Backup and disaster recovery for London organisations
A backup is not a copy of your data. It is the promise that you can get back to work.
Logic Networks sets up and looks after backup and recovery for organisations across London and the rest of the UK, and writes the plan for what happens on the bad day.
The difference between the two words matters. Backup is the copy. Recovery is how long it takes to be working again, and how much work you lose. Most organisations have the first and have never measured the second.
We start by asking two questions and writing the answers down: how long can you be without this, and how much of it can you afford to lose.
An untested backup is a hope, not a control.
Every organisation we take over has a backup that runs. Far fewer have a restore with a date on it, and that is the only evidence that any of it works.
- A copy held where your own accounts cannot delete it
- A restore tested and dated, not assumed
- An agreed recovery time for each system
- A written plan the person on duty can follow
What is included
Six pieces of work, from the copy to the plan on the wall.
Most organisations need the first three immediately and grow into the rest as the cost of a lost day becomes clearer.
Backup of servers and files
Everything that matters, on a known schedule.
Servers, file shares, databases and the applications your work depends on, copied on a schedule that matches how quickly the data changes.
The schedule is a decision, not a default. An accounts system that changes all day needs a different interval from a folder of archived documents, and paying for the same frequency on both is how backup budgets get wasted.
Backup of Microsoft 365
Because Microsoft does not do this for you.
Mail, files, sites and Teams content backed up separately from the tenancy itself.
Microsoft protects the service against their own failures. They do not protect you against a deleted folder, a leaver’s account removed to save a licence, or a file encrypted on a laptop and then synced to the cloud. Those are treated as valid instructions.
Copies kept out of reach
What the attack cannot delete.
At least one copy held where your normal systems and normal accounts have no power to change or delete it.
This is the single change that separates an organisation that recovers from ransomware in a day from one that negotiates. If the backup can be reached from the network it protects, it is part of the same problem.
Test restores
Dated evidence, not assurances.
A restore performed as a test, recorded with the date and what was recovered, on a cycle rather than once at the beginning.
We also test the awkward cases: a single file from six months ago, a whole mailbox, and a full system onto different hardware. Each one fails differently, and it is better to find that out on a quiet Tuesday.
The recovery plan
Followable by whoever is on duty.
A written plan covering what to bring back first, who to call, where the copies are and who can authorise a decision.
It is kept somewhere that does not depend on your systems working, including on paper, because a plan stored only in the environment it is meant to recover is not a plan.
Continuity for the whole organisation
Beyond the data, to the working day.
What your people do while systems are coming back: where they work, how they communicate, what they tell customers and which supplier they call.
This is the part most technical plans leave out, and it is what your insurer and your larger clients now ask about. We help you write it in plain words rather than as a document nobody reads.
How it works
Three arrangements, and the first one is worth doing even if you stop there.
A review takes a few days and produces something useful on its own: the honest answer to how long you would be down and how much you would lose.
Tell us what you have in place today, even if the answer is that nobody is sure, and we will say which of the three we would suggest.
-
A recovery review
What is backed up, what is not, how long a restore would actually take and where the copies are. Written up with the gaps in priority order.
-
Set it up properly
A one off piece of work: the backups configured, a copy put out of reach, a restore tested and dated, and the plan written.
-
Managed backup
Monitored daily, tested on a cycle and reported on monthly, as part of managed IT.
Who it is for
Everybody needs this and almost nobody reviews it, because backup is the one part of technology that appears to be working right up until the moment it is needed.
What differs by sector is how long you can be down before it becomes serious, and who you have to tell.
- Businesses where a lost day is a lost week of orders
- Charities and non profits holding records on beneficiaries
- Education organisations with coursework and student records
- Organisations asked about recovery by an insurer or a larger client
- Anyone who has just had a near miss and wants it not to happen again
- Organisations that have moved to the cloud and assumed it included backup
What we need from you
Two answers and one decision. The two answers are yours to give, not ours.
Everything else in this page follows from them. Without them, a backup is designed by guesswork and usually costs more than it needs to.
How long you can be without it
Per system, not overall. An hour for email and a day for the archive is a perfectly reasonable answer, and it saves a great deal of money.
How much work you can lose
If the last copy was taken at midnight, you lose the morning. Deciding that is acceptable, or not, sets the schedule.
