Gmail status: access issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Gmail 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 Gmail users through our website.

• Website Down (40%)
• Errors (32%)
• Sign in (28%)

Live Outage Map

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

City	Problem Type	Report Time
Montpellier	Errors	21 hours ago
Paris	Website Down	22 hours ago
Évry	Sign in	2 days ago
Lubbock	Website Down	2 days ago
Cardiff	Errors	2 days ago
Grièges	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.

Gmail Issues Reports

Latest outage, problems and issue reports in social media:

• 
  dusty (@Dtraintolomah) reported 3 minutes ago

  @Robyn3rfw @fib618nyc @alojohhardcore so i get these generic trade alert emails, but they dont pop up from my lock screen because i dont want every single gmail i get notifying me. furthermore, it doesnt say what trade it is, so i have to click the link into the AJ app, which requires me to login every time, unlike

• 
  Tony Tong | Founder | Ancient Systems x AI (@tonytonggg) reported 4 minutes ago

  @Mikadzyki_NFT Handing off a shift to a stack of workers only holds up if the plumbing actually connects. My own Nanobot instance runs on MiniMax 2.5, and getting a Composio MCP server wired in just for Gmail access ate a stubborn chunk of my week.

• 
  Curious Hobit (@Curioushobit) reported 4 minutes ago

  @JohnE_Fabb Off topic but my Gmail was having issues, I sent you something via X

• 
  Amine (@halalmails) reported 4 minutes ago

  Founders think BFCM prep starts in November By then it’s already too late Gmail, Yahoo, and Outlook don’t judge your emails on Black Friday. They judge you on what you’ve been doing in the weeks leading up to it The emails you send this month decide whether your November campaigns land in the inbox or rot in spam If you email subscribers who open and click, better placement, more opens, better reputation. The loop pays you If you blast a full list that mostly ignores you, promotions tab, then spam, then a reputation that keeps sliding. The loop bleeds you By the time Black Friday hits, the loop is already locked. You can’t fix it in a weekend What to actually do this month: → Send only to subscribers who opened or clicked in the last 90 days → If above 40% open rate ibcrease to 120 days if less decrease to 60 days → Exclude Marked as spam, unsubscribed, dropped and suppressed atleast once overall time. → Keep sending consistently 3-4 a week, 80% designed, 20% plain text →Clean your list every 2 weeks Black Friday is won in the weeks most founders spend ignoring email

• 
  JovanHuttonPulitzer™ אני לא סובל טיפשים! (@JovanHPulitzer) reported 10 minutes ago

  @stelzner_n1150 This has to be the stupidest comment ever sorry. 100% of cell phones, smart phones, cable boxes, electric meters, computer usage and email GOES THROUGH DATA centers and always has! Here are the facts: That claim is backwards. The internet and smartphones have always run on data centers (or the same thing under older names: server rooms, colo, hosting farms). What changed is the size, location, and branding — not whether the machines exist. The internet never ran without servers Commercial internet use took off in the early 1990s. From day one it needed: ISP modem banks and backbone routers Web/email/DNS servers Co-location and hosting facilities Those are data centers. Exodus, AboveNet, Equinix, and telco “carrier hotels” were already packing raised-floor rooms with racks in the mid-to-late 1990s. Google’s first big custom plants and Amazon’s early AWS rooms were 2000s expansions of that model, not a new species of building. You could run a static homepage off a PC in a closet. You could not run search, streaming, maps, payments, or “the internet” as people actually used it. Smartphones made the dependence worse, not optional The iPhone shipped in 2007. A phone is a thin client. Almost everything people treat as “the phone” lives off-device: iCloud / Google account sync Maps and traffic App Store / Play Store Gmail, Photos, Drive Instagram, TikTok, YouTube, Spotify Siri / Assistant Banking and payments That traffic hits someone’s data center. If those rooms went dark, the glass slab in your pocket would still make calls on a good day and not much else. Twenty years of smartphones is twenty years of more backend compute, not twenty years without it.

• 
  Gill (@Gilld) reported 14 minutes ago

  @Bakarattae @LaLiga @10ardaguler You can login through Facebook or Gmail

• 
  Aniruddh Dodiya (@aniruddhdodiya) reported 17 minutes ago

  @backlon I hope Gmail would allow to transfer server settings for my other added inboxes.

