Alaska Airlines Resumes Operations After System Glitch Grounds All Flights (gizmodo.com) 9
Alaska Airlines and Horizon Air grounded all flights Sunday night due to a major IT outage, prompting a system-wide FAA ground stop that lasted until early Monday. Although operations have since resumed, passengers are still facing delays and residual disruptions. Gizmodo reports: The airline requested a system-wide ground stop from federal aviation authorities at about 11 p.m. ET on Sunday night. That stop remained in effect until around 2 a.m. ET Monday, when the Federal Aviation Administration confirmed it had been lifted. But disruptions didn't end there. Alaska warned passengers to brace for likely delays throughout the day. [...] The FAA's website listed the stop as applying to all Alaska Airlines aircraft. Gizmodo notes that the incident comes nearly a year after the massive 2024 CrowdStrike crash, which has become known as the largest IT outage in history. "The July 2024 outage brought down an estimated 8.5 million Microsoft Windows systems running CrowdStrike's Falcon Sensor software, disrupting everything from hospitals and airports to broadcast networks."
"There's no word yet from Alaska on whether the outage ties into a broader software problem, but the timing, almost exactly a year after the CrowdStrike crash, isn't going unnoticed on social media, with users wondering if the events are related."
"There's no word yet from Alaska on whether the outage ties into a broader software problem, but the timing, almost exactly a year after the CrowdStrike crash, isn't going unnoticed on social media, with users wondering if the events are related."
Re: (Score:2)
Re: (Score:1)
I blame Lincoln. That motherfucker!
Seriously stupid (Score:2)
If you've got a multi-billion dollar company, it's not rocket science to have a live backup system in place that can just be switched over to. Just keep it two system updates behind and ready to switch over at a moment's notice.
Re: (Score:3)
If you've got a multi-billion dollar company, it's not rocket science to have a live backup system in place that can just be switched over to. Just keep it two system updates behind and ready to switch over at a moment's notice.
Doing that would be harder than rocket science. Behind applications there are databases and databases have schema. If not databases, then data access layers and those layers have interfaces. Since system updates can contain schema or interface updates, you can't always just roll back the system without also rolling back the data. Rolling the database back to a time two application versions ago would would lose a lot of information that shouldn't be lost (tickets, seat assignments, aircraft and crew status a
Re: (Score:2)
You need to conduct your transactions on multiple independent systems, simultaneously.
Think of it as RAID for applications and data.
Have your shit redundant, such that if any one goes down, it doesn't matter. Further, when that system comes back up, it's automatically brought up to date by the living systems.
Not cheap, or easy, which is why most places don't do it.
Why would these be related? (Score:2)
Unless Alaska Airlines with very high effort delays Crowdstrike updates by about a year, there is no way for these to be related. They are for sure related on the other aspect of the Crowdstrike disaster though, namely that Windows was crap, is crap and will remain crap and so is anything else by Microsoft.
Solution (Score:1)
The solution is to ban CloudStrike, CloudFlare, DoubleClick and all of these other services that are fucking up the internet.