Who decides on the day
One named person who can authorise a restore, and a second if the first is away. Recovery delayed while somebody is found is the most common avoidable loss.
What we usually find in a recovery review
These are the items that come up most often, and every one of them has been in an organisation that believed it was covered.
- A backup running for years and never restored from
- The backup device in the same room as the server it protects
- Microsoft 365 not backed up at all, because it was assumed to be included
- The backup account using the same password as an ordinary administrator
- One system left out, usually the one added most recently
- Retention set to a fortnight, so a problem noticed a month later cannot be undone
- No written recovery order, so everything would be restored at once
- The plan stored on the file server it is meant to recover
Our steps for working together
This work has a fixed order, because each stage is meaningless without the one before it. A backup designed before the recovery times are agreed is a guess.
- List the systems Everything your work depends on, including the application somebody set up themselves.
- Agree the two numbers For each system, how long you can be without it and how much work you can afford to lose. Written down and signed off.
- Design to those numbers The schedule, the retention and where each copy is held. Frequent where it matters, cheaper where it does not.
- Put one copy out of reach Held so that your own accounts, and anybody using them, cannot change or delete it.
- Test a restore A file, a mailbox and a whole system. Dated and recorded, with anything that failed fixed before we stop.
- Write the plan Recovery order, contacts, locations and who authorises what, kept somewhere that survives the outage.
- Test again on a cycle Because systems change, and the copy that worked last year may not include what was added since.
How we set priorities on the day
The recovery order is decided in advance so nobody has to argue about it during an incident. For most organisations it is communications first, then the systems that take money or deliver the service, then everything else. The archive comes last, however large it is, because nobody is waiting for it.
The questions your insurer will ask
- Are backups held separately Meaning off the network they protect, and unreachable from your ordinary accounts.
- When was the last successful restore They are asking for a date, not for reassurance.
- What is your recovery time Per system, and whether it has been measured or estimated.
- Who is responsible A named role, with a deputy.
- Is there a written plan And can somebody follow it while the systems are down.
The three copies rule, and why it still holds
The old guidance was three copies of your data, on two different kinds of storage, with one of them somewhere else. It was written before the cloud and it has survived because the reasoning behind it has not changed.
Three copies covers the case where one fails silently. Two kinds of storage covers the case where a whole type has the same fault. One copy elsewhere covers the fire, the flood and the theft.
What has changed is that a fourth condition now matters more than any of them. One copy must be unable to be deleted, including by somebody holding your own administrator password. Attacks look for the backup first, because an organisation that can restore does not pay.
In practice, for a small organisation, that means a local copy for speed, a cloud copy for the building risk, and a locked copy for the attack. The three are not expensive. Deciding to have them is the difficult part, because nothing visible happens on the day you do it.
What we will not claim
No backup arrangement makes an incident painless. What a good one does is turn a business ending event into a difficult week with a known end. We will tell you honestly what your current arrangement would deliver, including when the answer is that it would not deliver anything.
What good looks like
You can name your recovery time for each system and it is a measured number. Your last test restore has a date within the last three months. One copy is somewhere an attacker with your administrator password cannot touch. The plan exists on paper. And the person who would be on duty has read it.
Related services
Recovery depends on the rest of the environment being known, so this work is usually done alongside one or two of the pieces below.
Where you already have a backup product you are happy with, we work with it rather than replacing it.
- Cybersecurity because the modern incident is an attack, not a failure
- Microsoft 365 for the tenancy that also needs backing up
- Cloud and Azure for holding copies away from your building
- Managed IT to have it monitored and tested on a cycle
- Networks and infrastructure for the connection a restore depends on
Articles and news on backup and recovery
Notes on what a backup does and does not protect, written for the person who would be asked on the day.
The first hour after a suspected email compromise
What to do, in order, when you think somebody has got into a mailbox. Written to be followed by whoever is in the office at the time.
Read the articleThe five security controls worth putting in place first
Security spending often starts at the wrong end. These five controls remove the largest share of everyday risk and cost the least.
Read the articleTurning on multi factor authentication without stopping work
The single change that prevents most account takeovers, and how to introduce it across a team of thirty people in a fortnight.
Read the articleTalk to us about backup and recovery
Tell us what is backed up today, or that nobody is certain. We will review it, tell you how long a restore would really take and what we would change, in order.
If you are in the middle of an incident, say so in the first line.