Tumblr and Google App Engine down as US traffic plummets

The sky is falling!

12 Reg comments Got Tips?

Updated There are a fair few people cursing their displays and tablets this morning after a series of outages took down Google's App Engine, as well as the Tumblr blogging site and a host of smaller operators.

"Early this morning App Engine began experiencing slow performance and dropped connections," a spokeswoman told El Reg via email. "The Google App Engine team is diligently investigating this issue now and is working on restoring the service for affected users." Further updates are promised as the situation develops.

Google's problems kicked in at around 8:30am Eastern Time on Friday, with the Tumblr blogging site tumbling down at about the same time. Access problems at Dropbox and Instagram have also been reported, although both services appear to be up at running at time of our going to press.

Coinciding with the outages, the amount of US internet traffic took a major hit. According to data from Internet Traffic Report, the amount of traffic over US networks dropped by around 10 per cent this morning, with packet-loss rates soaring at around the same time.

So far there's no word as to the cause of the drop in traffic. Friday is traditionally one of the slower days of the week for many online properties, but not on this scale.

US internet traffic

Where have all the packets gone?

Meanwhile, service providers and data centers on the US East Coast are bracing themselves for the impact of Hurricane Sandy, which is due to come ashore at around midnight on Friday.

The category-two hurricane is most likely to hit in the New Jersey or New York area, and there's a lot of processing centers in its predicted path. After Amazon's outage earlier in the week, cloud providers could soon be making payouts over service-level agreements. ®

Update

Google reports the App Engine is now back up and running:

"Google App Engine has now been restored and users should see service returning to normal. Our team is still continuing to investigate and determine the root cause of the issue. We know that many of our customers rely on App Engine for their mission critical applications. We apologize for the inconvenience caused by this outage and we appreciate our customers’ patience," Google said.

Meanwhile Internet Traffic Report figures show that US data flows are now back up to normal levels.

SUBSCRIBE TO OUR WEEKLY TECH NEWSLETTER


Keep Reading

Hardworking Americans keep busy during the government shutdown driving up smut traffic

NSFW links You can make your own Yank joke here

Watch your MANRS: Akamai, Amazon, Netflix, Microsoft, Google, and pals join internet routing security effort

Filtering, anti-spoofing, coordination, validation to prevent crooks, spies hijacking victims' connections

Forget toilet roll, bandwidth is the new ration: Amazon, YouTube also degrade video in Europe to keep 'net running amid coronavirus crunch

Pair join Netflix to turn it down for what? Connectivity stability

Amazon's not saying its warehouse staff are dumb... but it feels they need artificial intelligence to understand what 'six feet' means

Vid The yellow markings on the floor aren't enough for real neural networks

Taiwan may turn traffic advice app into massive tracking system

To offer social distancing advice when people head out to play on a looming long weekend

As Amazon's stock price soars and Bezos adds to his billions, affiliates face massive cuts in their commissions

Times are hard...for you at least

Highways England waves around £62m contract for National Traffic Information Service after brief chat with vendors

Updated If you're anything like us, you dream of reorganising your country's highways. This is a new system they want to tighten up data streams, but still

IBM blames 'external' network provider, incorrect routing, traffic flood for its two-hour cloud outage

No data loss or attack detected. But aren't hyperscale clouds supposed to be more resilient than this?

Biting the hand that feeds IT © 1998–2020