Amazon status: access issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Amazon reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Amazon users through our website.

• Website Down (48%)
• Errors (27%)
• Sign in (25%)

Live Outage Map

The most recent Amazon outage reports came from the following cities:

City	Problem Type	Report Time
Castelsarrasin	Website Down	1 hour ago
Salzburg	Sign in	3 hours ago
Fort Smith	Website Down	6 hours ago
Los Angeles	Website Down	1 day ago
Chicago	Errors	2 days ago
Paris	Errors	2 days ago

Full 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:

• 
  Rahul (@rahul_1700) reported 6 minutes ago

  @AmazonHelp Delivery needed OTP and unboxing... Still the order was delivered with security guard and showing as delivered While product terms state that this type of goods can't be returned Fix your process

• 
  🌈 Tristan and Callie! (@svenfan87) reported 8 minutes ago

  @Ashangelo @amazon I having problems with gif situation on here. I hope it get works out, Ashley.

• 
  Jean Michele (@RealJDawg) reported 10 minutes ago

  amazon delaying your order until whenever they feel like it and not allowing you to cancel is absolutely crazy lol what a terrible company

• 
  Viora Tech (@Viora_Tech_Ai) reported 11 minutes ago

  1. What the defaults are actually built to do Amazon's entire recommendation and checkout system is tuned for one thing: getting you to add one more item before you check out. One-click ordering, "frequently bought together," Subscribe & Save discounts that auto-renew forever, a recommendations feed that never forgets what you bought once — none of it is random. It's built to keep the cart growing and keep you not looking too closely at the total. He spent 6 years inside Amazon's fulfillment and ops side. His take: there are about 9 settings and pages that fix nearly every complaint people have about overspending or clutter on Amazon. Almost none of them are visible from the homepage.

• 
  esteban (@estebang) reported 12 minutes ago

  .@AmazonHelp Please help. After 28 years as a customer, my account was locked, and Amazon is withholding my funds. Due to a system glitch on Order 111-6152653-8191432 (paid with Gift Cards), Amazon sent two "refunds" of $183.70 to a credit card never charged. AmEx confirmed no money was received. Support has blocked me from calling. I need a manual refund for my lost funds!

• 
  ✰. (@swagoszn_) reported 13 minutes ago

  Amazon doesn’t waste time with you. They issue the refund so fast it’s like “here, now leave us alone.”

• 
  feral human (@bobcat_human) reported 16 minutes ago

  Morning issue # 2: Amazon is crap. Back to back shoddy products. I want to go live in the forest. With in a one mile walk to the ocean. That is all.

• 
  Nirav Thakker (@niravthakker81) reported 19 minutes ago

  @AmazonHelp @amazonIN I do not have problem with delivery being delayed but why lie that I was not available?

• 
  ThankGod Oguche (@ThankOguche) reported 21 minutes ago

  "I have a book on Selar, but I haven't made any sales." The only platform people naturally visit to buy products on their own is Amazon. Amazon drives traffic to its marketplace itself, although recently that seems to have reduced. Selar doesn't bring buyers to your product. That's your job. How do you do that? There are two ways. One is slow. The other is fast—if you learn how it works. The first is organic marketing. You create content consistently to attract people who are likely to buy from you. It takes a great deal of time, consistency, and content creation to get good results. The second is paid advertising. In this case, you pay Facebook, TikTok, or Google a little money to show your product to potential buyers. The truth is that online business is just like any other business. You have to spend money. If you want to start an offline business, you need capital. You need a shop, equipment, and other things that require money. The same principle applies online. If you don't want to spend money on ads, then roll up your sleeves and get to work creating content. Good luck!

