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 (46%)
• Errors (29%)
• Sign in (25%)

Live Outage Map

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

City	Problem Type	Report Time
Irving	Errors	49 minutes ago
Lakeville	Website Down	9 hours ago
Zürich	Website Down	17 hours ago
Cali	Errors	1 day ago
Strasbourg	Errors	1 day ago
Lakeville	Website Down	1 day 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:

• 
  Adam Townsend (@adamscrabble) reported 4 minutes ago

  @TrendSpider Def liking Amazon down here, so well positioned going forward

• 
  NoFootsteps (@LahnJohnson) reported 6 minutes ago

  @bavedikian @BestBuy @FedEx Amazon only because it’s their drivers. Don’t order online from Big Box stores, as they use FOB Ship Point which means the customer takes ownership once it’s with the carrier and it’s the customer’s responsibility to deal with any carrier problems.

• 
  Sandeep🐦 (@UmmaSandeep) reported 9 minutes ago

  Reviews are actually more important for movies than products. Why? Amazon lo normal product neeku nachakapothe nuvvu return icheyochu. So review chudakapoina no problem. After watching a bad movie, you will neither get your money nor your time back, so review is more important.

• 
  Quinn Loveborne (@QuinnLoveborne) reported 12 minutes ago

  @NadiaLee @amazon Sadly this is one of those annoying admin errors. I hope they fix this soon.

• 
  Shubham Agrawal 🧢 (@shubhama133) reported 17 minutes ago

  @AmazonHelp Details shared via dm Plz help it's very urgent issue, .don't bother to call at any time even in midnight. I need a resolution asap so that I can get some relief from this mumbai heat..

• 
  Alexander Hill (@Alexand71423435) reported 18 minutes ago

  @KkeefekyKyle They need to fix the problem on Amazon!

• 
  Jorge Ugalde (@studymarketsai) reported 20 minutes ago

  @MerlijnTrader The core point is right: $2T at IPO isn't "early days." Google IPO'd at ~$23B and is worth well over 100x that today. For SpaceX to repeat that from $2T, it would have to become worth more than every stock market on earth combined. Entry valuation caps the math — that part's fair. But two things are loose. Huang didn't "tell retail to buy" — he compared buying at IPO to being an early large investor. A CEO's bullish framing, not a retail directive. And "retail buys the exit" is betrayed by the very examples cited: Meta IPO'd at $38 to a retail stampede and is up more than 10x; Google and Amazon IPO buyers did fine. The variable was never IPO-vs-private. It was the entry price — $23B vs $2T. The real problem is the one the post skipped: the messenger. Nvidia holds ~$30B of OpenAI and ~$10B of Anthropic, and all three run on its chips. Huang endorsing these IPOs is Huang talking his own book — he sells the shovels and owns equity in the mines. That's the conflict to flag, not "retail buys the exit." Skepticism on a $2T entry is fair. Just make the argument that actually holds.

• 
  Glen Bradley (@GlenBradley) reported 23 minutes ago

  @ThatEricAlper I would invest probably every penny of it into developing my local-only Amazon Alexa replacement, and my intrinsic Ethical AI solution to the alignment problem.

• 
  Rob Leff (@RobLeff) reported 23 minutes ago

  @Bhavani_00007 You need to think of AI not as a one size fits all but an ecosystem. It is cheap to do what we did before with Google searches. It will kill advertising on the open web but that is a different topic. The issue is using a sledgehammer for a nail. The cost of tokens is immense so if you use AI when it is not needed the costs will go through the roof. This is no different than going from on premise data centers to the cloud. When that happened our costs doubled at the company I was at since no one managed the usage which is called compute costs. And Google and Amazon made a fortune. Now you have the same with tokens and AI. AI will not explode in usage as stock projections say since it is to expensive for the companies using at as Microsoft learned. But it will be a great tool. All assuming it doesn’t become self aware, create Terminators and kill all of us which is a definite possibility

