Dropbox 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.

Dropbox Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Aakash Gupta (@aakashgupta) reported 6 minutes ago

  In April 2024, Dropbox disclosed that one compromised service account had given an attacker access to every active Dropbox Sign user's email, phone number, hashed password, API keys, OAuth tokens, and MFA data. Plus the names of everyone who had ever signed a document through the platform without even making an account. Syncthing has been around since 2013 and that breach is structurally impossible against it. 82,000 GitHub stars. MPL-2.0 license. Maintained by a Swedish non-profit foundation. Written in Go. The architecture is the whole product. Every device gets a cryptographic certificate. Traffic is TLS-encrypted end to end. Files move directly between machines you own through the Block Exchange Protocol. No central server gets compromised because no central server exists. Turn off the optional discovery and relay services and Syncthing has zero connection to anyone else's infrastructure. The reason cloud sync keeps producing breaches like the one above is structural. Centralized storage requires a single high-value target. The property that lets you log into Dropbox from a hotel computer is the same property that exposed every user when one service account fell. The convenience and the vulnerability are the same feature. Syncthing trades that property away. The cost is real. Both devices need to be online for sync to happen. There's no web UI you can hit from a borrowed laptop. No shareable link to text a friend. For most people that's a dealbreaker, which is why most people have never heard of Syncthing despite 13 years of open development. For files you actually care about, understand what the $120/year subscription is paying for. Storage at scale is close to free. The price covers an account, a server, a database, and a team that has to keep all three secure forever. The same surface area that made the 2024 breach possible. Dropbox can read your files. So can Google. So can Apple. Their architecture requires it. Syncthing literally cannot. Its architecture forbids it.

• 
  Hany (@kmhaneem) reported 9 minutes ago

  Dropbox launched in 2008 with a simple promise. Put your files in this folder and we will sync them everywhere. Every sync goes through their servers first. Their infrastructure. Their terms. Your files sit on their machines until you need them back. A developer named Jakob Borg decided that was the wrong architecture. Not inconvenient. Wrong, at the level of who owns what. In December 2013 he shipped the first public release of Syncthing. Peer-to-peer file sync. Your devices talk directly to each other. No company in the middle. No server reading the transfer. Syncthing is free, open-source, and has 67,000+ GitHub stars. The project's own stated mission: your data is your data alone, and you deserve to choose where it is stored. Most sync tools list speed and storage first. Syncthing's README lists data protection as priority 1 and priority 2. Speed does not even make the list. That tells you exactly what this project is. -> Your files go from your laptop to your phone. Nowhere else. -> Works on Windows, Mac, Linux, Android, and more. -> No storage cap. Your limit is your own hardware. -> Peer-to-peer sync. Direct device to device, encrypted in transit. -> Runs silently in the background. Zero clicks after setup. -> Web UI included. No command line required to use it daily. -> Open protocol means no vendor can quietly change the rules on you. -> GPG-signed releases. You can verify every binary before running it. -> Versioning built in. Deleted something? You can get it back. -> Self-hostable discovery servers if you want to go fully off-grid. By 2019, Syncthing was getting roughly a million downloads per stable release and syncing hundreds of terabytes of data every day. It is now backed by the Syncthing Foundation, a Swedish non-profit, so no company can buy it, pivot it, or shut it down. Last commit: this week. Shipping continuously since 2013. 300+ contributors. Still pushing updates in 2026. Dropbox: $9.99/month. Google Drive: $9.99/month. Syncthing: $0. Forever. No account to create. No server holding your files hostage. No price hike email arriving on a Tuesday morning. No terms-of-service update quietly granting them new rights to your content. Cancel Dropbox and your access dies with it. Run Syncthing and nothing changes. Your files are on your machines. They stay there whether you open GitHub tomorrow or never again. That is not a feature. That is a different relationship with your own data. 67,000+ stars. MPLv2 license, which means no corporation can quietly close it down. 300+ contributors across a decade. Updated this week. The people who switch to Syncthing are not always the most technical. They are the ones who read the terms of service once and could not unsee them. If that sounds like you, the link is worth a look. (Link in the comments)

• 
  Muneeb Naseem (@MuneebNaseem) reported 11 minutes ago

  The most honest data point on consumer AI economics right now is a YC batch. Of 175 companies in the most recent cohort, only 16 built for consumers. That is a 91% enterprise skew inside the accelerator that historically launched Dropbox, Airbnb, and Reddit, all consumer-first. This is a structural verdict on where the money goes when founders do the math. The unit economics of consumer AI are genuinely broken at the moment. Subscription tiers for a product like ChatGPT compress quickly toward a local revenue maximum because the same users who pay $20/month for Plus would pay $200 for the same output embedded in a workflow they already fund through their employer. Enterprises pay per seat, per token, and per integration without the churn rate that plagues direct-to-consumer apps. Founders at YC read this signal faster than VCs publish it. Brian Chesky himself called out that there is no consumer business model for AI he has seen that scales past a local maximum. The second-order consequence is a talent concentration effect. The 16 consumer-focused companies in that batch will recruit from the same pool as the 159 enterprise ones, at lower expected revenue multiples. That means consumer AI as a category runs lean or runs out of runway before it finds distribution. The parallel to 2012 mobile is instructive. Enterprise dominated early SaaS on mobile too, until one consumer behavior, photo sharing, unlocked a new monetization surface. The category that unlocks consumer AI monetization has not shipped yet. Until it does, every YC batch will look like this one.