• 
  ☆~ ASHTHA ~☆ (@DivyaKashy43898) reported 19 minutes ago

  @XSupport My friend is unable to access their X account @cuteshivi77. The password is correct, but X keeps showing “Incorrect Password.” We have requested a password reset several times, but the OTP/reset email is not being received on the registered Gmail account. Please look into this issue and help restore access to the account. Kindly fix the password reset/OTP issue as soon as possible. 🙏 #XSupport #AccountRecovery #PasswordReset

• 
  Martin Szerment | Practical AI (@MartinSzerment) reported 20 minutes ago

  Claude can now run your Mac in the background while you do something else, and that quietly ends the debate about whether AI agents need your full attention. Everyone assumed autonomous computer control meant giving up your screen, watching a cursor move while you wait. That assumption is wrong now, at least on macOS 15. Sonnet 4.6 scores 72.5% on OSWorld, up from under 15% when computer use first shipped in late 2024. Two years took the model from novelty to something close to reliable. Yes, it's Pro and Max only, macOS only, still marked research preview. None of that changes the direction, it only changes the rollout schedule. This isn't a new feature, it's a shift in where automation lives. For a decade we treated automation as an API problem. Now the interface itself is the integration layer. Give it three years and clicking through a UI will be the fallback path, not the exception, for every agent that lacks a direct connector. Connectors like Slack and Gmail stop being the main event and become a shortcut Claude reaches for first, before it falls back to acting like you would. Engineers who spent years building bespoke API integrations are about to watch screen control eat a chunk of that work. Multitasking on a Mac just got a second set of hands, and most people haven't noticed yet.

• 
  franleplant (@franleplant) reported 22 minutes ago

  @poteto @SpaceXAI The gmail connection is not working for personal @gmail accounts

• 
  𝐓𝐡𝐞𝐂𝐫𝐲𝐩𝐭𝐢𝐜𝐖𝐨𝐥𝐟 (@TheCrypticWolf1) reported 22 minutes ago

  @Jaywash_1NFT @Cointelegraph @Yubico Yubikey is a security key that's acts as MFA for accounts such as Gmail, social media accounts, etc, it can't be used with Hardware wallets, unless there is an account login where you sign up with email or phone number and password. Most hardware wallets are not compatible with security keys because they simple don't require it as the hardware wallet acts as the authenticatior in that situation.

• 
  Lukas Caha (@lukas_caha) reported 24 minutes ago

  @lingxi @bot To be more specific, using the Gmail connector bot has huge trouble sending multiple screenshots from app he just made to my own Gmail. I thing he could make an app for doing this 10 times in the time it took him to use Gmail to do it (unsuccessfuly).

• 
  Jake Parente 🎸 (@Jake_Parente) reported 24 minutes ago

  @TraderLion Hey TraderLion congrats on the launch of your new app. I just have an issue logging in as there is no option to login via google gmail. If you could add this feature that would be great. Thanks

• 
  Steven Pu (@reedvoid) reported 26 minutes ago

  @pitdesi Summer of 2004 I built an app that used gmail attachments as cloud file storage, I called it Goggles. I showed it to a few classmates who worked at Google, they told me to shut it down immediately b/c I'll get sued. Keep wondering what've happened if I didn't listen.

• 
  Praetor (@FourVork) reported 26 minutes ago

  It's crazy how everyone is praising Grok Bot, but let me be the annoying person who tells you why it still kinda sucks. 1. The actual capabilities of agents are not novel. Most of the demos I've seen can already be done with Hermes and similar setups, often with smarter models and more control. The difference is that Grok Bot gives you everything out of the box ,without setup and wiring things together. 2. The agent itself is still pretty weak. The UX makes you want to delegate ambitious tasks to it, but the underlying model/agent doesn't always live up to that promise. It takes dumb paths, burns through context/tokens surprisingly fast, and needs more babysitting than the "AI employee" framing suggests. Also, there's no native X search which is especially funny for a product coming from (you have x_search on @grok but not on @bot which needs to use x connection and burn credits). 3. The security model deserves way more attention. All your Bots share the same computer, including files, cookies, browser sessions, and logins. So once you start logging into Gmail, GitHub, Stripe, internal dashboards, etc., a prompt injection or compromised webpage suddenly has a much bigger blast radius. Well, these were the main issues I had with grok bot (apart from smaller issues like absence of read aloud) and why I stick with hermes. But on the other hand, the product is genuinely good, and this is why.

• 
  Andrew Linder (@andrewmlinder46) reported 28 minutes ago

  @TeamYouTube Yes that link does not work I tried that a few times but I cannot login and go through the process of accessing my Gmail. So I will need some additional help please.

