Topic: Why is reddit down so much?
Anonymous A started this discussion 14 years ago #22,931
???
Anonymous B joined in and replied with this 14 years ago, 19 minutes later[^] [v] #359,246
It runs on Ubuntu servers. Not Debian, not RedHat, but Ubuntu.
It's pretty funny really. I can excuse Ubuntu use on a desktop, since it is the easiest install of them all, but for a server? I don't get it. It's certainly not the most reliable flavor of GNU/Linux.
It's pretty funny really. I can excuse Ubuntu use on a desktop, since it is the easiest install of them all, but for a server? I don't get it. It's certainly not the most reliable flavor of GNU/Linux.
(Edited 40 seconds later.)
doctor dreamworks md !jdTkvovFCQ joined in and replied with this 14 years ago, 2 minutes later, 21 minutes after the original post[^] [v] #359,247
Barack Obama
Killer Lettuce !!iNo3FkiZx joined in and replied with this 14 years ago, 1 minute later, 23 minutes after the original post[^] [v] #359,249
Valdair.
Anonymous E joined in and replied with this 14 years ago, 11 minutes later, 34 minutes after the original post[^] [v] #359,261
Anonymous B replied with this 14 years ago, 1 minute later, 36 minutes after the original post[^] [v] #359,263
Anonymous E replied with this 14 years ago, 10 minutes later, 46 minutes after the original post[^] [v] #359,274
Anonymous F joined in and replied with this 14 years ago, 26 minutes later, 1 hour after the original post[^] [v] #359,285
Anonymous G joined in and replied with this 14 years ago, 9 minutes later, 1 hour after the original post[^] [v] #359,298
ITT: bullshit layman speculation
It's a problem with the complexity of their application architecture, not the hardware or OS. If they could afford more hardware, the high loads would be less problematic.
Also, they're using a NoSQL cluster as their back end. Those have been shown inseparable from effects of the CAP theorem. They have probably chosen availability and partition-tolerance over consistency.
TL;DR: extremely high throughput with complex architecture and significant budget constraints means their load balance system rejects requests and their cluster has genuine failures
It's a problem with the complexity of their application architecture, not the hardware or OS. If they could afford more hardware, the high loads would be less problematic.
Also, they're using a NoSQL cluster as their back end. Those have been shown inseparable from effects of the CAP theorem. They have probably chosen availability and partition-tolerance over consistency.
TL;DR: extremely high throughput with complex architecture and significant budget constraints means their load balance system rejects requests and their cluster has genuine failures
Killer Lettuce !!iNo3FkiZx replied with this 14 years ago, 41 minutes later, 2 hours after the original post[^] [v] #359,321
@previous (G)
How is Valdair "layman speculation"? That man is pure evil. If there's trouble on Reddit then he's more than likely to be the cause.
How is Valdair "layman speculation"? That man is pure evil. If there's trouble on Reddit then he's more than likely to be the cause.
doctor dreamworks md !jdTkvovFCQ replied with this 14 years ago, 11 minutes later, 2 hours after the original post[^] [v] #359,323
it's back!
Anonymous G replied with this 14 years ago, 1 minute later, 2 hours after the original post[^] [v] #359,326
doctor dreamworks md !jdTkvovFCQ replied with this 14 years ago, 10 minutes later, 2 hours after the original post[^] [v] #359,329
it's down!
Killer Lettuce !!iNo3FkiZx replied with this 14 years ago, 2 minutes later, 2 hours after the original post[^] [v] #359,330
Anonymous H joined in and replied with this 14 years ago, 1 hour later, 3 hours after the original post[^] [v] #359,349
Down now Up? 4Sure Reddit has income issues. Back end income continues to crash dive off cliffs. Suspect advanced hardware is hard to cum by.