The call comes at 8:14 a.m. Someone cannot reach the shared drive. Neither can anyone else. Your server is down.
For a firm with ten or fewer employees running a single on-premises server, this is not a minor inconvenience. It is a near-total work stoppage. What happens next depends almost entirely on decisions made before this morning.
A single on-prem server failure can halt billing, file access, and critical applications for an entire small firm within minutes, even when Microsoft 365 keeps email running. Without a tested continuity plan, a recovery measured in hours typically extends to days. Structured backup and disaster recovery closes that gap and makes the difference between a bad morning and a business-threatening event.
What stops working — and what doesn't
Microsoft 365 Business Premium puts email, Teams, SharePoint, and OneDrive in Microsoft's cloud. Those services keep running regardless of what happens to the physical box in your server closet. Everything else tied to that box does not.
- Shared file storage. Every file saved to a mapped drive — the G: or S: drive your staff uses throughout the day — is inaccessible.
- Line-of-business applications. Billing software, practice management platforms, QuickBooks, and any application that lives on or connects to the server stops responding.
- Local databases. Client records, matter files, or patient data stored in an on-prem SQL database go offline with the server.
- Network printing. If the server manages print queues, printers stop accepting jobs.
- Backups stored on the server. If the backup destination was the server itself, it went down with everything else.
Without a continuity plan: hour by hour
Minutes 0–30: Confusion and reboots
Staff confirm the problem with each other. Someone reboots the server. It does not come back. The IT contact — possibly someone the firm has not spoken to in months — gets a voicemail.
Hours 1–3: Diagnosis
A technician connects remotely or arrives on-site. Identifying the failure takes time: a failed drive, corrupted RAID array, power supply fault, or something more serious. Once the failure mode is clear, the technician asks for a backup. This is the moment you find out whether last night's job actually completed and whether it has ever been tested against a real restore.
While the diagnosis runs, your staff is idle or improvising. Email works. Teams works. But no one can open a client file, post a payment, or produce an invoice.
Hours 4–8: Waiting on hardware and data
If a drive has physically failed, a replacement must be sourced — same-day availability is not certain. If the operating system is corrupted, reinstallation begins from scratch. A full server restore from image backup, assuming one exists and is current, can consume several additional hours depending on data volume. The business remains effectively closed to any work that touches the server.
Day two and beyond
If the backup is incomplete, outdated, or untested, recovery extends into the next business day or further. Some data may be unrecoverable. The firm faces missed deadlines, delayed billing, and the uncomfortable call explaining to a client why their file is unavailable.
With a continuity plan: the same morning
Automated monitoring detects the failure before staff arrive. Because the server's full image has been replicated continuously to a cloud environment, the failed server's workloads can be brought online in the cloud within a defined recovery window — often before the workday is fully underway.
Employees are redirected to the cloud-hosted environment. Files are accessible. The line-of-business application reconnects. Work continues while a dedicated team that knows your environment addresses the physical hardware separately, without the clock running against billable time or client deadlines.