cruising for burgers Posted July 19 Share Posted July 19 in which thread are u discussing this? Link to comment Share on other sites More sharing options...
zazen Posted July 19 Share Posted July 19 can't discuss it, my computer wont boot up 4 Link to comment Share on other sites More sharing options...
BobDobalina Posted July 19 Share Posted July 19 Have you checked your motherboard for TCP/IP interrupts at the kernel level? Link to comment Share on other sites More sharing options...
cruising for burgers Posted July 19 Author Share Posted July 19 yeah kermit levels, s/n ratio levels, thd levels, boss level, every level bruv... 1 Link to comment Share on other sites More sharing options...
Silent Member Posted July 19 Share Posted July 19 Its Y2K bug, just as nosterdamus predicted in the Mexican calendar, the ends is here. Döner Trump, beamus up. to the big blender in the sky we fly. Hulk Hogan I love you. Kisses 1 2 Link to comment Share on other sites More sharing options...
Scaramouche Posted July 19 Share Posted July 19 It caused a major incident at work but I wasn't personally impacted. There were/are some very stressed people though. Good excuse for not travelling into the office today, just in case. All social commitments cancelled, just to be sure. News highlight so far: 1 2 1 Link to comment Share on other sites More sharing options...
aencre Posted July 19 Share Posted July 19 A single software update bug caused by a capitalistic monopoly can singlehandedly paralyse the whole international economy and provoke financial panic, but the main risk for economic stability is... moderate left-leaning parties? 5 1 2 Link to comment Share on other sites More sharing options...
auxien Posted July 20 Share Posted July 20 https://www.theverge.com/24202037/microsoft-crowdstrike-outage-blue-screen-error-photos Link to comment Share on other sites More sharing options...
user Posted July 20 Share Posted July 20 12 hours ago, BobDobalina said: Have you checked your motherboard for TCP/IP interrupts at the kernel level? Heh. Funny when you consider how many people leave the BAUD rate of their SCSI ports unlocked. Link to comment Share on other sites More sharing options...
EdamAnchorman Posted July 20 Share Posted July 20 16 minutes ago, user said: Heh. Funny when you consider how many people leave the BAUD rate of their SCSI ports unlocked. These are the same type of jagoffs who play fast and loose with their IRQs, RAID their PnP devices, and disable their pagefile. Link to comment Share on other sites More sharing options...
Rubin Farr Posted July 20 Share Posted July 20 Link to comment Share on other sites More sharing options...
Rubin Farr Posted July 20 Share Posted July 20 1 Link to comment Share on other sites More sharing options...
cichlisuite Posted July 20 Share Posted July 20 (edited) well they really did strike the crowd.... Edited July 20 by cichlisuite 1 Link to comment Share on other sites More sharing options...
cruising for burgers Posted July 20 Author Share Posted July 20 5 hours ago, Rubin Farr said: 1 Link to comment Share on other sites More sharing options...
zkom Posted July 20 Share Posted July 20 2 1 Link to comment Share on other sites More sharing options...
zkom Posted July 20 Share Posted July 20 Some rumors I've seen is that CrowdStrike laid off a large part of their QA team earlier this year and the management "didn't believe in test automation" but instead relied only on manual testing. The bug itself seems to be a null pointer issue. Anyway, the question that how did this big of a fuck up get deployed in so many large scale and critical systems without anyone catching the issue before deployment is mind-boggling. 5 Link to comment Share on other sites More sharing options...
zazen Posted July 20 Share Posted July 20 Yes anything that auto-updates, theres a big responsibility on the makers to test properly. Especially if its a low level thing like an OS or security software that sits close to the kernel. They should have a whole battery of tests, different environments and then phased roll outs with a close eye kept looking out for problems. You can bet that Apple, Microsoft and Google/Android test their OS updates exhaustively. Sometimes their updates break things but its usually subtle or obscure. CrowdStrike were supposed to be very trustworthy which is why lots of highly-regulated industries like airlines and banks and hospitals were using them. They literally had an update that instantly blue screens any PC it touches, and somehow they sent it out to the whole world without fucking noticing? Complete cowboys. And they are supposed to be enterprise level trusted purveyors of security software. It'll be interesting to see how CrowdStrike try and justify it (they need to think of the shareholders, after all) but if there's any justice in the world CrowdStrike deserve to go bankrupt the minute they've finishing cleaning up this global bullshit that they've caused. 1 Link to comment Share on other sites More sharing options...
zazen Posted July 20 Share Posted July 20 This shit is serious 4 Link to comment Share on other sites More sharing options...
chaosmachine Posted July 21 Share Posted July 21 Classic mistake, this is exactly what happens when you update your routing table after freeing memory: dangling pointer references all over your BGP stack leading to kernel panic. They should be using a memory-safe language like Lua or Haskell. Fortunately it's an easy fix, just run the latest terraform ansible playbook on the docker cluster, wipe the container registry, and push out the images. As long as there's no md5 collisions on the new pods, most self-healing serverless job queues should reinitialize in under a few minutes, and you should see stable herd cohesion in under an hour if you're provisioned correctly. If not, you can bypass the kernel by holding left alt and the f8 key while rebooting. 1 Link to comment Share on other sites More sharing options...
auxien Posted July 21 Share Posted July 21 14 minutes ago, chaosmachine said: Fortunately it's an easy fix, just run the latest terraform ansible playbook on the docker cluster, wipe the container registry, and push out the images. As long as there's no md5 collisions on the new pods, most self-healing serverless job queues should reinitialize in under a few minutes, and you should see stable herd cohesion in under an hour if you're provisioned correctly. 3 1 Link to comment Share on other sites More sharing options...
EdamAnchorman Posted July 21 Share Posted July 21 (edited) 3 hours ago, chaosmachine said: Classic mistake, this is exactly what happens when you update your routing table after freeing memory: dangling pointer references all over your BGP stack leading to kernel panic. They should be using a memory-safe language like Lua or Haskell. Fortunately it's an easy fix, just run the latest terraform ansible playbook on the docker cluster, wipe the container registry, and push out the images. As long as there's no md5 collisions on the new pods, most self-healing serverless job queues should reinitialize in under a few minutes, and you should see stable herd cohesion in under an hour if you're provisioned correctly. If not, you can bypass the kernel by holding left alt and the f8 key while rebooting. Edited July 21 by EdamAnchorman 1 2 Link to comment Share on other sites More sharing options...
Scaramouche Posted July 21 Share Posted July 21 (edited) 6 hours ago, chaosmachine said: Classic mistake, this is exactly what happens when you update your routing table after freeing memory: dangling pointer references all over your BGP stack leading to kernel panic. They should be using a memory-safe language like Lua or Haskell. Fortunately it's an easy fix, just run the latest terraform ansible playbook on the docker cluster, wipe the container registry, and push out the images. As long as there's no md5 collisions on the new pods, most self-healing serverless job queues should reinitialize in under a few minutes, and you should see stable herd cohesion in under an hour if you're provisioned correctly. If not, you can bypass the kernel by holding left alt and the f8 key while rebooting. “You fell victim to one of the classic blunders! The most famous of which is, ‘never get involved in a land war in Asia,’ the second is never update your routing table after freeing memory: dangling pointer references all over your BGP stack leading to kernel panic, but only slightly less well-known is this: ‘Never go in against a Sicilian when death is on the line!’” Edited July 21 by Scaramouche 2 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now