ec2 Archive

On Tuesday morning, hosting service Media Temple was out of service while the company dealt with a DNS resolution issue. Media Temple staff provided an overview of what happened in the comments section of my post.

Now today Amazon Web Services is down — actually only parts of the Amazon Web Services cloud computing offering is down. I use their S3 storage service and that is up and running. As of 9:00AM Eastern, the following message provides the current status related to the downtime, “6:09 AM PDT EBS API errors and volume latencies in the affected availability zone remain. We are continuing to work towards resolution.”

It’s interesting to watch bloggers including Martin Bryant and Mike Butcher only point to the same set of sites (Quora, Foursquare, Reddit and Hootsuite) that use Amazon’s EC2 and are down. There are plenty of other sites that are down including Easybib, a site run by a close friend of mine that is the largest online bibliography site in the world. And Easybib is just one example of time-critical services that are affected. So while it’s easy for us to look at a few social media services that are affected, just think about all of the business, professional and utility services that are also currently down.

Update 10am Eastern – it looks like the problems are expanding – here’s the latest update:

6:59 AM PDT There has been a moderate increase in error rates for CreateVolume. This may impact the launch of new EBS-backed EC2 instances in multiple availability zones in the US-EAST-1 region. Launches of instance store AMIs are currently unaffected. We are continuing to work on resolving this issue.
7:40 AM PDT In addition to the EBS volume latencies, EBS-backed instances in the US-EAST-1 region are failing at a high rate. This is due to a high error rate for creating new volumes in this region.

You can follow the updates around the Amazon EC2 issues on the Amazon Health Dashboard. As with all downtime, please feel free to report in the comments if your site is down due to the Amazon issue.

Last week Amazon Web Services (that’s the cloud computing division of Amazon) announced a reduction in pricing for the largest versions of their EC2 server offering.

Today Amazon has launched a new version of their EC2 server — the “micro instance”.

Amazon’s Jeff Barr explains how the Micro instances work, “Micro Instances (t1.micro) provide a small amount of consistent CPU resources and allow you to burst CPU capacity when additional cycles are available. They are available now in all Regions. You can buy Reserved Micro Instances and you can acquire Micro Instances on the Spot Market. Interestingly enough, they are available in both 32 and 64 bit flavors, both with 613 MB of RAM. The Micro Instances have no local, ephemeral storage, so you’ll need to Boot from EBS.”

It looks like the Micro instances could also be a good option for starting with EC2 as you can always move to one of the larger server offerings if the need arises.

Jeff also notes that the Micro instances could be used for non-traditional uses including:

1. DNS servers, load balancers, proxies, and similar services that handle a relatively low volume of requests.
2. Lightweight cron-driven tasks such as monitoring, health checks, or data updates.
3. Hands-on training and other classroom use.

Pricing for the Micro instances starts at $0.02 (two cents) per hour for Linux/Unix and $0.03 (three cents) per hour for Windows.