Crowdstrike outage
-
@Kruse said in Crowdstrike outage:
What self-respecting BANK runs critical systems on Windows servers? WTAF?
Depends what they're doing with their development capabilities and environment management.
For the customer-facing side the deployment is baked in. Software updates are easy - in this case too easy - and you have a widespread and familiar OS that is low-cost compared to alternatives (nobody can seriously suggest Mac OS would be viable in financial terms for fleet).
Once you do that, the path to Windows Server - particularly cloud integration with Azure - is fairly easy.
And let's remember that even if the Servers weren't running the back end, a front end that can't serve customers isn't worth much. That's where the work is: redploying software manually to thousands of Windows PCs and middleware that delivers the critical info to UI.
A lot of financial institutions were still running mainframes until the early 2000s and the ability to support that was eroding. So they had to do something. At the same time, MS uplifted their cloud capabilities because they wanted a cut of Amazon's lunch.
A lot of ATMs are still running XPe because updating them
is notwas not cost effective.We could go back to Terminals, but most of the guys who wrote the software are dead
-
The question isn't: "why Windows?" but "How the fuck did this get past every single test and dev environment without causing an issue?"
(I managed to once get a piece of shit code into a bank system because we didn't test enough decimal places. The day we found it I was off my tits from a work lunch and wouldn't fix it until the next day - fortunately only a couple of hundred thousand dollars affected and the customer never noticed).
-