Major Outage At the Amazon Web Services 247
ralphart writes "The Northern Virginia datacenter for Amazon Web Services appears to be having a major outage that affects EC2 services. The Amazon Forums are full of reports of problems. Latest update from the status page: 2:49 AM PDT We are continuing to see connectivity errors impacting EC2 instances, increased latencies impacting EBS volumes in multiple availability zones in the US-EAST-1 region, and increased error rates affecting EBS CreateVolume API calls. We are also experiencing delayed launches for EBS backed EC2 instances in affected availability zones in the US-EAST-1 region. We continue to work towards resolution."
Emergency Plan (Score:5, Interesting)
I didn't even realize that one of our partners was using Amazon EWS until suddenly they were down all day. Amazon is really stable historically, but it's frustrating when you're out of business and all you can do is wait and see if Amazon will fix it soon.
In the "old school" thinking, smart companies have a redundant data center somewhere, humming along and waiting to be switched on if the main data center ever goes down. "The cloud" was supposed to solve that - massive redundancy within Amazon's services were supposed to protect you from outages. Not the case, apparently, since it looks like Amazon is going to fall below their promised 99.95% uptime (4.38 hours per year downtime).
I think the answer is to have redundant cloud services online, so you could switch from Amazon to Google or DevGrid if you had issues. The problem is, there's nothing quite like Amazon right now, it's not easy to switch from Amazon to some random service. This might be the biggest argument against virtual services - lack of standardization makes it hard to move from one to another, and hard to set up backup services in case of emergency.
6 weeks before the AWS summit 2011 (Score:4, Interesting)
Gotta wonder what kind of flack Amazon is going to take for this one. I've had a couple clients looking into cloud services including moving to AWS and have already had one of them call me and cancel a meeting about it. While I understand stuff happens, the entire sales pitch for AWS was redundancy and build as you grow. Redundancy has obviously not worked in this case, while I usually support cloud services, this is definitely going to be a hard example to counter when trying to sell it to potential customers.