For those who like to maintain some degree of privacy, Snapchat is a top pick especially if you have something snarky to say to your friend and like to include a photo. Recently, an AWS outage caused the popular app to go dark for a while along with Reddit, Fortnite, and a few other services.
Amazon Web Services is a cloud computing platform that powers apps like Snapchat and, when it goes down, it feels like the world has come to an end.
One reason for that is we have come to the point in the evolution of technology where there is an expectation of persistent service. When we glance at Facebook or check our Gmail, we expect those apps to work as reliably as the electricity in our homes or cell service for an iPhone.
After the AWS outage, Amazon was quick to point out that the cloud computing platform was back online within hours, although they mentioned that some users may not be able to connect right away. The outage was caused by an “operational” disruption at data centers in Virginia and other locations. I contacted Amazon about the outage for more details but have not heard back yet.
Other apps and services that were impacted include the Amazon Ring doorbell camera and the McDonald’s app. Users on the East Coast and other areas reported not being able to see friend lists or daily streaks in Snapchat, a sure sign that something is awry with the backend cloud service since it means app data is not available.
Why the AWS outage is cause for concern
NBC reported that the outage was caused specifically by Amazon DynamoDB, which is a database service that maintains all of the tables for things like those Snapchat friends lists. In total, there were 6.5 million reports from apps, services, and websites across the globe about the AWS outage that could impact popular apps like Snapchat.
Is it really the most troubling thing to lose access to Snapchat and Reddit? For millions of users, the tension is not from a disruption in banking or other vital services. At the same time, the consternation comes from our perception that these services are almost infallible. We rely on them as a social construct to stay in regular communication with friends, so losing the ability to do that feels like losing your voice or dropping your phone off a cliff. It’s hard to know when things will go back to normal and when you can resume your regular social communication.
Reddit in particular has grown tremendously over the last year. Daily active users now number around 110 million, which is 21% higher than last year. Importantly, users are also spending more time on Reddit. They post and chat on the service for about 34 minutes per day on average, which is about a 3% increase from last year. This means, when Reddit goes down, users have to find a different way to engage with other users.
Snapchat has also gained users globally even in the last quarter, although there’s been a decline in the United States. The app became easier to use about a year ago thanks to some design changes meant to draw in a larger crowd.
What to do when there’s an AWS outage
Many of these hardened services that we rely on, from Reddit to Snapchat, are often available with a click of a mouse or a quick finger tap. They are part of our daily digital routine. When we lose access, it can feel like a monumental blockage. The best course of action, though, is to simply wait it out. Amazon had already restored AWS within a short period of time once the outage became widely known.
Another tip is to find alternate methods of communicating during an outage. For example, while Snapchat was down, you could still send private messages using Instagram. Reddit is not the only discussion board around. Quora is a good option as an alternative, although that company reportedly also uses AWS. It’s worth maintaining an account on multiple social media channels even if they are all hosted by AWS because you may regain access to one app faster than another.
In the end, these apps and services have become like electric power or the internet in our homes. Then again, both of those can also experience disruption similar to an AWS outage. Fortunately, this disruption seems like a temporary blip.