Amazon Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

Amazon Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Rohit Mohite (@RohitMo53046335) reported 50 seconds ago

  @AmazonHelp @AmazonHelp, However, my issue is that your team previously claimed no refund was processed, while I have an official email from Amazon stating that the refund was initiated. I am attaching that email here as proof. Please look into this discrepancy and confirm the status.

• 
  Market Participant (@amktparticipant) reported 4 minutes ago

  Amazon is raising at least $25 billion in a bond sale, but says it won't issue more debt this year $AMZN

• 
  Jon Doe (@JoN__D0E) reported 5 minutes ago

  @spreetail robs people using platforms like Walmart and Amazon. I ordered a small above ground pool for my toddler. They shipped a flat empty box to my house and refused to issue a refund or send me the right product. I have ring camera footage proving this. #scam

• 
  Mr. Marshall (@Neo_Marshall) reported 6 minutes ago

  @otomporo8701 Wow, Amazon is fast for you? It’s quite slow for me o, am I doing something wrong?

• 
  Biggs (@Brandonbigger) reported 10 minutes ago

  @amazon just fyi GOFO express has workers using your vests to make deliveries. This is giving them access to areas they wouldn’t and also represents your delivery drivers poorly with their terrible quality. I have ring door cam showing them using your vest as well if needed

• 
  Mahesh Jayanth (@maheshjayanth) reported 12 minutes ago

  @amazonIN I have been delivered a wrong product and not able to return the Order # 171-5575264-1953903. Can you help me resolve the issue? @amazon

• 
  Mike Vancouver 🍁 (@MikeVancouver2) reported 15 minutes ago

  @CodySevertson They haven't taken down all the SN 650 shows from Amazon yet. (I'm curious how long they will be left up).

• 
  Puneet Patwari (@system_monarch) reported 16 minutes ago

  Horizontal scaling: more power, more problems Instead of one big machine, you run many smaller ones. Traffic gets distributed across them by a load balancer (Day 1 of this series). The upside is real: Near-unlimited scaling. Need more capacity? Add more machines. There's no ceiling like there is with a single server. This is how Google, Netflix, and Amazon handle their scale. No single machine could ever do it. Fault tolerance built in. One machine dies? The other nine keep serving traffic. The load balancer just stops routing to the dead one. Your users never notice. This is the big one. Horizontal scaling gives you redundancy for free. Elastic scaling. Traffic spikes on Black Friday? Spin up 50 more machines. Traffic drops on Monday? Scale back down to 10. You pay for what you use. But here's the complexity you're signing up for: Your application has to be stateless. If a user's session lives in the memory of Server 3, and their next request hits Server 7, they're logged out. You have to move state out of individual machines and into shared systems like Redis. You now need a load balancer, and it needs to be highly available too, or it becomes your new single point of failure. Data consistency gets hard. Ten machines reading and writing to the same data means you're now dealing with everything we covered in the CAP theorem thread (Day 7) and replication (Day 9). Debugging spans machines. A request might touch five different servers. Which one had the bug? You need centralised logging and distributed tracing just to find out what happened.

• 
  RingleaderVon 🐶🐑🏳️‍⚧️ (@VonyaVG) reported 16 minutes ago

  @LucifersFavDuck I think it's Amazons doing, not Vivziepop, Amazon took down the New alastor Song without Viv having a say in it so i think the merch ban was Amazon too

• 
  ˙⋆✮ Miri ♱ ༻📻🪦 (@Shojosacrifice) reported 17 minutes ago

  @ivypie07 It’s perhaps a knock on effect from Amazon taking down Black Gryph0n’s video, which was worked on by Spindle employees. If so, it’s awful that it’s come to this. I really feel for the artists too, they created something fantastic because of their love for the shows they work on and their reward is this? It’s so gross.

• 
  Gatoexpert 🇧🇷 (@gatosabidobr) reported 19 minutes ago

  What a goal from Elon Musk here in Brazil ANATEL approved the starlink signal directly on the cell phone. Imagine yourself in the middle of the Amazon or another forest and your cell phone working. As long as there is energy you will be saved. Is the problem that the other operators will lose to Starlink? Will everyone in Brazil use Starlink? Scenes from the next chapter.

• 
  Transform Labs (@TransformLabsHQ) reported 20 minutes ago

  3/ Amazon is winding down the service because AI agents finally crossed the capability threshold. Nuanced data validation, exception handling, and contextual decisions that needed human judgment 5 years ago can all be automated now.

• 
  Crisjola—Radio Deer Trash (@tawnyshedu) reported 27 minutes ago

  @debergerac_art @ranksauce This is Amazon leveraging that they “own” HB/HH by airing them, & I assume are threatening to destroy & absorb SpindleHorse. SpindleHorse went years not having an issue & suddenly do when Amazon is having strikes just weeks ago? I’m almost positive Viv is legally gagged. 2/2

• 
  ✨🍩 dönut 0u0 🍩✨ (@trauma_donut) reported 27 minutes ago

  @AmazonHelp hey u can go away, u guys literally made me sit thru 11 different agents that basically ignored my problem and kept transferring me to a different department

• 
  Finance Spot (@financespotnews) reported 31 minutes ago

  $AMZN has now raised approximately $54 billion in bonds so far in 2026, plus $10 billion in a Canadian bond raise in June and $15 billion in a US offering in November. Amazon has said it does not plan to issue more debt in 2026.