Amazon's Alexa goes quiet on Echo devices amid AWS disturbances
Alexa, Amazon's digital assistant used by the Echo device range, is suffering from service outages that are affecting a number of users in the United States, issues that are likely to be caused by a regional outage of Amazon Web Services in Northern Virginia. (Update from Amazon)
Some users in the United States have received failure responses from their Alexa-equipped devices for their queries, reports TechCrunch, with the occasional outages affecting both Amazon's Echo devices as well as third-party hardware that uses Alexa, including the Sonos One. Along with verbal apologies and chimes from Alexa, users are also seeing instances of a red light ring, another notification that something went wrong with the query.
DownDetector shows a spike of reported issues with Alexa from users starting from approximately 11am eastern, with thousands of complaints made over the following hours. The site's live outage map indicates the vast majority of the Alexa reports stem from the United States, with a few reports stemming from European users.
Amazon has yet to publicly confirm the issue with Alexa specifically, but it is probably connected to the "Network Connectivity" event on the Amazon Web Services status page. Starting from 7:29am PST (10:29am eastern), the status page notes instances of packet loss affecting AWS Direct Connect customers in the US-East-1 Region, based in Sterling, Virginia.
According to later status updates, the connectivity issues affected connections for multiple Amazon services located in the region. By 9:43am PST (12:43pm eastern), connections between CoreSite VA1 and VA2 as well as other nearby locations are said to be inactive, with Amazon working to restore service through the connections.
A 10:24am PST (1:24pm eastern) update advises some of the network connectivity has been resolved.
According to Twitter user Daniel Pentecost, Amazon's online support team are apparently aware of the problem, and it should be resolved in the next hour or two. It is alleged the phone-based technical support is also being flooded by users, with a reported 18-minute hold time for those waiting to speak to a representative.
The Alexa outage occurs one month after Amazon ran a Super Bowl ad jokingly suggesting it could replace the digital assistant with celebrities if it lost its voice.
Update: An Apple spokesperson reached out to AppleInsider reiterating AWS "did not suffer a widespread or massive outage" during the period. The issue was caused by a loss of power to one of AWS' redundant Internet connection ports in Virginia, which created connectivity issues for AWS customers using Direct Connect services in AWS US East.
AWS has since restored the issue, and says it is working with its partner to prevent a recurrence.
Some users in the United States have received failure responses from their Alexa-equipped devices for their queries, reports TechCrunch, with the occasional outages affecting both Amazon's Echo devices as well as third-party hardware that uses Alexa, including the Sonos One. Along with verbal apologies and chimes from Alexa, users are also seeing instances of a red light ring, another notification that something went wrong with the query.
DownDetector shows a spike of reported issues with Alexa from users starting from approximately 11am eastern, with thousands of complaints made over the following hours. The site's live outage map indicates the vast majority of the Alexa reports stem from the United States, with a few reports stemming from European users.
Amazon has yet to publicly confirm the issue with Alexa specifically, but it is probably connected to the "Network Connectivity" event on the Amazon Web Services status page. Starting from 7:29am PST (10:29am eastern), the status page notes instances of packet loss affecting AWS Direct Connect customers in the US-East-1 Region, based in Sterling, Virginia.
According to later status updates, the connectivity issues affected connections for multiple Amazon services located in the region. By 9:43am PST (12:43pm eastern), connections between CoreSite VA1 and VA2 as well as other nearby locations are said to be inactive, with Amazon working to restore service through the connections.
A 10:24am PST (1:24pm eastern) update advises some of the network connectivity has been resolved.
According to Twitter user Daniel Pentecost, Amazon's online support team are apparently aware of the problem, and it should be resolved in the next hour or two. It is alleged the phone-based technical support is also being flooded by users, with a reported 18-minute hold time for those waiting to speak to a representative.
The Alexa outage occurs one month after Amazon ran a Super Bowl ad jokingly suggesting it could replace the digital assistant with celebrities if it lost its voice.
Update: An Apple spokesperson reached out to AppleInsider reiterating AWS "did not suffer a widespread or massive outage" during the period. The issue was caused by a loss of power to one of AWS' redundant Internet connection ports in Virginia, which created connectivity issues for AWS customers using Direct Connect services in AWS US East.
AWS has since restored the issue, and says it is working with its partner to prevent a recurrence.
Comments
FWIW, Apple uses AWS for iCloud, etc. -- But the only issue they show is Maps Traffic:
https://www.apple.com/support/systemstatus/
Here is Amazon's, which details time and issues with recent events as well as shows a per day history for all their services for various continents.
EDIT: Apple doesn't show you even three days now? If so I couldn't see it.
Apple used MS Azure for iCloud until recently - they just switched to Google Cloud. Do you not read the tech news?
So why do you think Apple is the outlier when it comes to being transparent about the downtimes and frequency? They used to keep at least three days logged. Apparently they don't even do that anymore? You're not offering any possible reason of your own for it.
If no one has done it, it probably isn't that interesting as you think.
2) A PERL script? Really?
3) So you're argument is if it doesn't exist it shouldn't exist or the even worse if Apple isn't doing it, it means it shouldn't be done. Man, that sounds like when you people made claims that Apple would never use OLED, to choose just one example, simply because they hadn't yet.
4) If Google didn't offer a comprehensive system status page but Apple did I don't doubt for a second that you (and many others on this forum) wouldn't bend over backwards to proclaim that Apple was being transparent and Google was being sneaky. At one point I think most of you actually liked technology for its own sake and looked at it through an objective lens instead of this fanatical Apple one. When did this happen to you? Why is there a problem with removing a brand from your decision making in order to look at a situation without bias? These are just fucking companies! They owe you nothing outside the agreed upon product or service agreement.
And that information can actually be pulled easily if you actually care that much about it.
If 99.99% of people don;t gives a shit about something they're not oblige to produce anything for the 0.01% who do.
The whole thing about OLED is specious, OLED now is not the same thing as it was 5 years ago and knowing "uptime" is something that has existed since the damn of computing and the internet. The meaning and means of getting the info and publishing it hasn't changed much at all.
That's why I pulled out Perl , cause it would routinely be used for those kind of systems tasks that far back.
If you want to be cool and use Rust, Swift, Python, Node or Lua whatever works to produce the same thing if it ACTUALLY BOTHERS YOU THAT MUCH.