• 
  JJV (@JJV0d) reported 24 minutes ago

  **** pushing protein powder. **** pushing gadgets. **** pushing serums. The fattest-margin thing an AI persona can move is a $12 digital PDF 100% margin. No warehouse. No tracking numbers. No refunds. No support tickets. No amazon clipping 92% off the top of ur cut. A $12 PDF u build once in google docs + move 300 times a month = $3,600/mo in CLEAN profit off a file that took u an afternoon Now run that across 6 personas pushing different PDFs to different audiences n the digital layer alone prints $20-40k/mo before u even touch affiliate Why is nobody on here saying this out loud Bc "selling a $12 PDF" reads broke next to "$50k/mo in affiliate." The ego wants the screenshot. The account wants the spread. Affiliate on a $42 collagen tub at 7% = $2.94 a sale. U need ~12,245 sales to land $36k/mo. Digital PDF at $12, 100% margin = $12 a sale. U need 3,000 sales to land $36k/mo. No middleman. No shipping lag. No stock risk. No platform leash. She taps download. The sale closes inside the DM. Stripe pays out in 48h. What digital products actually move to this demo The PDFs that print money are NOT ebooks. They're templates, checklists, trackers, n routines she can print at home n use that same day. - "My 5am wind-down stack" (1 page, the supps + timing + the habit chain) = $7 - "Fridge reset checklist" (2 pages, shelf-by-shelf w product links baked in) = $9 - "21-day stiff-knee protocol" (3 pages, supp schedule + mobility list + food swaps) = $12 - "Sunday meal-prep board for women 50+" (4 pages, printable weekly grid) = $9 - "Anxious-pup starter kit" (2 pages, calming routine + product list + vet script) = $11 - "Homeschool year planner" (6 pages, grade-by-grade layout) = $14 Each one is 1-3 hours in google docs or canva. Each one sells 6-18 months untouched. Each one ships through a gumroad or payhip link dropped straight in the DM. How it bolts onto the persona u already run The persona already has the DM funnel live. Keyword in comments. Meta business suite auto-DM. 4-message sequence. Drop one line in: "btw i put together a free checklist of my whole sleep setup. want it?" The "free checklist" is the mouth of the digital funnel. She grabs the free cut (1 page). Bottom of that page: "the full 5am stack w timing + doses + my own notes is $12 here: [gumroad link]" Free-to-paid on this exact flow runs ~14-22% bc she already trusts the persona + the freebie already proved she knows her stuff. The compound across the roster 6 personas x 1 flagship PDF each x $12 avg x ~300 sales/mo = ~$21,600/mo in pure digital revenue Sitting on top of: - Affiliate ~ $30-50k/mo per persona - Subscribe & save ~ $8-18k/mo per persona - Brand deals ~ $5-15k/mo per persona The digital layer tacks $20-40k/mo onto the roster w zero marginal cost after that first afternoon per PDF (btw the PDFs double as list builders. Every $12 buyer hands u their email through gumroad/payhip. That list becomes a channel the persona owns outright, independent of any algorithm. tiktok nukes the page — list survives. instagram flips the algo — list survives. 3,000 emails per persona x 6 = 18,000 owned emails at a 38-44% open rate bc they opened to get the thing they paid for.) Sometimes the richest-margin product in the whole AI persona game is a 2-page google doc someone could throw together over lunch + every operator here is too proud to touch it bc $12 doesn't flex on a screenshot... The screenshot doesn't pay. The margin does. link in bio if u want the digital layer wired into ur current setup or keep handing amazon 92% while a $12 PDF pays u the whole hundred lmaoooo

• 
  Darshan Rajguru (@DarshanRajguru5) reported 32 minutes ago

  @vikpai @AnshulGains That's a brand side problem I thought Amazon/Flipkart affiliates programs are successful tools for creators for money making. Even fintech sources a lot of customers from affiliate creators I get a lot of feed on insta from tier 3 cities with affiliate links all over their pages

• 
  Representative Press (@RepPress) reported 35 minutes ago

  @TechSavvy59787 @CuisineInfamous @ClownWorld The vehicle in the video is a standard Amazon delivery van—not a massive semi or heavy commercial truck. It's one of their common step-in vans (often Rivian electric or similar), with a gross vehicle weight typically around 7,000–9,500 lbs. That's comparable to a loaded pickup truck or large SUV. any residential driveways (especially concrete ones poured to standard 4-inch thickness) are designed to handle 8,000–10,000+ lbs per axle under normal use. A quick partial pull-in by a delivery van for 30–60 seconds is unlikely to cause structural damage in most cases, especially if the driveway is in good condition. AND ONLY THE BACK WHEELS ARE ON THE DRIVEWAY, SO HOW MUCH WEIGHT DO YOU THINK WAS THERE? When you order a package for delivery to your home, you grant the delivery company and driver temporary, limited permission to enter your property (including the driveway and walkway) solely to complete the delivery. This stems from common law principles: delivery personnel are treated as "invitees" (or at least licensees) with a legitimate business purpose tied to your order. They are not trespassers during this reasonable, brief access. This applies to Amazon, UPS, FedEx, USPS, DoorDash, etc. Courts and practices recognize that e-commerce relies on this access. The driver had an implied license for brief, reasonable access. The homeowner can set rules going forward, but the confrontation style doesn't help enforce them and risks escalation. Bottom line: The homeowner was in the wrong here in terms of approach and proportionality. Delivery drivers deal with tight schedules, and a quick driveway pull-in for one package is normal courtesy in most neighborhoods. If damage is a real worry, best options are: Add specific instructions in the delivery app ("Leave at porch, no driveway"). Put up clear (but polite) signs. Contact Amazon support for repeat issues. Most people (including many in the video comments) side with the driver, the homeowner was rude and overreacting.

