serious threats to your IT infrastructure, for example, system failure, staff error, fire or power loss. Identifying these can help you put procedures in place that will reduce the risk and determine the course of action needed in a recovery. If fire is a serious risk, for example, then it is quite clear that recovery needs to take place on a different site.
Prioritise your recovery
You should always prioritise the order in which things should be done in a recovery. What are the most mission-critical services you require (manufacturing, website, email) and what order do you need these restoring? This can help you create a recovery plan that has the least impact on the business.
As part of your prioritising, look at each of your services to determine your Recovery Time Objectives (RTOs). This means putting a maximum limit on the time you are willing to see that service go offline. For mission critical services, this won’t be very long, however, you have to realistically evaluate what the most likely cause of the disaster would be and how much work would be involved in bringing the service back online.
Back up data to meet your objectives
Besides RTOs, another factor which has to be taken into consideration is the Recovery Point Objective (RPO). Essentially, this means how much data your company is prepared to lose in a disaster. A key part of a disaster recovery plan is ensuring that you have your data backed up frequently enough to meet your RPO. If you are an ecommerce company and do not wish to lose any of your sales, your recovery point will need to be the very last second your site was online. In which case, you will need a constant backup in operation for your recovery to meet its objectives.
Create a critical response team
An essential element of any disaster recovery plan is to create a critical response team the personnel you need to get your system back online. This may also include third party contacts, such as software vendors or your web host, as well as internal staff.
To ensure you have someone for every role, it is always best to have a backup for each member. This way, if someone on the recovery team is on holiday or ill when a disaster occurs, you know there is someone else that can step in and undertake the job.
Ideally, you should have multiple means of contacting each member of the team and it should be made clear on the recovery plan who is responsible for calling each member.
Avoid confusion by creating a written, step by step DR manual.
When a disaster occurs, your staff are under pressure to get your system back online quickly. You have various people, each with their own agendas, all trying to carry out their tasks simultaneously. It can be hard in these stressful circumstances for team members to communicate effectively with each other and sometimes they forget that what they do has to fit into the process. If something is done out of sequence, it can put you back to square one.
To ensure the disaster recovery process goes smoothly there should be a step by step action plan put in place which specifies the order in which things should be done and who should carry out each task.
Test your recovery plan
No matter how well you plan your recovery, in reality, putting your system back online always works out differently. You should test it, in full. This way, anything that does not work out in practice can be modified and things you overlooked can be added to the final version. Always build in contingencies.
Have your backup resources ready and waiting
Always have a full set of backup resources in place. If the cause of your disaster is a failed hard drive, getting it fixed will be a lot easier if there is a spare server on site. The last thing you need is to wait for a replacement to be delivered.
In addition, you should also have full documentation available for all the hard and software you need as well as complete selection of any tools needed.
Have full details of all your software
If you run lots of different software programs, it can be easy to forget one when in the recovery process. You should have a complete list that gives the details of each application, how they should be configured, the contact details of the application owner and your contract details.
Make sure you have an up-to-date network diagram
A network map can save you hours of work and prevent the trial and error of looking for specific faults or rebuilding a system to help, clearly identify each node on the switch and panels.
Make the most of virtualisation
If you have short recovery time objectives and want up to the second recovery points, then you can benefit by using high-availability virtual machines, such as VMware, for your system backup. These can offer a far easier disaster recovery solution than physical servers, as virtual machines are capable of automatically restarting an application on alternative hardware without loss of data or availability.
If you are looking for a cost efficient way to meet your disaster recovery objectives, take a look at our VMware and HyperV cloud hosting packages.