AWS postmortem: a first-hand account of the outage

The AWS outage dealt a massive blow to companies worldwide. To understand the full effects, Cybernews spoke to engineering experts who were directly impacted.

Issues affecting Amazon Web Services (AWS) caused massive outages affecting a multitude of services, including Signal, Snapchat, Amazon, Ring, and many others.

It started at night for Honeycomb.io’s engineering experts Stephanie Hippo and Emily Nakashima.

“I was blissfully asleep,” Platform Engineering Director Stephanie Hippo told Cybernews, “but when I woke up in the morning and signed on, I actually had to hop right into a meeting with Emily.”

That’s when Hippo knew that something serious was going on.

As it was the middle of the night, Hippo and Nakashima’s team of engineers was working tirelessly, rotating responsibilities so that each member could get some rest.

The outage started at around 3 a.m. (ET) on Monday, the 20th of October, 2025. It began with a select few AWS services experiencing problems, until the incident escalated to over 70 services being impacted at its peak.

However, Hippo and Nakashima quickly figured out that the issues were with the AWS servers, not their internal systems.

“Within the first couple of minutes, Alania, our platform engineer, was being paged to investigate the incident,” Hippo told Cybernews.

“It was clear very quickly that something was happening upstream.”

However, the outage “wasn't severe enough” to wake up Senior Vice President of Engineering at Honeycomb, Emily Nakashima, in the middle of the night.

“If our incident responders are responding and taking care of the situation, then generally things don't escalate to me. But as soon as I woke up in the morning, I saw that in the night, one of our platform engineering team members had been paged because she noticed something anomalous within Honeycomb,” Nakashima told Cybernews.

Curious what others think about this story? Contribute your thoughts to the debate below.

Honeycomb was able to get more information from AWS directly, as their own incident responders were communicating with companies via their status pages.

“We were able to get better details on what would be the right next step. So, between our tool and the communication of the AWS team, we were able to decide the next steps,” Hippo said.

Nakashima explained what can happen in an outage like this and how Honeycomb mitigated the effects while not creating any additional load, which could further strain operations.

“We were reallocating services behind the scenes…to make sure the critical parts of our system are running, and not every team has the capability to do that quickly…we were able to make the best of the incident.”

What happened during the AWS outage?

Thousands of websites, platforms, and applications were down on Monday, the 20th of October, due to a DNS issue within AWS.

Signal, Snapchat, Fortnite, Starbucks, Reddit, Coinbase, Ring, Amazon, Amazon Alexa, Apple TV, and Apple Music were down for tens of thousands of users.

The reason many sites and apps were affected by the AWS outage is that apps and websites like Snapchat, Fortnite, and massive streaming companies, including Netflix and Twitch, use Amazon’s cloud-based services to host their apps.

Many of these applications and websites are hosted on Amazon’s web services, meaning that when Amazon’s servers have issues or experience an outage, a domino effect occurs.

AWS identified the possible root cause of the incident in the early morning, which was “related to DNS resolution of the DynamoDB API endpoint in US-EAST-1.”

DNS resolution involves translating domain names, such as "www.amazon.co.uk," into machine-readable IP addresses.

The DNS issue has supposedly been fully resolved, and “most AWS Service operations are succeeding normally.”

Unlock more exclusive Cybernews content on YouTube.