Businesses and services around the world are gradually recovering after a global blackout that affected computer systems for hours on Thursday and Friday.

Businesses, banks, hospitals and airlines have been hit hardest after a “bug” in cyber security firm CrowStrike’s software update affected Microsoft Windows.

Crowdstrike CEO George Kurtz apologized for the problems, saying the problem is being fixed, but admitted it could be “a few days” before all systems are up and running.

While some air services are beginning to return to normal after thousands of flights were cancelled, carriers expect some delays and cancellations will continue into the weekend.

Many businesses now face delays and lost orders that can take days to resolve.

Health services in Britain, Israel and Germany also experienced problems.

The problems were first noticed in Australia and were probably most acutely felt in the air travel industry. There were delays at airports, with long queues as flights were canceled or delayed, planes grounded and passengers stranded.

Worry about future dangers

The global chaos has fueled concerns about the vulnerability of interconnected technologies worldwide and the extent to which a single software bug could have such a far-reaching impact.

As the recovery continues, experts say the blackout has raised concerns that many organizations are ill-prepared to implement contingency plans when a single point of failure, such as an IT system, or a piece of software within it, goes down.

But these outages will recur, experts warn, until networks are better prepared for such contingencies and organizations introduce better backups.

By Friday night, problems had eased in many parts of the world, with many airports saying that while there were still problems with check-in and payment systems, most flights were now operating normally.

And the website Downdetector, which identifies websites that may have technical problems, showed fewer websites in the UK experiencing problems by the end of the day.

Crowdstrike CEO George Kurtz also said in a post on X that he was “deeply sorry” for the problems caused. “We understand the seriousness of the situation and are very sorry for the inconvenience and disruption,” he says.

He even noted once again that the problem was not caused by a security breach, stressing that CrowdStrike’s customers are “fully protected.”

“We are working with all our customers to ensure that systems are back up and able to provide their services,” he concluded.

Microsoft also said that users may have to perform multiple reboots, up to 15 before the problem is resolved on their network.

Shares of Crowdstrike fell about 12% on Friday, at the expense of rivals SentinelOne and Palo Alto Networks.

Who is CrowdStrike and what does it do?

CrowdStrike is active in the field of cyber security. In this context, it develops software that aims to help businesses detect and prevent hacking. Most of the world’s Fortune 500 companies work with it, including major banks, healthcare companies and energy companies.

The company provides what is known as “endpoint security” as it uses cloud technology to protect internet-connected devices in cyberspace. This activity differs from alternatives provided by other cyber security companies, which include protection directly on central server systems.

CrowdStrike – which previously reached a market cap of around $83 billion – is one of the most popular cyber security providers in the world, with nearly 30,000 subscribers worldwide. But the global problem could force its customers and investors to reconsider their reliance on the company, opening the door to potential rivals such as Palo Alto Networks PANW.O, which saw its stock rise 1, 7% on Friday and Sentinel One, which jumped as much as 3.6%.