• 
  The Dr. Margaret Show (@DrMargaretShow) reported 29 minutes ago

  @JohnBeaudoinSr Wow. Gmail shares attachments with AI, and yesterday, GoDaddy’s Outlook was down for 8 hours. I was unable to log in on my laptop, and the mobile app showed no new emails all day. Looks to me like they don’t want us emailing one another, in preparation for the next thing.

• 
  J Bret Edwards (@jbecompslns) reported 30 minutes ago

  @specialkdelslay I'm not quite sure what GCP defines as a vCPU, but I DO know how a vCPU in EC2 is defined and I would almost certainly bet they're either identically defined or very similar. It's probable that a "vCPU" is a hyperthread (or in non-intel parlance, an SMT thread, or just a thread). I would highly suspect D.O. oversubscribes their customers (probably not to 2 or 3x physical threads), but probably 1.5x. This is an educated guess, but without seeing how they define it, and knowing EXACTLY how EC2 does it, the way you get "faster" vCPUs is a very small number of ways: 1. dedicate a single hyperthread per customer and only over-subscribe the micro-series instance types (t-series in EC2). These dedicated hyperthreads are on the same physical core on a "medium" instance type, so you get one dedicated pCPU with 2 vCPUs. 2. newer hardware with high clock speeds. even though D.O. may be using the same generation CPU, they are likely not getting custom SKUs, like google, amazon, and microsoft all get. seriously, go look up the sku for any CPU in any of AWSs EC2 servers (you can see it with a /proc/cpuinfo)...then go look up the SKU...it doesn't exist on Intel's website. Custom higher-clockrate SKUs SPECIFICALLY for cloud vendors that they do EXTENSIVE testing on and get the premium binned silicon wafers (they buy and reserve their own wafers). Technically, newer hardware and higher clock speeds/custom skus is two separate ones, but they're more or less the same. The point is, there's basically two ways to boost vCPU performance. One is improving hardware, the other is the way you load down the hypervisor (also, it depends on whether or not they're isolating the hypervisor resources or not or if they're shared by a customer). Also, the perceived performance gap could be that they have bad luck on where they've landed and they're getting either a noisy neighbor and/or they're sharing infra with the VMM, which needs dedicated resources in Xen. In KVM, it doesn't need dedicated resources, but will still consume some resources and may cause performance hits depending on where the VMM task(s) run(s). As for what google is doing, yeah, looks like they're juicing their cloud customer numbers. I haven't used GCP in over 10 years, but I'm certainly still counted as a user, because it was tied to my personal gmail when I started my account. Also, youtube creators have to go into the google cloud console to set certain settings for API search and request limits. I suspect they're counting those users as well.

• 
  Robin Swanson (@robin_swanson) reported 31 minutes ago

  @MSFT365Status I haven't gotten emails since 9:11 AM. What are you doing to fix this issue? How are you communicating to customers? This is beyond unacceptable. Yet again a major malfunction by Microsoft. I should just switch my business email to Gmail, which has never gone down.

• 
  AncienCarnelian (@AncienCarnelian) reported 34 minutes ago

  @krainboltgreene @Speedkicks @slime_machine Unfortunately, it's basically impossible to create your own email server. The software and hardware is easy enough to get setup, but the major email service providers (Microsoft Outlook, Google Gmail, contracted emails through those 2) block all but a shortlist of providers.

• 
  Kaz (@ThisKaz) reported 34 minutes ago

  @GergelyOrosz Really? When tech companies have an outage they sort of shrug and the economy goes on. AWS out - “we’re sorry” Robinhood down - “we made mistakes” Microsoft/Gmail down - “we’re working on it” Bank can’t dispense money? Bank runs, economic fall out, congressional hearings.

• 
  Pers (@0xPerser) reported 36 minutes ago

  Gave Grok Bot a real assignment: run parts of my business like an actual employee, not a chatbot answering questions. First mistake, immediately corrected. I almost had it spin up its own Gmail account — Google's already catching and disabling Gmail accounts created by agents like this. Switched to AgentMail instead, since it's already a listed plugin in Grok Bot's browser. OAuth, three minutes, 24 email tools live and working. For payments, Stripe has a live MCP server — the agent can create customers, issue invoices, run transactions, all through plain language, no custom integration built by hand. Before any of it touched something real, I wrote the rules first. What it's allowed to read. What it's allowed to send. What it's allowed to spend, and the exact caps. What needs my approval before it happens. One command that shuts the whole thing down if I need it gone. Passport first. Rules second. Never the other order. The mistake almost everyone makes isn't giving an agent too much access. It's giving it access before the rules exist to contain it.