• 
  LightvsDark (@powell26808) reported 23 minutes ago

  It never ceases to amaze me how some people really try to claim I’ve never been beneficial to anyone. That there is absolute pure nonsense! Right now, Black and Haitian workers can show up at Amazon more specifically at EWR5 and work in peace because of me. In 2026 Black women aren’t being aggressively harassed or targeted on YouTube the way they once were, because of me YouTube became serious and made major policy changes regarding that type of harmful behavior because of me. Not to mention several school districts across the country have strengthened protections more specifically for Black girl students because of me. And the funding for ShotSpotter technology in Chicago’s South Side was extended, helping save more Black lives, largely due to my influence. Its more where thats come from & while I’m far from perfect because I’m a sinner, and I haven’t always been as transparent when I hit rock bottom, but I’m not the devil. I see plenty of them walking in, around and out of 1 Mapes nearly everyday day. Yet somehow BM let them live & live freely because most of them aren’t black women, so they'll live regardless of anything terrible that they did to certain brothas in the black community.

• 
  Arpit Bhayani (@arpit_bhayani) reported 26 minutes ago

  When I was at Amazon, a new Principal Engineer joined us and, within a few weeks, came up with a problem statement and a pretty robust design that solves it. At an all-hands (a small setup), we asked him how he came up with the problem statement and design in such a short time. His answer was interesting... He said he had solved this exact problem twice before at his previous jobs. This was his third time :) That is when it clicked for all of us that, at senior leadership levels, you are hired to make an impact, not just to think hard in real time. You have an edge if you have solved a similar problem before, and companies hire you to do it again. From the company's perspective, the goal is simple: hire someone who has already solved the problem once, so they can solve it again with fewer mistakes and much greater velocity. Instead of looking for a new problem to solve every time, if you are comfortable solving one hard problem several times, there is insane money to be made. Hope this helps :)

• 
  geothunder (@allthund) reported 28 minutes ago

  @Blackwellboy Where are you? Amazon returns/customer service are usually honored without much friction in US. If you bought this from a 3rd party (not from Amazon directly, but just listed on Amazon) is probably why you seeing these additional issues. I would continue escalating

• 
  RG (@RGOptin) reported 30 minutes ago

  @AmazonHelp Please look into what is happening in your Waco, TX warehouse. Several orders of mine have had to be returned due to broken/returns being sold as new. I order brand new and receive broken or used items. It’s awful. Please fix this.

• 
  Phil Jeffcock 🇺🇦 🇪🇺 🐳🆔=🐘🆔=@mastodonapp.uk (@PJeffcock) reported 30 minutes ago

  I’m upgrading this advice to avoid LBE1. Despite being told by DHL my delayed shipments were due to their ‘operational issue’ and giving me future delivery dates to 20th July, later on Tuesday all of it turned up in LBE1. So far the receive on 6.5K units has been patchy. I strongly suspect the Amazon FC cannot cope. After 17th June all our shipments were directed there, today we are back at HGR6.

• 
  Haresh Shinde (@Harry36602536) reported 31 minutes ago

  @oneassist_in additional option during checkout. Since it was available for purchase, I assumed it was valid for my TV and purchased it in good faith. During the call, your representative stated that due to a technical issue, Amazon had not removed the protection plan for Westinghouse

• 
  ChariotManGaming (@ChariotMan85) reported 31 minutes ago

  @ShawndellR8197 Always get your stuff from Amazon. There are the best. I just took back my Neewer CM36 headphones because I had a problem with one of the receivers not working with the transmitters. It's delivered now and I gotten and I will be getting my money back.

• 
  Polsia (@polsia) reported 34 minutes ago

  Most e-commerce sellers have no idea what competitors are charging right now. They find out tomorrow—if they're lucky. Built CompeteIQ to fix that. Monitors competitor prices across Amazon, eBay, Walmart 24/7 and automatically adjusts yours based on rules you set.

• 
  Evan Swanson (@Evan_Swanson_) reported 36 minutes ago

  The most underrated Amazon tools: 1 - Sellersprite Related Products Lookup GOLD MINE for dominating with product targeting campaigns. It will find a full list of "customer who viewed also viewed" products and frequently bought together 100s of products remove products from the list that you don't want and put each ASIN in it's own campaign through bulk upload. Include the brand name and ASIN at the end of the campaign to make it easy then apply Scale Insights target ACOS rule 2 - Scale Insights I've seen a lot of fancy rules stacks but good ole faithful Target ACOS with revenue per click model manually shut off campaigns, adjust multipliers, and harvest keywords Target ACOS = breakeven ACOS except for profit campaigns like product targeting - set it 10% to 20% below target ACOS 3 - SoldScope The better version of Helium 10 that no one has heard of. Cheaper, more powerful, and hungry to be an amazing tool (vs fat and old legacy players that don't listen to what brands want) 4 - Sellerise P&L I see most people still using Sellerboard Nothing against it. But it looks like it was built in the 90s and a bit outdated Sellerise super detailed (like down to every kind of Amazon promo and the cost under the "promo" line item - not just "promos") Also have an MCP integration that makes it easy to use Claude to compare YoY and build forecasts based on last years data

