Google App Engine struggles against bug

By Tim Conneally | Published June 18, 2008, 1:54 PM

Since early morning yesterday, Google's App Engine Web application hosting service has been forced into a limp, yielding persistent errors for users attempting to access their applications.

In the App Engine Forum at 6:35 pm PT yesterday, a team member posted a brief explanation of why users were having difficulty: "This outage was the result of a bug in our datastore servers and was triggered by a particular class of queries. We have isolated the bug and we're currently working on a fix. Going forward, we're also working to further isolate queries so that in the future a bug like this won't affect the stability of the system as a whole."

Apparently troubles began 12 hours prior to this post, and were not isolated until 1:40 pm. Google has not yet announced a fix.

Google App Engine opened to the public in May of this year, following a test period with a cast of 10,000 developers. The service allows developers to host their Web-based applications on Google's infrastructure, offering 500 MB of persistent storage for free and "enough bandwidth for 5 million monthly page views." Applications are implemented using the programming language Python, no other languages are yet supported.

Google's cloud services still need time to mature, as Amazon's Web-based infrastructure has illustrated. The online retailer's "cloud" services have been around for longer, but also suffer from periodic downtime. An incident in February of this year Amazon's Simple Storage Service went down for about two hours, taking with it large chunks of sites like SmugMug and Twitter. Again, two months later, an hour-long outage affected Elastic Compute Cloud customers across the country.

While these services do not yet guarantee 100% uptime, many customers' services which are built upon them often require it, and even a single hour of downtime can be a crushing blow to their users.