• 
  Donrado (@donrado) reported 36 minutes ago

  I've changed my mind once again. Instead of selling the book or keeping it private/locked down. I'm going to tack a somewhat condensed version onto the back of the pocket manual. Basically I'm going to combine the pocket manual and full manual into one living document, that I can tweak if necessary as time goes on. This will be great for you since everything is free, and be great for me since I don't have to worry about updating 2 separate documents. The other advantage is I won't have to deal with Amazon rules or policies, it'll be a book strait from me to you. Why give the full manual away for free? Because if the enough people request it, I can make money in other ways like creating pre-mixed powders (Bonobo Nitro Fuel) for people looking for more convenience etc. So to be clear I will be combining the pocket manual and full manual into one unified manual that I can update freely over time without having to deal with Amazon.

• 
  Ayush (@ayuhzkishan) reported 37 minutes ago

  Started with Amazon EC2 basics. It’s AWS’s virtual server service. You can launch, manage and scale virtual machines in the cloud within minutes. Feels like having your own server but without the hardware headache.

• 
  R2 (@R2D2_beepbeep) reported 37 minutes ago

  @AmazonMGMStudio Amazon is defo broken.

• 
  SwipeYield.in (@swipeyield) reported 38 minutes ago

  If you use an ICICI Bank debit card for international trips or online shopping, heads up: a major fee hike is hitting your account on June 21, 2026. ICICI is raising its Dynamic Currency Conversion (DCC) fee from 1% all the way up to 3.5%. That is a massive jump, and it’s specifically designed to target a classic trap a lot of people fall into. Here is what’s happening and how to avoid it. The Trap: "Pay in INR" When you are swiping your card at a hotel abroad, or checking out on a global website (like Netflix, Amazon US, or booking an international airline), the payment terminal or website will often recognize your Indian card and ask: “Would you like to be billed in US Dollars or Indian Rupees (INR)?” It feels safe and convenient to choose INR because you see the exact amount hitting your bank account right away. That is DCC. But it’s a massive ripoff. The merchant's local bank uses a terrible exchange rate with a hidden 4% to 7% markup. And now, on top of that bad rate, ICICI is going to slap a 3.5% fee just for processing that INR payment through a foreign network. It applies even if you are sitting at home in India This doesn't just affect travelers. If you buy software, apps, or subscriptions online from a company that operates in India but is registered legally in a foreign country, you will get hit with this 3.5% fee if you pay in Rupees. *How to protect your money:* The fix is incredibly simple but requires you to be alert: Always choose the local currency: When a foreign POS terminal or website gives you a choice, never choose INR. Always choose the local currency of the country you are in (USD, EUR, AED, etc.). Let your card handle the conversion: When you pay in the local foreign currency, you completely bypass the DCC trap. Your transaction will go through standard cross-border processing, which is significantly cheaper than a bad merchant exchange rate combined with ICICI's new 3.5% penalty. Double-check your checkout screens and recurring international subscriptions before June 21!

• 
  Malky (@Malky0010) reported 42 minutes ago

  @atlanticesque How does Amazon solve a problem of "I want to buy something at 3 AM"? You'll get it delivered next day (at best) anyway.

• 
  Jonathan Condon (@RealVentGuy) reported 43 minutes ago

  @AmazonMGMStudio Amazon is broken. Only way to save it is to bring Stargate back with @martingero

• 
  Abdullah Altamimi (@KsaTmtm) reported 43 minutes ago

  You can’t see this is my account around one years. I cannot do anything my account pending. I think some people will do something inside my account when I tell you help please help. Please help me please you send to me some link the link when I follow up all be wateryou don’t understand my situation problem he cannot fix 30 bucks until one year. What do you want? please do something. Do you need money? What do you need? I will give you but fix my account please @amazon_policy @amazonnews @AmazonHelp

• 
  Gina (@sweettart024) reported 45 minutes ago

  @bavedikian @BestBuy @FedEx Ran into same problem - a computer ordered from Amazon, tracked to a UPS warehouse in TX,en route to me, just "disappeared". Amazon said to deal with Dell & UPS to solve it. I tried with no luck & finally convinced them THEY made the deal with UPS to ship the computer. Not me.

