Backups. One of the most boring, and important, aspects of running a modern business.
When I’m discussing backup and disaster recovery (DR) options with a client one of the first questions I ask is “How much data can you afford to lose?”.
Almost invariably the answer is “None”. To which my response is almost invariably “OK. So how much data can you REALLY afford to lose?”. Now, this is, of course, an overly simplified way of looking at it – most companies will have multiple applications, file repositories, and systems, and these will have different requirements for backup and DR.
Backup and DR, in their simplest forms, can be broken down into 2 requirements:
- Recovery Point Objective (RPO) – essentially how much time can pass before the amount of company data lost exceeds the maximum tolerance, and the business is at risk. If the most recent backup is older than this amount of time, then the RPO has been missed.
- Recovery Time Objective (RTO) – this refers to how much time can pass before the business process that has been disrupted can be restored to service.
An excellent example of backups gone wrong is the recent outage at source code collaboration company GitLab.com. You can read what happened is here, but the bottom line is that the actions of one GitLab.com employee led to accidental deletion of company data, resulting in an outage of the platform, and that – in GitLab’s own incident report – “out of 5 backup/replication techniques deployed none are working reliably or set up in the first place”.
In this case, the data that was lost was not production code, but rather less-important things like comments and bug reports, but what if it wasn’t? Even though the impact of this data loss was fairly low, the damage to the company’s reputation is going to be very difficult to recover from, and there’s no telling how much money they will lose from this outage in the long run.
This is just one company that experienced an outage resulting from lack of proper backup procedures, but believe it or not, this is happening all over the place. It may even be happening within your own business, without you being aware of it.
Consumption of Office 365 products is growing at a tremendous rate. Of these products, OneDrive and Exchange Online have the highest penetration rate.
But would it surprise you to learn that deleted items in Sharepoint/OneDrive libraries are kept for only 90 days by Microsoft, and deleted mailboxes for only 30 days?
Yep, unless you’ve backed up user mailboxes and files yourself, or have eDiscovery and Hold enabled in the case of mailboxes, that data is gone forever. In a typical on-premises Exchange and file server environment, this would be completely unacceptable.
If you, like many IT managers/Infrastructure managers/CIOs, are shocked by this, good. At least now you’re aware of the risk. If you were aware of the risk and have already put measures in place (and tested them!) to manage this, even better. But if you have not, may I shamelessly suggest contacting our team at firstname.lastname@example.org