• 
  Mathieu (@mathlleu) reported 39 minutes ago

  LinkedIn bought the best Gmail extension ever made. Then killed it. Twice. 2012: LinkedIn buys Rapportive for a reported $15M. The sidebar showed you who any email sender really was. Salespeople used it for something else entirely: type a guessed email into Gmail, and if Rapportive lit up, the address was real. 2017: rebranded into "Sales Navigator Lite for Gmail". 2020: shut down. Official reason: low adoption. Then it happened AGAIN. The most recommended replacement, Clearbit Connect, died in April 2025. (HubSpot bought Clearbit, and that was that.) Rapportive died twice. The job never did. Find a work email, prove it's real before you hit send. People did that by hand, address by address, in the To: field. Today it's verification against live mail servers, and if the lookup misses, you don't pay. We wrote up which famous "Rapportive replacements" are actually still alive in 2026. The graveyard is longer than you think. Link in the replies 👇

• 
  Julian (@julianfaceless) reported 39 minutes ago

  I once posted the SAME video on 2 freshly created 0-sub channels. Channel 1: 87 impressions Channel 2: 52,330 impressions Only variable: the infrastructure behind it. Here’s what that tells us about YouTube’s algorithm: Your channel carries a score that no one will EVER show you. Think of it like credit at a bank, nobody cares if you pull up in a nice suit and a Richard Mille. They pull your history and decide how much they're willing to lend you. YouTube does the same. A trusted account gets its video in front of a REAL test audience, and if it performs YouTube scales distribution hard. If the account is not trusted, it won’t get shown to anyone, meaning nobody clicks. And when nobody clicks, the algorithm just confirms what it already suspected and moves on. Now here’s what actually feeds that score: → IP reputation and history → Fingerprint consistency across sessions → How the account behaved in week one → Upload velocity → Retention and CTR over time → Whether your engagement looks bought → How long you've stayed in one niche It’s reasonable that people obsess over the bottom half… those are the numbers YouTube Studio shows them. The top half are as important though, yet they’re invisible to the untrained eye So here’s how to spot if your infrastructure sucks: Open your channel's setup and look for anything shared. → Same IP as another channel? → Same browser profile? → Same recovery email? → Same phone number? → Same AdSense? Any of those and the channel isn't isolated. People overcomplicate this but it's simple. YouTube is trying to figure out one thing: "Is this a real separate person, or 1 guy spinning a bunch of accounts?" If your setup looks like one guy wearing 10 masks, you're cooked. A proper setup looks like this: Channel 1 → Profile 1 → IP 1 Channel 2 → Profile 2 → IP 2 Channel 3 → Profile 3 → IP 3 Each channel on its own dedicated residential IP. Not a proxy or a cheap VPN you can find anywhere. That's the whole reason I built @tubeproxies. Every IP is sourced through residential Internet Service Provider networks for one use case only: Faceless YouTube channels. Because nothing is as demanding as YouTube when it comes to IP quality. Then each channel must get its own: → phone number → recovery email → browser fingerprint → login history → normal usage pattern Next you must act normal across all of them: Don't log into Channel 1 from Profile 2 because you were lazy. Don't open your personal Gmail in the same profile. Don't check an old banned channel from the clean browser. Don't use the same SMS number across the portfolio and call it separated. This is how people poison clean channels without realizing it. They buy a good IP, then attach it to a browser that already has ***** history. They make a new Gmail, then add the same recovery email they used on 6 other channels. They get a new phone number, then log in from the same Chrome they used for a terminated account. Then they: → Post for 30 days → Get 0 impressions → Blame content/algo Could be the content. Could even be the algo. But if the infra is messy, YouTube never gave the channel a real test anyway. And if any part of your channel touches another another one such as an old account, a banned account, your personal account, or a reused setup… You must fix it before you upload again. Don't wait until the channel is doing 300k views a day to think about this. That's the worst time to start swapping IPs, emails, phones, browsers and AdSense. Sudden changes look like inauthentic account behavior. Build it clean from day one, then act normal. Browse before uploading. Warm the account up. Don't upload 3 videos in 24h on a fresh channel. Don't log in from Denmark at 9am and Texas at 9:20. Don't make the channel look like a bot farm and then wonder why YouTube treats it like one. The channel with 87 impressions was from a Nigerian IP. The one with 52.3k was a US one.