• 
  Killiorr (@killiorrNJ) reported 47 minutes ago

  @dextor_fenix @Demolord25 @ANerdNamedJerry amazon lily fodder being knocked unconscious with everyone freaking ******** out isnt being treated differently at all to you? even in the Motobaru scene they make sure you realize something deeper is happening by Zoro's reaction you are the one with comprehension issues

• 
  Cold Blooded Charter (@ColdBloodedONx) reported 48 minutes ago

  $SPCX (SpaceX) - my strategy It is impossible to avoid SpaceX at this point, it already feels like you cannot even open the fridge without seeing it, but this is exactly where great setups can appear if you have patience. FOMO entries obviously do not exist on this blog, you need to think a bit and play it smart, because the first months will be pure chaos until the first earnings report. The exact date is not known yet, but likely around one quarter after IPO. Ok, let’s start from basics, the most important factor: supply vs demand. At the IPO, around 4.3% of shares hit the market, while the rest is locked until the first earnings release (then they start unlocking gradually - after the report around 20% or even 30% can be released, and if price is above $175 on earnings day, the unlock can be significantly larger, huge difference). On top of that, Elon Musk himself cannot sell for one year. So supply available on day one is extremely low. Demand, on the other hand, is obvious, everyone sees it - massive. Even during the $135 allocation there was roughly twice as much demand as available shares, meaning around $75 billion in capital that could not get filled at that price. Because of that, I mark $135 as a structural support level on the chart. The chart itself comes from MexC, where the stock has already been traded on leverage for about 3 weeks, which gives some early insight into potential key levels. We will see how valid these levels remain once real equity trading starts. $135 is the base, realistically you will not get a second chance to buy it at the same level as BlackRock and other institutions. And definitely not cheaper - I strongly assume that if price ever comes back there, a huge wave of buyers will step in with that same $75B of unmet demand. $175 is the key level around the first earnings report. A price above this level potentially unlocks up to 50% more shares into circulation due to lock-up mechanics. $190 is the first level where price already acted as support multiple times, then broke down with high volume and confirmed it as resistance. Even though this comes from a crypto exchange chart, in a highly hyped asset like this there will be a lot of algorithmic and cross-market participants, not just retail crypto traders, so I treat this level as important. $200 is the obvious psychological level, and also a major POI (point of interest) where most trading volume happened over the last weeks, with multiple S/R retests. Interestingly, if $SPCX breaks above $201, the company would immediately move into 5th place among the largest companies globally, essentially competing with $AMZN for position. Interestingly, $SPCX will compete directly with $AMZN, and this competition will also exist in physical space - low Earth orbit satellite infrastructure. Amazon has only around 300+ satellites while SpaceX already has ~10,000+ and the gap is still widening. Amazon’s Leo project is ambitious, but Musk has the operational advantage. China is also entering the race aggressively. As an investor, it is important to think long term here. SpaceX is effectively three companies in one, and only Starlink is currently generating real meaningful revenue. However, I am not trying to overanalyze fundamentals too much here because in the first months price will be completely detached from fundamentals. The chart will be driven mainly by hype, macro conditions and how long the initial distribution phase lasts. The first major test will be earnings day, but that does not mean there will not be great opportunities before that. Patience is key, otherwise you end up like those who already went long at $215, which you can clearly see on the first candle of the chart. 💙👽

