Crowdstrike fix for blue screen issue Workaround Steps: 1Boot Windows into Safe Mode or the Windows Recovery Environment 2Navigate to the C:WindowsSystem32driversCrowdStrike directory 3Locate the file matching โC-00000291*.sysโ, and delete it. 4Boot the host
— SANS.edu Internet Storm Center (@sans_isc) July 19, 2024
It started on a Thursday USA time
I'm in awe of the scale of the Crowdstrike / Windows BSOD issue.
Here are the most startling images I've seen morning.
Let's start with this: at 10pm PT yesterday, famous @troyhunt notices that something odd is happening to Windows systems:https://t.co/0KWDELUycT
Note "channel updates …bypassed client's staging controls and was rolled out to everyone regardless" https://t.co/UecaAmJdqc
A few IT folks who had set the CS policy to ignore latest version confirmed this was, ya, bypassed, as this was "content" update (vs. a version update)
"They pushed a new kernel driver out to every client without authorization to fix an issue with slowness and latency that was in the previos Falcon sensor product. They pissed over everyone's staging and rules and just pushed this to production"https://t.co/XVEJoLTBeMhttps://t.co/eYq3Fy0fAS
— ๐ฆ SchizoDuckie ๐ฆ (@SchizoDuckie) July 19, 2024
https://t.co/daXQLipeLv "This is going to turn out to be the biggest cyber incident ever in terms of impact, just a spoiler, as recovery is so difficult," says one expert
Am I reading this right? This news story came out _yesterday_ about how companies are getting sloppier with reviewing major app updates. The story was based on research done to promote a cybersecurity company called……. CrowdStrike https://t.co/hQ85SPEsmz
Crazy visual: 12-hour timelapse shows plane traffic over the US with the FAA grounding Delta, United, and American Airlines flights during this morning's outage pic.twitter.com/KRuL3HjZVf
CrowdStrike CEO is getting pummeled for his response to the global outage.
Why everyone hates it:
1) WEAPONS-GRADE CORPO SPEAK
Letโs be clear. Legalese doublespeak is designed to dodge and obfuscate rather than inform or communicate. This statement was obviously written by aโฆ pic.twitter.com/oLua908QR2
This statement conveys that the CEO thinks youโre overreacting. Everyone calm down; it was only a global outage that took down emergency rooms and the London Stock Exchange.
The smart thing in light of the Crowdstrike global outage is to look not to Crowdstrike, but your own company:
What happens when someone (anyone!) pushes code that passes all internal tests but crashes prod for most customers? When do you discover it? Is it before customers?
Hereโs the thing folks. Iโve been coding 32 years. When something like this happens itโs an organizational failure. Yes, some human wrote a bad line. Someone can โgit blameโ and point to a human and itโs awful. But itโs the testing, the Cl/CD, the A/B testing, the meteredโฆ
For those who don't remember, in 2010, McAfee had a colossal glitch with Windows XP that took down a good part of the internet. The man who was McAfee's CTO at that time is now the CEO of Crowdstrike. The McAfee incident cost the company so much they ended up selling to Intel. pic.twitter.com/DgWid6MSK0
Seems legit. Per LinkedIn he worked in 2010 as McAfee CTO and in April 2010 there was a faulty antivirus update that sent WinXP into a reboot loop.https://t.co/2zOyRun6P8https://t.co/hlq45zIWIA
We created a no-prompt bootable ISO with WinPE that auto-deletes the bad crowdstrike file. Then automount to VDI machines and have them boot to it. We've done hundreds this way.
Rebooting 3 and up to 15 or more times is working on a large percentage of machines. It appears that sometimes the network stack is up long enough and crowdstrike update mechanism is able to fix the broken .sys file. Try rebooting over and over and over and over. Seriously.
How we did this in the old days: When I was on Windows, this was the type of thing that greeted you every morning. Every. Single. Morning.
You see, we all had a secondary "debug" PC, and each night we'd run NTStress on all of them, and all the lab machines. NTStress wouldโฆ pic.twitter.com/rZkvpujbcr
Yup. Config (and input data generally) is just another form of control flow, only it's chunks of your application code that are the control flow primitives. Data format interpreters are not fundamentally different to virtual machine interpreters.
(Others may have mentioned this?) but we find many references "channel files" in @CrowdStrike's patents that provide more insight into their purpose, format, etc.
I worked as a Linux distro dev 25 years ago. We tried migrating the whole company to our OS, but our core business functions like sales & HR could not work so we switched them all back. Even if you migrated only servers, business users would still be on Windows. Stop fantasizing.
— Katie๐ปMoussouris (she/her/she-ra/she-hulk) (@k8em0) July 20, 2024
So am I right in thinking the whole thing about that .sys file not being a kernel driver is bullshit because itโs a configuration file for a kernel driver?
IMHO the root of today's kernel issues with Windows go back to Windows NT 3.51. Then MS isolated the kernel from 3rd party drivers but the result was games, printer drivers, and AV sw stopped working & would have to be rewritten. So MS relented & changed architecture with NT 4.0
Too bad many webcare / social media corporate accounts respond to every question (even ones of importance to the general public) along these lines:
we extremely apologize for the challenge encountered, customer relations haven’t forgotten about you, it wasn’t our intention, kindly follow back, share full names, reachable WhatsApp number via DM, so we can assist further
When you join a Wi-Fi network that shows a captive portal on the iPhone, iOS opens the captive portal in a Safari web view regardless of which default browser you’ve chosen.
This little experiment implements a captive portal that detects the browser. Brave was the default browser, but iOS opened the captive portal in Safari. The second screenshot shows what this sample website displays if opened with Brave.
Sergey is the single maintainer of fail2ban, the open source project that protects countless (likely a majority) ofย public facing servers facing on the Internet.
Please don’t let fail2ban become another XZ Utils and support Sergey: we don’t want the project to become unmaintained, or worse: being backdoored like XZ was.
I can’t even comprehend how many servers are protected by fail2ban, how many compromises are avoided, how many people who run hobby things all the way up to major sites that get to sleep soundly every night… because of this single project.
Puppeteer is a Node library which provides a high-level API to control Chrome or Chromium over theย DevTools Protocol. Puppeteer runsย headlessย by default, but can be configured to run full (non-headless) Chrome or Chromium.
Crawl a SPA (Single-Page Application) and generate pre-rendered content (i.e. “SSR” (Server-Side Rendering)).
Automate form submission, UI testing, keyboard input, etc.
Create an up-to-date, automated testing environment. Run your tests directly in the latest version of Chrome using the latest JavaScript and browser features.
Capture aย timeline traceย of your site to help diagnose performance issues.
One reason is archival of books. Even (or maybe especially) in IT, books already have historic meaning especially in narrower fields where they often are not available in the Internet Archive or have been scanned by Google Books.