Amazon's EC2 Servers Go Down Again

[kjk]

http://techcrunch.com/2011/08/08/amazon-ec2-outage/

Are you trying to use the web right now? Just stop. It’s largely broken.

As indicated by about 20 tips in the last few minutes and pretty much all of Twitter, Amazon’s EC2 service appears to be totally down. That means services like Reddit, Heroku, Foursquare, Instagram, Fab, Quora, and many, many others are down.

2nd time in 4 months for Amazon's servers.

One of the realities of the inevitable upcoming cloud computing era-sometimes, stuff just might stop working for a while.

[TimW]

Originally Posted by kjk

http://techcrunch.com/2011/08/08/amazon-ec2-outage/


2nd time in 4 months for Amazon's servers.

One of the realities of the inevitable upcoming cloud computing era-sometimes, stuff just might stop working for a while.

Looks like it was down for 40 minutes. Either it wasn't a complicated problem or they've gotten more efficient at solving problems.

[pidgeon92]

Originally Posted by kjk

One of the realities of the inevitable upcoming cloud computing era-sometimes, stuff just might stop working for a while.

Why should their servers be any more reliable than my internet connection?

[kjk]

Originally Posted by pidgeon92

Why should their servers be any more reliable than my internet connection?

lol! or my AT&T phone/data connection for that matter

But Amazon's EC2 server arrangement seems pretty cool in theory-lots of backups/auto-rerouting/etc...but even a 99% uptime means at some point it won't be up.

My other concern is that Amazon has a LOT of experience in servers, and even they have hiccups. Apple's new iCloud is pretty new-and their MobileMe service didn't inspire a lot of confidence.

[joenunya]

This took out Netflix streaming to me here on the west coast of the US for a few hours (guess at duration) so I fell asleep to old Art Bell shows via Winamp instead of South Park or Family Guy.

[OtterBooks]

ah, that explains the Netflix issue I had.

[elcreative]

Originally Posted by TimW

Looks like it was down for 40 minutes. Either it wasn't a complicated problem or they've gotten more efficient at solving problems.

Yes, that's why they have warned customers using the system that it will be 24-48 hours before they can get their functionality back... guess self comes first... really helps give you faith in the virtues of cloud computing...

[RainingLemur]

Thanks to this, I had to go outside for a change. GET IT TOGETHER, AMAZON!

[OldGunbunny]

Originally Posted by kjk

...but even a 99% uptime means at some point it won't be up.
.

You're right.
1 percent downtime is equal to:

14.4 minutes a day
1 hour, 40 (almost 41) minutes a week.
87 hours, 36 minutes in a year.

for a lot of enterprise networks, 99 percent uptime requirements are on the low side.

[kjk]

Originally Posted by OldGunbunny

You're right.
1 percent downtime is equal to:

14.4 minutes a day
1 hour, 40 (almost 41) minutes a week.
87 hours, 36 minutes in a year.

for a lot of enterprise networks, 99 percent uptime requirements are on the low side.

Nice math work!

The other thing is-what can you really do if it does go down? Amazon et.al. can give you a rebate for the time it was down, but it really isn't a lot.

Netflix recently reimbursed me for when they went down for an evening; I think 3% of my $8 streaming bill? And I think Microsoft gave me a free game when the Xbox Live network was down for two weeks.