• 
  Representative Press (@RepPress) reported 53 minutes ago

  @1119GJAMG @ClownWorld The vehicle in the video is a standard Amazon delivery van—not a massive semi or heavy commercial truck. It's one of their common step-in vans (often Rivian electric or similar), with a gross vehicle weight typically around 7,000–9,500 lbs. That's comparable to a loaded pickup truck or large SUV. any residential driveways (especially concrete ones poured to standard 4-inch thickness) are designed to handle 8,000–10,000+ lbs per axle under normal use. A quick partial pull-in by a delivery van for 30–60 seconds is unlikely to cause structural damage in most cases, especially if the driveway is in good condition. When you order a package for delivery to your home, you grant the delivery company and driver temporary, limited permission to enter your property (including the driveway and walkway) solely to complete the delivery. This stems from common law principles: delivery personnel are treated as "invitees" (or at least licensees) with a legitimate business purpose tied to your order. They are not trespassers during this reasonable, brief access. This applies to Amazon, UPS, FedEx, USPS, DoorDash, etc. Courts and practices recognize that e-commerce relies on this access. The driver had an implied license for brief, reasonable access. The homeowner can set rules going forward, but the confrontation style doesn't help enforce them and risks escalation. Bottom line: The homeowner was in the wrong here in terms of approach and proportionality. Delivery drivers deal with tight schedules, and a quick driveway pull-in for one package is normal courtesy in most neighborhoods. If damage is a real worry, best options are: Add specific instructions in the delivery app ("Leave at porch, no driveway"). Put up clear (but polite) signs. Contact Amazon support for repeat issues. Most people (including many in the video comments) side with the driver, the homeowner was rude and overreacting.

• 
  Danny (@Danny__Hoffman) reported 55 minutes ago

  @ThomasConnorsJr @amazon Almost every brand new book I've bought from them comes in their paper bags and pages are bent while covers are down or the entire book is bent. Or shipping shampoo with boos and dog food. So much time chatting with them always promising it'll never happen again. It always does

• 
  Zack Riley 🇦🇺 (@ColdHeart_Prj) reported 56 minutes ago

  @NoahRayWrites I would say they probably do if I can buy directly from Germany. This is /com.au vs .com so I am wondering if mid week it pulls them from the US store and the price comes down, but one would think that shipping would be more. It cost $70 to send a package to/from the US to Australia. I guess if amazon does it in bulk it would be different.

• 
  Mason Marauder (@MasonMarauder) reported 57 minutes ago

  @kayellen17 @LiberaNos_AMalo @nosoupforgeorge Pritzker had no problem doing deals with Amazon and Boeing....he cries about billionaire owners to gain support of the rank and file taxpayer. His many Billions does Bezos have?

• 
  omesh (@omeshmangnale) reported 57 minutes ago

  @AmazonHelp The machine is leaking from its inlet continuously. The Urban Company is refusing to service it dismissing the issue altogether. I asked for replacement they said it's bought from Amazon and they are not responsible

• 
  Milk Road AI (@MilkRoadAI) reported 59 minutes ago

  Bill Ackman made one of the most important and counterintuitive observations of the year (Save this). "A similar thing is happening today to Amazon and Meta and Microsoft. These are old-fashioned companies in kind of this OpenAI era." He was identifying an opportunity that almost everyone in the market is currently sleepwalking past. Here is the historical parallel he drew, and it deserves to be taken seriously. In 2000, while everyone was chasing internet stocks, Berkshire Hathaway hit the lowest valuation it had ever traded at in its entire history because investors looked at it, decided it was full of old-economy businesses, and wrote it off as irrelevant. Berkshire's B shares were trading around $29 in March 2000 but today they closed at $488, a gain of roughly 1,520% from that moment of maximum dismissal. Ackman was there in 2000, watched it happen in real time, and now believes he is watching the same psychology play out again. The new new thing right now is chips, semiconductors, and energy, which is where the shorter-term capital is flooding and when capital floods toward one thing, it leaves high-quality businesses behind. Microsoft is down 11% year to date despite its AI revenue hitting a $37 billion annual run rate, up 123% year-over-year. Amazon is down despite AWS growing 28%, its fastest pace in 15 quarters, with advertising crossing $70 billion TTM. Meta is down year to date despite posting 33% revenue growth in Q1 2026 and reaching 3.56 billion daily active people across its apps. These are not struggling businesses but rather are the most profitable enterprises in human history, getting written off as old fashioned at the exact moment they are embedding AI into the core of everything they sell. That is the Berkshire Hathaway dynamic playing out again with the same dismissal, the same rotation logic, and the same opportunity for investors who are paying attention. Come join Milk Road Pro for our full breakdown of which mega cap AI compounders we hold, how we're thinking about the SaaS disruption spectrum from Microsoft to Salesforce, and exactly which companies in our portfolio are positioned to benefit from this capital rotation when it reverses. Link below.

• 
  Wired Weasel (@Wardance68) reported 1 hour ago

  @ClownWorld He just enjoys being an a-hole. UPS, FedEx, and Amazon use my driveway regularly, no problems.

• 
  Sҽαɳ 🇺🇸⚓️ (@doc_1029) reported 1 hour ago

  I hate calling Amazon bc i NEVER get an American on the phone. I don't want to speak to someone with broken English. This is ridiculous. Amazon's customer service is ******* annoying.