Ask HN: Is the Delta outage a 'bug' or a malicious hack? | Hacker News Reader

Ask HN: Is the Delta outage a 'bug' or a malicious hack?

Today, Delta was the last company being hit by some kind of system-wide outage, causing hundreds of flights cancellations. Last week it was Southwest suffering from some mysterious system-wide "bug" that caused mayhem, and last year United had a similar system-wide crash.

My question is: is there a pattern? Are airlines targeted, perhaps as part of either proving our infrastructure is hackabale, or an actual act of sabotage? Or is just that all airlines run old, buggy software whose shelf life has expired?

View on HN

I am really surprised they don't have a high availability setup for their database, etc being a fortune 500 company. Hours of downtime is serious money for them it would seem. Well if they did have such setup, it failed.

Even a bootstrapped startup could get a MongoDB Cluster or something similar going in multiple datacenters.

The power company Tweeted about it also. https://twitter.com/GeorgiaPower/status/762662968572215296

So I guess it was a power outage only at their place...

But the whole idea of not having working redundancy in a multibillion dollar business really makes me question their story.

So not to sound like a conspiracy theorist, but I do think there is a possibility they got hacked. And if that hunch is true then calling it a power outage instead of admitting it is a insult to our intelligence.

> But the whole idea of not having working redundancy in a multibillion dollar business really makes me question their story.

It really isn't that far fetched, IT is a cost to be minimized to them.

According to the Ars Technica article http://arstechnica.com/business/2016/08/data-center-disaster... a fire broke out in their main datacenter. Assuming that was true, employees were most likely evacuated and at the very least would have been distracted. Their disaster recovery plan may have assumed that there would be admins onsite to run a manual switch, may have assumed a complete failure instead of a partial failure or may have had per application disaster recovery plans. Disaster recovery is a difficult and stressful process in normal times, much more so when your regular workstation(s) are unavailable and you're faced with a disaster-level distraction like a fire.

Reply on news.ycombinator.com