Apparently, Amazon Decided On A Long Weekend; Goes Down Again
from the life-on-the-amazon dept
While much of the focus on “uptime” today was over whether or not various web 2.0/tech sites could handle the onrush of traffic concerning the Steve Jobs/iPhone keynote, it appears that Amazon.com was actually stumbling and having trouble staying up just a few days after its still unexplained Friday meltdown. It’s unclear if the two events are related, but given that Amazon is positioning itself to help keep other sites up and running, no matter what the situation, this isn’t the sort of ringing endorsement it was probably looking for.
Comments on “Apparently, Amazon Decided On A Long Weekend; Goes Down Again”
Yeouch! I really don’t envy their server maintenance guys right now; if called upon to speculate, I’d say that whatever happened on Friday was absolutely catastrophic, and they threw together a hacky temporary fix that they’re still trying to replace with something permanent. I’d love to know what could have flattened them this badly, though; we’d have heard about it if there’d been a fire in one of their datacentres or something, and one doesn’t earn Amazon’s market share without some very strong backup systems in place.
Hmmm
Time to forward them all my viagra and cialis spam so they can improve their uptime and handle new giant loads that the female persuasion seems to be clamoring for.
isn’t it customary for sites with such a heavy load to automatically forward its users to mirror sites?
to help reduce server loads?
Something more than Amazon is revealing?
It could have been any idiot with a prepaid cell phone calling in, but on Leo LaPorte’s Sunday tech show (radio), someone called in claiming this had to be a very sophisticated denial of service attack that could have only been initiated by someone with hundreds of computers available and the skill to initiate this kind of a blitz on a retailer like Amazon. When LaPorte pressed the man for details, asking why the backup systems that Amazon must have had in place didn’t work, the callerclaimed that this DOS attack was enough to also take out all of the backup/mirror servers that Amazon usually relies upon to keep their servers going in a crisis.
The caller also pointed out that at the same time as Amazon’s first outage, IMDB was also taken offline with an identical service outage problem. Turns out that Amazon owns IMDB, something they don’t go out of their way to reveal to the public. He didn’t sound like a crackpot, BTW.
Re: Something more than Amazon is revealing?
I don’t listen to Leo LaPorte’s radio show, but from what you say the caller on that show said, the caller definitely knew what he was talking about (except that, IMHO, there really isn’t much skill involved).
Re: Re: Something more than Amazon is revealing?
But what exactly do they stand to gain from keeping this under wraps? If what this caller suggested is accurate, it sounds to me like Amazon have nothing to be embarrassed about; it’s one thing to be reluctant to own up to getting taken down by some teenage 4chan regular trying to make his e-penis look bigger, but it doesn’t reflect badly on their infrastructure if it’s not designed to withstand a premeditated act of large-scale sabotage.
“Turns out that Amazon owns IMDB, something they don’t go out of their way to reveal to the public.”
It’s right on imdb.com’s homepage…”An amazon.com company.”
It should be noted that Amazon.com may have been down and/or slow, but by many accounts, their hosting services were not. Maybe you can still call into question their ability to provide the hosting services, but ti doesn’t seem to follow that because Amazon was down (for reasons unknown at this time), they aren’t capable of providing hosting services. Especially given hosting is still up.
Woadan