Physical and virtual servers
Image-level backup of Windows Server and Hyper-V guests, so recovery restores a working machine rather than a folder of files.
Backup & Recovery
A backup is only real once it restores
Nearly every business we assess has a backup. Far fewer have one that has been restored from in the last year, and some have been failing silently for months. We design, monitor and regularly test backups so the answer on the bad day is a timeline, not a hope.
🛡 3-2-1 By Design⏱ Immutable Offsite Copies💰 Tested Restores★ Monitored Daily
The Design
Two numbers decide your entire backup design
Before choosing any product, answer two questions. How much work can you afford to lose, and how long can you afford to be down? Everything else follows from those.
The first number is your recovery point objective. If backups run nightly at 10pm and a server dies at 4pm, you have lost eighteen hours of work, every invoice, chart note, quote and email processed that day. For some businesses that is survivable. For a medical practice or a busy dispatch operation it is not, and the answer is more frequent snapshots rather than a bigger nightly job.
Recovery time, and what it really includes
The second number is recovery time objective: how long from failure to working again. People underestimate this badly because they think about copying data and forget everything around it. A real recovery includes sourcing replacement hardware, rebuilding or booting a virtual machine, restoring the operating system and applications, restoring data, reconnecting authentication, testing, and then getting staff back in. Copying files is often the shortest part.
This is why we design around instant recovery where the budget allows: a local appliance that can spin your server up as a virtual machine within the hour while the permanent rebuild happens in the background. It converts a multi-day outage into a lunch break.
3-2-1, and why the offsite copy must be immutable
The rule is three copies of your data, on two different media types, with one copy offsite. It has survived because it fails gracefully, no single event takes out everything. In Arizona the offsite element earns its keep against fire, theft and monsoon power surges that kill a NAS and the server it sits beside.
Ransomware changed one detail. Modern attackers look for backups first and delete them before encrypting, and a network share that your server can write to is a network share the attacker can wipe. The offsite copy therefore needs immutability, storage that cannot be altered or deleted for a defined retention window, even with valid credentials. Without that, you have a backup that an attacker gets to veto. Our ransomware page covers the rest of that chain.
Monitoring closes the gap between configured and working. Backup jobs fail for boring reasons: a full disk, an expired credential, a new server nobody added to the job, an agent that stopped after an update. We alert on failures and on suspicious successes too, because a job that completes in three minutes when it used to take two hours is not good news.
Then we test. Scheduled restores of real files, real mailboxes and full virtual machines, with results written down. Once a year we recommend a full recovery rehearsal, which is where disaster recovery planning takes over, see disaster recovery and continuity, or read more at backup.orcait.io.
Untested backup is not a control. It is a belief.Orca IT, Gilbert AZ
What We Protect
Everything that would hurt to lose
Six workloads we back up, each with different mechanics and different restore expectations.
File shares and NAS
Versioned protection for shared drives and Synology-class NAS units, with retention long enough to survive quiet corruption.
Cloud data
Exchange Online, OneDrive, SharePoint and Teams. Microsoft replicates your data but does not protect it from your own deletions.
Workstations and laptops
Coverage for the local files people never moved to the share, and for the field laptops that live outside the office entirely.
Line-of-business databases
Application-consistent backup for SQL and practice or accounting systems, so restores come back usable rather than corrupt.
System configuration
Firewall, switch and server configurations, so rebuilding infrastructure does not depend on someone remembering how it was set up.
How We Run It
Six practices that separate real backup from theater
Configuration is the easy part. These are the habits that make it survive contact with reality.
Daily job verification
Every job is checked for success, duration and size. Sudden changes in run time or data volume get investigated, not celebrated.
Immutable offsite storage
Cloud copies written to storage that cannot be deleted or altered within the retention window, even by an account with valid credentials.
Scheduled restore testing
We restore real data on a schedule and record the result, including how long it took. Recovery time stops being a guess.
Retention that matches your risk
Long enough to catch slow corruption and quiet deletion, short enough to stay affordable. Reviewed rather than set once and forgotten.
Coverage audits
New servers, new shares and new cloud workloads get added to protection. Gaps usually appear because something was created after the design.
Documented recovery steps
Written procedures for each workload, so recovery does not depend on one engineer being available and awake.
Backup and recovery questions
What do RPO and RTO mean in plain terms?
RPO is how much work you can afford to lose, measured in time. If backups run once nightly, your worst-case RPO is nearly a full day. RTO is how long you can afford to be down before you are running again. Those two numbers decide your design and your budget, and everything else is detail.
Does Microsoft back up our Microsoft 365 data?
Not in the way most people assume. Microsoft protects the platform and replicates data for their own resilience, but they do not protect you from deletion, ransomware encryption of synced files, or a departing employee clearing a mailbox. Retention windows are limited. Third-party 365 backup fills that gap and we consider it standard.
Is a NAS in the server room enough?
It is a good first copy, and it is fast to restore from. It is not sufficient on its own, because fire, theft, a power surge or ransomware with network access can take the NAS and the server together. It needs a partner copy offsite in immutable storage.
How often should backups run?
For most businesses, hourly or every few hours for critical servers and nightly for everything else. The right answer comes from your RPO. A practice generating chart notes all day needs tighter intervals than a firm whose data changes weekly.
What does immutable actually mean?
The storage refuses to modify or delete the backup data for a set retention period, regardless of what credentials are presented. That matters because ransomware operators specifically hunt for and delete backups before encrypting. Immutability removes their ability to do that.
How long does a full recovery take?
It depends on your design. With a local recovery appliance, a failed server can often run as a virtual machine within an hour while the permanent rebuild happens behind it. Restoring from cloud alone with no local copy can take days for large datasets, mostly limited by your internet speed.
Do you test our backups or just monitor them?
Both. Monitoring catches failed jobs. Testing catches the harder problem, which is a job that reports success but produces an unusable restore. We restore real data on a schedule and record the outcome and duration.
Can you take over a backup system we already own?
Usually yes. We support common business backup platforms and Synology-class hardware. We start by verifying whether the existing design meets your recovery objectives, and we will tell you if it does not rather than assuming a rip and replace.
Let us try restoring something
A single test restore tells you more than a year of green status emails.
Talk to Your Pod
When did you last restore something?
We will review what you have, attempt a real restore, and tell you honestly how long a full recovery would take today. If the answer is uncomfortable, better to learn it now.
(602) 677-0779Family owned in Gilbert, AZ since 2015 · onsite across the Phoenix metro · remote support nationwide · never outsourced
Same-day response
No long contracts
Flat, honest pricing
Five-star service
Get your free IT consultation
A few details and your pod gets right back to you, usually the same business day.