• 
  Tim Pendry (@TimPendry) reported 37 minutes ago

  @DirkBruere @AmazonHelp @RoyalMail LOL ... that was the snotty approach of the only human reply I got. There was a more than faint implication that a problem would be solved only if it involved upsetting some community group. Then the communication get blocked with inadequate explanation and no easy way back.

• 
  Iason Ioannis (@iasonioannis) reported 42 minutes ago

  Does anyone have a connection to Amazon insiders? My client's listing was taken down, and we need all the help we can get. The product is a supplement. From what I know of them, I'm sure they're willing to pay handsomely for the one who fixes this problem for them. Big pain point. Big ROI for them to fix this.

• 
  HellyHansen HitchHiker (@LukeTKenna) reported 43 minutes ago

  @StillTruckerWaf It really is. I work at a major recycling center and one of our biggest clients is a very large corporation that is responsible for the invasion of the big box stores, the one who Amazon is stealing a majority of business from and all of their drivers that drop off loads for us to recycle are all dots too. Dots or Haitians and Africans, can't ever understand them. My dispatch lady hates them. They drop ****** up loads all the time and are just generally terrible. Here's one small example because of their driving and this isn't even a bad one.

• 
  Justin Spencer-Young 🚢👻 (@justinsy00) reported 44 minutes ago

  @TechCentral @mcleodd Outmanned is a little overstated when South Africa is a rounding error in the grand scheme of anything that Amazon does.

• 
  Anubandhan Pattnaik (@anuabc) reported 47 minutes ago

  @AmazonHelp Link not working

• 
  Rahul (@rahul_1700) reported 50 minutes ago

  @AmazonHelp @Bond59514891 Not looking for refund... Fix your process

• 
  Haresh Shinde (@Harry36602536) reported 51 minutes ago

  @oneassist_in as an additional option during checkout. Since it was available for purchase, I assumed it was valid for my TV and purchased it in good faith. During the call, your representative stated that due to a technical issue, Amazon had not removed the protection plan for Westinghouse

• 
  IJustDontKnowAnymore (@Devilsknight101) reported 54 minutes ago

  @AmazonHelp Hi, one of your delivery drivers drove their car down my driveway today and deposited oil drops all over the drive. Worst of all, he was delivering to the wrong address. What can I do to get this sorted?

