Microsoft outages caused by CrowdStrike software glitch paralyze airlines, other businesses. Here’s what to know.

Banks, airlines, television networks and health systems around the world that rely on Microsoft 365 apps were hit by a widespread outage linked to CrowdStrike early Friday. Thousands of flights and rail services were canceled around the world, including more than 1,800 in the United States, and many other public and retail services were disrupted.

Here’s what we know about the outage:

What caused Microsoft’s global outage?

The issue was caused by a technical problem at a global cybersecurity company. Crowdstrike The company said it had identified the issue with its software and was working to fix it. CrowdStrike provides antivirus software for Microsoft’s Windows devices.

“Earlier today, a CrowdStrike update caused numerous IT systems around the world to go down,” Microsoft said in a statement to CBS News.

Later on Friday, Microsoft He said on social media “We have completed our mitigation efforts and our telemetry indicates that all previously impacted Microsoft 365 apps and services have been restored. We will enter a monitoring period to ensure that the impact has been fully resolved,” the company said in a statement.

CrowdStrike CEO George Kurtz said in a statement that the issue has been identified and a solution is being implemented. He added, “This is not a security incident or cyber attack. The issue has been identified, isolated and a fix has been deployed.”

In an update shared Friday afternoon: On social mediaKurtz again apologized to those affected and promised to “provide full transparency about how this happened and the steps we are taking to ensure this never happens again.”

What was affected by the Microsoft outage?

Over 3,000 people The flight was cancelled More than 11,400 flights within, to or from the United States were delayed as of Friday night, according to flight tracking services. Flight AwareMore than 42,000 flights were delayed worldwide on Friday.

American Airlines, Delta Air Lines and United Airlines had suspended operations earlier in the day due to the outage but resumed at least some flights late Friday morning.

“We are restoring service today and expect to have reliable operations for our customers tomorrow,” American Airlines said in a statement late Friday. United Airlines also said in a statement that it was proceeding with its recovery “cautiously over the course of the day.”

Time-lapse footage shared by the Federal Aviation Administration showed flights resumed Friday afternoon.

Brigham and Women’s Hospital in Boston canceled all scheduled surgeries and appointments that were deemed non-urgent because many of its computer systems were down and important digital records could not be accessed.

“We do everything on the computer now,” Meghan Mahoney, a neurocritical care nurse, told CBS News, “…patient history, allergies, etc. When that breaks down, we have to go back to paper charts.”

Memorial Sloan Kettering Cancer Center in New York City has postponed all procedures that require anesthesia. Emory Healthcare System in Atlanta has postponed certain types of surgery. Seattle Children’s Hospital has closed its outpatient clinics.

The Department of Health and Human Services said Friday afternoon that it was “working to assess the impact of the CrowdStrike outage on patient care and HHS systems, services, and operations.”

“Department of Homeland Security systems are currently being affected by a global outage of cybersecurity software CrowdStrike. As you have seen in the news, this is affecting many companies around the world. This was not a cyberattack,” a memo to Homeland Security officials obtained by CBS News said.

Portland, Oregon, Mayor Ted Wheeler declared a state of emergency for the city on Friday, allowing for “immediate response and resources to restore the city’s systems to a fully operational level.” New York CityMayor Eric Adams said the outage did not significantly affect city operations because training had been conducted earlier to teach city staff how to respond to IT issues.

In Europe, Lufthansa, KLM and SAS reported disruptions. Switzerland’s largest airport reported that no planes were allowed to land, according to CBS News partner network BBC News. In India, check-in was all done by hand at the country’s main airport in Delhi. Electronic check-in kiosks were not working and gate information was being updated by hand on whiteboards, the BBC reported.

German hospitals said on Friday they would halt elective surgeries and doctors in Britain said they were having problems accessing online booking systems. Pharmacists in Britain said they were experiencing disruptions to medicine deliveries and prescriptions.

The Federal Bureau of Investigation said it had no operational issues related to the outage, although some of its systems were affected and have workarounds that can be used during the outage.

The London Stock Exchange said it had experienced an outage in its regulatory news service but that trading was not affected. A spokesman for the New York Stock Exchange said the market was fully operational and was expected to open as normal.

Starbucks is experiencing a power outage Blocking Customers The company also banned the use of mobile ordering functions. Delivery companies such as FedEx also Some shipments may be delayed Due to a power outage.

“FedEx has activated contingency plans to mitigate the impact of the global IT outage experienced by a third-party software vendor. However, package deliveries may be delayed through July 19, 2024,” the company said in a statement.

When will Microsoft’s outages be fixed?

Even if CrowdStrike implements fixes, some of the issues it caused will likely take time to resolve, Omar Grossman, chief information officer at identity security company CyberArk, told Reuters, because the problem involves endpoint detection and response (EDR) products that run on individual client computers, he said.

“We found that because the endpoints were crashing — blue screens — we couldn’t update them remotely and had to manually fix the issue on each endpoint, a process that we expect to take several days,” Grossman said.

In an interview with CNBC, CrowdStrike’s Kurtz said many of the affected systems have already started to recover, and that for some customers the fix could be as simple as rebooting their computers or servers.

But he acknowledged that “some systems may not fully recover and we are working with each customer individually to ensure their systems remain operational.”

Kurtz did not give a deadline, but many companies and organizations around the world will likely have to rely on their own technology departments to get their systems up and running again.

— Chris Van Cleave contributed to this report.