CrowdStrike have effectively a mini root cause analysis out
Pretty much as everybody knows, they did a channel update and it caused the driver to crash.
If they blame the person who did the update.. they shouldn’t, as it sounds like an engine defect.
https://www.crowdstrike.com/blog/technical-details-on-todays-outage/
For the people thinking ‘shouldn’t testing catch this?’, the answer is yes. Clearly something went wrong.
This isn’t CrowdStrike’s first rodeo on this, although it is the most severe incident so far.
Eg just last month they had an issue where a content update pushed CPU to 100% on one core: https://www.thestack.technology/crowdstrike-bug-maxes-out-100-of-cpu-requires-windows-reboots/
Truthfully these issues happen across all vendors - I’ve had my orgs totalled twice now by AV vendors, one while I was on holiday abroad and had to suspend said holiday.
For the people thinking ‘shouldn’t testing catch this?’, the answer is yes. Clearly something went wrong.
This isn’t CrowdStrike’s first rodeo on this, although it is the most severe incident so far.
Eg just last month they had an issue where a content update pushed CPU to 100% on one core: https://www.thestack.technology/crowdstrike-bug-maxes-out-100-of-cpu-requires-windows-reboots/