• 
  SYNTHLEX (@SYNTHLEX_) reported 40 minutes ago

  xAI just handed Grok Bots the right to read, write, and act inside your Microsoft accounts. The timeline is fighting about whether that is safe. The real story is a single permission scope that almost nobody clicked through to read. Grok's Outlook connector asks for two things at sign-in: Mail.ReadWrite and Mail.Send. Full send authority from the first Accept. Grok's Gmail connector, by comparison, starts read-only and only earns send access when you turn it on. Read that again. The Microsoft account got the more permissive default. That is not a footnote. That is the exact line where an assistant becomes an employee. Now look at the screenshot everyone scrolled past. A left sidebar of named workers. Schedule Manager. Inbox digest. Bug triage. Expense digest. Launch post. That is not a chat window. That is a staff directory. First. The unit of AI work stopped being the prompt. You are not typing questions into a box, you are assigning roles to a roster. The interface is an org chart wearing the skin of a messaging app. Second. The permission is the product. For two years the bottleneck was never capability, it was trust. A model that drafts a flawless email but cannot send it is a brilliant intern who still needs you to press the button. Mail.Send deletes the button. The value was never in the draft. It was always in the authority to act without you standing there. Third. Each bot runs on its own cloud computer that keeps working after you log off. Persistence changes the verb. A tool waits for you. An employee works while you sleep. The 9:41 AM digest in that screenshot is a report on labor done in the hours the user was gone. Fourth. OneDrive and SharePoint access is gated to Business and Enterprise tiers, and setup routes through an Azure admin consent. xAI is not selling you a chatbot. It is selling your IT department a workforce. The Copilot war will not be won on model quality. It will be won at the admin approval screen. Fifth. The skill this rewards is not prompting. It is delegation. Deciding which bot gets Mail.Send and which stays read-only. Knowing what "make it 45 min instead" means when a human types it to a manager instead of a machine. The person in that screenshot is not a user. He is running a standup. The model can already do the work. The only decision left is what you sign it the authority to do without asking you first. That was never a prompting problem. It is a management one.

• 
  gkemp (@gkemp) reported 47 minutes ago

  @CreeK_ @MSFT365Status Gmail had a self inflicted outage on Apr 8, and is garbage for business (I've used it at several companies).

• 
  Gabriel Mican (@youwishmusic) reported 51 minutes ago

  @cartierfamilyZ Why are people so pressed about data centers all of a sudden? All of your online information from X account, to gmail, any online storage, website etc is stored in a data center. We've had data centers for decades, all of a sudden they're an issue? Take a deep breath, and tune out the idiotic noise.

• 
  Matt Wright (@MattWrightMCT) reported 54 minutes ago

  Every business that reaches seven or nine figures has an aha moment. Mine came two hours into day one. An aha moment is when you find the one thing that gives your business an edge. For Facebook, it was working out that a new user who made seven friends in their first ten days was unlikely to ever leave the app. So that became the only number the company chased. For Gmail, it was working out that somebody typing their password was almost certainly about to open their inbox, so they started loading it right then and killed the wait everybody else was making you sit through. At Cars4Us, it came two hours after we opened the doors, when I bought the first car and two things were obvious. Buying cars was far easier than I'd assumed it would be. And what made that bloke say yes had far more to do with how fast we moved than with the price I gave him. Everything else about us was worse. We were four people using a cheap transport company, and nobody had ever heard our name. He sold to us anyway, because we got to him first. That was the edge. Speed. I couldn't be bigger than my competitors on day one and I couldn't outspend them. But I could be the quickest, and that was the one area nobody was competing in. So the whole company got built on that. Enquire at 9:30 in the morning, inspector standing at your car by 1pm, money in your account by 4. Within 60 days we were the largest online car buying company in the country. So stop trying to be better at everything, because that's the one game every competitor in your market is already playing. Go and find what your industry is slow at, and own that instead. (And yes, I used the F1 car pick because I'm talking about speed here)

• 
  Loscauses2 (@handsomejason09) reported 54 minutes ago

  [REPOST] @ScopelyExplore FIX THE PROBLEM WHERE ALL THESE HIGH LEVELS ARE GETTING HACKED! I’m Level 79 and I don’t wanna be a target. Remove the option where you can change your Gmail or account info in game.

• 
  a (@adidogCEO) reported 59 minutes ago

  and as we continue to see retail adoption events, the less crypto becomes crypto ppl don't want to know what is under the hood, they just want it to work they don't want to remember seeds, private keys, or even use a wallet address to signin they just want to login with gmail and have it work and they are not fee-sensitive as long as its not exorbitant cheap and fast doesn't matter as long as it is cheap and fast enough