Skip to content
Healthcare IT & HIPAA

When the Cloud Goes Down: What to Do in the First Hours of an Outage You Can't Control

November 17, 2026

By the Galleon IT Solutions team

Healthcare-first managed IT engineers serving DFW and Houston since 2017.

Last reviewed: November 17, 2026

Within two days in mid-November, two of the systems businesses trust most stopped working. On November 12, Microsoft 365 began failing for users around the world. Outlook and Exchange Online were hit hardest, with Teams and other Microsoft services affected too, and the disruption stretched across roughly a day before Microsoft reported everything restored. Microsoft said there was no indication of a cyberattack and pointed to a network configuration change it had to roll back. The next afternoon, the FAA ordered a nationwide ground stop after radar and communications problems in the air traffic control system. Departures stopped across the country for more than four hours, and thousands of flights were cancelled. Officials blamed overloaded network equipment in an aging system.

Neither outage was caused by anything a business owner did. No practice in Dallas or Houston misconfigured Microsoft's network, and no traveler could have prevented the ground stop. That is exactly the point of this article. Some failures are completely outside your control. What is inside your control is how your practice responds in the first few hours, and whether patients notice.

Healthcare felt it directly. Reporting on the Microsoft outage described hospitals in Toronto losing email and UK National Health Service staff falling back to paper records, with some referrals and pharmacy messages delayed. A medical practice runs on the same tools: email for referrals and records requests, Teams for staff coordination, cloud fax that often lands in an inbox, and a patient portal that sends its notices by email. When the platform underneath stops, the front desk feels it first.

Step 1: Confirm it is not you

The first fifteen minutes are usually wasted on the wrong question. Staff restart computers, reset passwords, and call the internet provider, when the real problem is hundreds of miles away in someone else's data center. The faster you confirm a provider-side outage, the faster you stop chasing local fixes and switch to your plan.

  • Check the provider's status page. Microsoft publishes service health in the Microsoft 365 admin center and on its public status account. Your IT partner should be watching both.
  • Check whether other services work. If websites load and your EMR opens but Outlook will not, the problem is almost certainly upstream.
  • Ask your IT partner one question: is this us or them? A good partner answers that in minutes, because they see many clients at once. When every client reports the same failure at the same moment, it is a provider outage.

Step 2: Tell your staff what is happening and what to use instead

Confusion costs more than the outage itself. Within the first half hour, the practice administrator or office manager should send one clear message through a channel that still works, usually a text message group or a phone call to each location. It should say three things: what is down, what to use instead, and when the next update will come.

Something as simple as this works: "Outlook and Teams are down nationwide. It is a Microsoft problem, not ours. Use the office phone and the staff text group for anything urgent. Referrals go by phone and fax until further notice. Next update at 11:00."

Step 3: Protect the patient-facing work first

Not everything matters equally during an outage. Sort the work into what touches patients today and what can wait.

  • Referrals and urgent results. If referrals normally arrive by email or email-based fax, call the referring offices you work with most and give them a phone or fax number that does not depend on the outage.
  • Prescriptions. Confirm e-prescribing still works. If it does not, follow your downtime procedure for prescriptions rather than improvising.
  • Scheduling and check-in. If your EMR is unaffected, carry on. If it is affected, move to your paper downtime forms immediately; waiting an hour in hope makes the backlog worse.
  • Patient communication. If appointment reminders or portal messages go out by email, expect gaps and have the front desk confirm the next day's appointments by phone.

Step 4: Do not create a second problem while fixing the first

Outages are when practices make their worst security decisions. Under pressure, staff forward patient information to personal Gmail accounts, text photos of charts, or use free file-sharing tools to keep work moving. Every one of those can turn an inconvenience into a reportable HIPAA incident.

Attackers know this too. Major outages are followed quickly by phishing emails and fake support calls that use the outage as cover: "Microsoft needs you to re-verify your account" or "click here to restore your mailbox." Remind staff that no legitimate vendor will ask for a password during an outage, and that any unexpected request to log in, install software, or grant remote access gets reported to your IT partner, not acted on.

Step 5: Keep a simple log

Write down when the problem started, what was affected, what workarounds were used, and when service returned. It takes five minutes and pays off three ways. It documents your response for HIPAA purposes if any patient information was involved. It gives your IT partner what they need to check for missed messages afterward. And it becomes the raw material for improving your plan, which is the subject of our companion article, Not If, But When: The Outage Preparedness Plan Every Medical Practice Needs.

Step 6: Recover carefully, not all at once

When service comes back, it rarely comes back cleanly. During the Microsoft outage, recovery rolled out region by region over many hours, and services that appeared to be working could fail again. Expect a flood of delayed email to arrive out of order.

  • Have one person review delayed referrals, results, and records requests that arrive after the outage, oldest first.
  • Enter anything recorded on paper during downtime into the chart, and have a second person check it.
  • Ask your IT partner to confirm that nothing was lost, including fax-to-email messages, voicemail-to-email, and portal notifications.
  • Do not switch staff back from workarounds until service has been stable for a while. A false recovery that fails again mid-afternoon is worse than staying on the backup a little longer.

What your IT partner should be doing while you do all of this

A practice administrator should not be the one refreshing status pages. During a provider outage, your IT partner should confirm the cause, tell you whether your systems are affected, watch for phishing that uses the outage as cover, advise which workarounds are safe, and verify a clean recovery afterward. If your current IT provider went quiet during the Microsoft outage, or learned about it from you, that is worth noting.

At Galleon, our healthcare clients across DFW and Houston get an initial response within 15 minutes, and during a widespread outage that response includes a straight answer about whose problem it is and what to do next. If your practice was caught flat-footed this month, call us at (972) 776-6366.

Frequently asked questions

Confirm it is a Microsoft-side outage through the service health page or your IT partner, tell staff what to use instead, move referrals and urgent communication to phone and fax, and avoid moving patient information to personal accounts. Then log what happened and verify that no messages were lost once service returns.

An outage on its own is generally not a breach. The risk comes from what staff do to work around it, such as sending patient information through personal email or unapproved apps. A downtime procedure that names approved workarounds in advance prevents that.

If one workstation or one office is affected, it is usually local. If every user is affected and other businesses report the same thing, it is the provider. Your IT partner should be able to tell you within minutes.

Yes. Attackers routinely use well-known outages as cover for fake password reset and support messages. No legitimate vendor asks for your password during an outage.

No. A practice cannot prevent a provider outage, but it can decide in advance how it will keep seeing patients when one happens. That is what a downtime and continuity plan is for.

Get Started

Ready for a straight answer about your IT?

Schedule a 20-minute discovery call. We will tell you what is working, what is not, and what the gaps would cost.