It’s the first time that a number has been put on the glitch that is still causing problems around the world.

Microsoft says it estimates that 8.5m computers around the world were disabled by the global IT outage.

It’s the first time a figure has been put on the incident and suggests it could be the worst cyber event in history.

The glitch came from a security company called CrowdStrike which sent out a corrupted software update to its huge number of customers.

Microsoft, which is helping customers recover said in a blog post: “We currently estimate that CrowdStrike’s update affected 8.5 million Windows devices.”

  • RentlarEnglish
    2·
    5 months ago

    https://forums.rockylinux.org/t/crowdstrike-freezing-rockylinux-after-9-4-upgrade/14041

    The bug seems to have only affected certain Linux kernels and versions. Of course no one cared, because it didn’t simultaneously take out hospital systems and airline systems worldwide to an extent that you’d only think you’d see in movies.

    Linux has comparitive advantages for being so diverse. Since there are so many different update channels it would be hard to pull off such a large outage, intentionally or unintentionally. Yet, if we imagine a totally equivalent scenario of a CrowdStrike update causing kernel panics in most Linux distribitions, this is what could be done:

    • Ubuntu, Redhat, and other organizations who make money from supporting and ensuring reliability of their customers’ systems, would be on the case to find a working configuration, as soon as they find out it’s not an isolated incident or user error.
    • If one finds a solution, it will likely quickly be shared to other organizations and adapted.
    • The error logs, and inner workings of the kernel and where it fails are clearly available to admins, customer support personnel and tech nerds, so they aren’t fully at the mercy of the maintainers of the proprietary blobs (both Microsoft and Crowdstrike, for Windows, but only Crowdstrike for Linux) to determine the cause and potential solutions that would be available.
    • The Linux internet-facing component updates can be rolled back and inspected/installed separately to the Crowdstrike updates. The buggy update to Microsoft Azure and from Crowdstrike happening together on the same day muddied the waters as to what exactly went wrong in the first several hours of the outage.
    • There’s more flexibility to adjust the behaviour of the kernel itself, even in a scenario CrowdStrike was dragging its feet. Emergency kernel patches could just set to ignore panics caused by the faulty configuration files identified, at least as a potential temporary fix.