• 
  Sanarsh (@sanarsh11) reported 13 minutes ago

  Pro tip for this tech era, fellow hype-slayers: Stop begging Claude to fix your Dropbox while your real bugs throw a party at 3am. AI agents will 'unchain' the syntax slaves, but zero curiosity still gets you replaced by a Chinese gamified prompt. Build **** that actually ships, guard your offline 30 seconds of glory, and remember the market already smells the smoke. We're all just scripting the interview while CEOs whoosh past reality. Stay skeptical, ship anyway.

• 
  JoshEcho (@RealJoshEcho) reported 18 minutes ago

  Right several things this time: 1. Managed to fix the audio commentary. 2. No music or game audio on the vod 😡🤬 3. Now I need to work out how to un submit a streamlabs ticket. 4. Downloaded Dropbox in order to store my 90GB 😱 of stream footage. 5. There's more, a lot more. 1/6

• 
  Inzo Technologies (@InzoTechHQ) reported 22 minutes ago

  Where's your most sensitive data right now? A server? A laptop? Someone's personal Dropbox? An email from 2023? If you don't know where critical data lives, you can't protect it.

• 
  Last Satoshi 🇨🇦 (@iamlastsatoshi) reported 22 minutes ago

  @chamath If the knowledge base or skills are in md file based. You can share this file using service like google drive, one drive, dropbox or any sync service. This way same file but shared between different agents. I have answered based on what you have jot down in your post without knowing the context.

• 
  SHIRISH SHIROOR (@Shirish1004) reported 23 minutes ago

  @vfsglobalcare Filled DS 160. Filled Application for visa appointment and paid fees. We are eligible for interview waiver. After clicking dropbox button it goes to VFS global contact us page and then says 404 error. We want appointment for drop box at VFS global Pune India center. Plz guide us

• 
  CPASteve (@BuffCpa) reported 23 minutes ago

  @AccountingAsArt @cordes_tax We use UltraTax on a remote server. We print PDFs, move them to DropBox (our internal storage). Admin puts them through the Tax Return Deljvery system in TaxDome. Easy.

• 
  Andreas (@abrkn) reported 31 minutes ago

  @Dropbox Dropbox paper logs me out every single day starting a few weeks ago. Login method is passkey. Please fix

• 
  Kirksville Attorney Patrick Nolan (@PatTalksLaw) reported 31 minutes ago

  @kirkjangel I agonized about that for a long time too. Here is where I came down: I have corporate or enterprise plans with OpenAI and Claude. That give the same privacy protections as mycase and dropbox. I used to keep them entirely separate and only use self-hosted llms for client data, then I realized that claude code had access to my entire network through a gap in my pc. So I modified my policy because my use case uses AI. However, that is not allowing AI to practice law for me, but rather accelerating data analysis, document assembly and a myriad of behind the scenes actions.

• 
  Red (@rubelr44) reported 31 minutes ago

  you're paying google $10/month to sit in their server room. dropbox gets $12/month. apple gets $10. the kicker? they can all see your stuff. and when dropbox got breached in 2024? emails, passwords, and tokens were just... out there. there’s this tool called syncthing and it’s honestly kind of a cheat code. no cloud. no company servers. no middleman watching you. it just syncs your files directly between your own devices. peer-to-peer. it's got like 81k stars on github so it’s legit. here is why it wins: direct sync: files go from your phone to your pc. they never touch a 3rd party. privacy: encrypted with tls and crypto certificates. zero friction: no accounts. no sign-ups. just install it and share a device id. everywhere: works on windows, mac, linux, android... even solaris if you're into that. safety net: it has file versioning. if you accidentally delete something, you can just roll it back. the wildest part is that syncthing isn't even a company. it's a swedish non-profit. there is no "cloud" to shut down. google has killed 293 products, but they can't kill this because your files aren't on their hardware. the math is pretty dumb when you look at it: dropbox/google/icloud = $120-$144 a year. syncthing = $0. unlimited storage. unlimited devices. it's been around since 2013 and it's 100% open source. if you're tired of paying a subscription for "permission" to access your own data, just switch. your hardware. your files. forever.

• 
  Anirudh Sharma (@anirudhology) reported 32 minutes ago

  @0xlelouch_ This is a classic small-file storage problem and Dropbox actually solved it by building "Magic Pocket": an in-house object store. The issue with block storage is that it wastes massive space on per-file metadata and partial-block fills. For 10B files, this overhead can cost up to $ 90K/month. Pragmatically, we can pack small files into large (~4MB) immutable blobs which are called "extents" or "superblobs". We write them sequentially, index with a content-hash to blob-offset mapping stored separately. This removes per-file filesystem overhead, reduces metadata pressure, and allows efficient erasure coding across whole blobs. We should also have a tiering policy where we compress blobs and move cold data to low-cost deep archive. Dedup at file level can also be incorporated. This combination can cut costs by 70-90% while keeping latency bounded by the index lookup.

• 
  scooter (@torsten9103) reported 38 minutes ago

  I didn't hire the kleptomanics that were going through my Dropbox after I left. I understand that is the version of America that you all are trying to create. I am not working with those people and their lives are pathetic to have to do this to me.

• 
  Statutorynx (@statutorynx) reported 38 minutes ago

  10 little tips to keep your divorce lawyer fees down. How to manage your divorce on a budget, these habits will save you 1. Use a shared folder Upload documents to Dropbox or Google Drive. Don't print things and drop them off.