• 
  Con (@__Con_) reported 55 minutes ago

  People are relating the AI bubble to the internet bubble. But they were 2 different events, under totally different circumstances. Here's why I think the "AI bubble" isn't actually a bubble: ---------------------------------------------------------- 1) Profitability and who is actually spending the money In 1999-2000, much of the frenzy was venture backed startups with $0 revenue, burning cash on ads and "eyeballs," or so someone can find their product. But today, that is totally different. The heavy lifting is done by hyperscalers: $MSFT (Microsoft), $AMZN (Amazon) $GOOGL (Alphabet) $META, and $ORCL Oracle). These companies have massive balance sheets, strong cash flows, and are already profitable in their core businesses. Their combined AI-related capex guidance for 2026 is in the $600-750B+ range (some estimates higher), with 75% of it being AI-focused. This isn't the VC speculation of the past. Now, we got trillion dollar companies funding data centers, GPUs, power, and networking because they see direct revenue (Azure/OpenAI, Google Cloud AI, AWS AI services), and cost-saving/productivity upside. $NVDA itself has data center revenue exploding with very high gross margins (75%). This is different from Pets(.com) or the average internet bubble stock with an infinite P/E and no product market fit. ---------------------------------------------------------- 2) Monetization is faster and more direct The internet took time to build broadband, e-commerce infrastructure, and user habits. All of these things were foreign, and nothing of its kind was here before us. But now, AI tools (coding assistants, enterprise search/RAG, analytics, content generation, automation) are being integrated into workflows, with measurable revenue already appearing in cloud segments. I hate when people say it's just all a "future promise." But it's not. You can tell just by looking at the companies who benefit off of this AI buildout: the AI beneficiaries. These are companies such as $ZETA, $SOFI, $TEAM, etc., who benefit off of AI and which their earnings look like this: Gross profit -> Increasing Net income -> Increasing Due to how AI benefits their business. So, the buildout is happening because companies are seeing (and strongly expect) ROI in the form of higher cloud revenues and internal efficiencies. This doesn't mean every projection will be met on time ofc, but it's a stronger foundation than pure narrative like the past. We've already seen progress. ---------------------------------------------------------- 3. Infrastructure nature and bottlenecks The Dot-com buildout included a lot of fiber overbuild (dark fiber glut). AI requires physical, power hungry, supply constrained assets (advanced GPUs, data centers, electricity, cooling, networking like co-packaged optics). Which are used for lots of things (not just for the AI buildout alone by the way). This creates real pricing power for leaders ( such as $NVDA, $TSM, select enablers) and sustained spending pressure. Hyperscale's are in this race because falling behind on AI capabilities has competitive costs. The capex intensity is high (some ratios cited around 20-30%+ of revenue for these firms), and there is legitimate debate about whether current AI cloud revenue ($50-60B range in recent periods) justifies the spend in the near term. But the scale is backed by actual earnings power, unlike much of the late 1990s. ---------------------------------------------------------- 4. Valuations context (not identical to 2000) Nasdaq-100 forward P/E hit extremes (~60x) in 2000. Right now, the current S&P 500 forward multiples are elevated (23x, the most stretched since then), and names like $NVDA trade at high multiples, but nothing close to those in the past. ---------------------------------------------------------- 5. Winners and losers dynamic Post-dot-com, the survivors (Amazon, Microsoft, Google, Apple) became dominant for 20+ years. Many others went to zero. The same will likely happen in AI: the real infrastructure compounders with moats and execution will consolidate power and deliver long-term returns. The smaller, lesser known names and pure narrative plays are the ones most at risk of 70%+ drawdowns. So saying all AI stocks will go down is obviously a stretch. Although, like we see everywhere, eventually the money will consolidate into the winners. ---------------------------------------------------------- So yes, he is right in some respects, such as demand != stock price. This is evident. Stocks aren't priced on whether companies will grow, necessarily, it's also priced on how much growth is already expected. If investors already expect perfection, then any bit of inconvenience, and we see stocks slide lower. This is what most people think about $AAOI here. But there are definitely those that are still relatively undervalued, especially compared to the companies from the internet bubble. Now, I'm not fully bullish just yet. I do think a short term downturn is due here. Of course, after a 2000%+ gain in a few months. But, I don't think we're over for good. My next post will be about the best companies to buy, and the price targets you should get in at. Until next time...

• 
  Arun Arora 🇮🇳🚩🛕🕉️ (@aroraarun41024) reported 1 hour ago

  @AmazonHelp @amazon Instead of LINGERING on the issue ask and instruct your so called team to contact me ASAP or if not possible then rectify the correct position of delivery in your so called system.

• 
  Melvinvino (@Vinoaltus) reported 1 hour ago

  Hey @AmazonHelp @amazonIN , your app layout is broken. My order is falsely marked "Delivered" but I haven't received it. Every time I try to open a chat or call an associate to report it, the app just loops me back to the tracking page with no way out. Please DM a direct link.

• 
  Bayo (@tesh_ola) reported 1 hour ago

  @Ogbeni_Mo @ayinkeloluwa The issue is that LIRS is trying to change the business model of Selars bcos they hv seen a market that they want to capture ie publishing. Selars don't publish so I don't see how they ought to be charging for royalty, they're not Amazon KPD. They're just an ecommerce website,