Is Google Cloud down?

Problems in the last 24 hours

Google Cloud signals over the past 24 hours. The dashed line is the service-wide baseline used to detect unusual activity.

• Service-wide signals
• Service-wide baseline

Most Reported Problems

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

• Cloud Services (71%)
• Domains (14%)
• Web Tools (14%)

Live Outage Map

The most recent Google Cloud outage reports came from the following cities:

City	Problem Type	Report Time
São Paulo	Cloud Services	12 days ago
Township of Evan	Cloud Services	15 days ago
Catania	Cloud Services	24 days ago
Guayaquil	Web Tools	1 month ago
Mexico City	Cloud Services	2 months ago
Chhindwāra	Domains	2 months 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.

Google Cloud Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Garrett Smith (@gsmithw91) reported 29 seconds ago

  @googlecloud @ThomasOrTK can someone help me get some accountability for a $3118.19 refund that I have been waiting 93 days for ?

• 
  Baki (@jessyjulien_) reported 11 minutes ago

  @googlecloud, I think my code is not the problem. `google-api-core` package might be having a little bit of a rough day.

• 
  Rimple (@RimpleMalik) reported 12 minutes ago

  @googlecloud @GoogleCloud_IN Need urgent help. We have raised a P1 case for quota increase. No one is answering the call. Need to get our Persistent Disk SSD Quota updated. Case number for reference: 73964507. We even bought enhanced support plan for this. Very disappointing.

• 
  Shawn Sully (@aroogle) reported 14 minutes ago

  @googlecloud @GMapsPlatform @Google I need urgent help with an API billing issue case 73262938

• 
  Guy Leech (@guyrleech) reported 14 minutes ago

  The problem with a home lab and @Azure and @googlecloud access is that I sometimes forget where I've done a particular piece of work. All scripts, etc come back to my laptop & @OneDrive but I was using PS2EXE the other day but cannot remember where but I want to re-use the command line as I'm lazy (had parameters for the version resources)

• 
  Shailendra Varma (@_shailendra_var) reported 18 minutes ago

  Here's the message: Hey @googlecloud @GoogleCloudTech, my project "Tailor Book" (ID: tailor-book-25259) was hacked, and support is refusing to waive the unfair billing spike caused by unauthorized bot traffic! 🚨 Please look into Case #74350542 and fix this. #GoogleCloud #GCP

• 
  Usman Bashir (@usman_basheers) reported 18 minutes ago

  @DataChaz @googlecloud The 50% token cut is nice, but the failure mode I keep hitting is agents retrying the whole workflow instead of just the failed step. Does ADK 2.0 fix that granularity, or just the guessing part?

• 
  Ai Trend (@Ai_Trend_) reported 20 minutes ago

  Your agent memorized the training environment, leading to production failures. EnvHarness solves this with a 200-line wrapper and an LLM that designs your curriculum. 🎯 The static trap: Most RL agents overfit to training. New observations or slightly different physics cause catastrophic failure. EnvHarness treats this as an environment bug, not an agent one. 🔧 How it works: EnvRigger, an LLM designer, analyzes your base environment and generates perturbation strategies like noise injection, parameter drift, and scenario mutation. The wrapper applies these dynamically during training. 📈 Results on real benchmarks: We saw a 9.0 point improvement on held-out tasks and nearly 10% faster convergence. These are not synthetic gains; they're on standard agent benchmarks reported in RL papers. 💰 Cost of doing nothing: You're either hand-designing robustness (slow, expensive, incomplete) or accepting brittle agents. EnvHarness is Apache-2.0, plugs into existing stacks, and avoids technical debt and accuracy costs. Stop training agents that fail when it matters. The wrapper ships today. #ReinforcementLearning #AIAgents #GoogleCloud

• 
  Antonio Alonso-Stepanov (@antonioalst) reported 23 minutes ago

  I wonder if there’s a network effect to compute where running more GPUs/TPUs gives you more data on which machines fail and how, which lets you train better models to predict failures and route workloads around them. I’ve heard @googlecloud uses these kinds of predictive health signals.

• 
  BlockClaimed🍚 ⛓ (@BlockClaimed) reported 23 minutes ago

  @onchainantonio @googlecloud Post quantum migration is a trust architecture problem....not merely cryptography

• 
  Carles Sala (@csalacat) reported 27 minutes ago

  @googlecloud Can you please stop renaming things arbitrarily? The names might be cool, but not even @GeminiApp keeps up with your branding changes, and I ask it for help navigating the menu maze it gives me outdated routes that still use Vertex AI

• 
  Adel Bucetta (@adelbucetta) reported 27 minutes ago

  @googlecloud the honest answer is that many companies think they can just bolt ai on top of their existing data mess, but it's not that simple. you need a robust data strategy from day one to support autonomous agents.

• 
  NaoX | Post-Quantum Chain (@NaoXprotocol) reported 29 minutes ago

  @rjmuneeb101 @googlecloud Exactly. Migration is the trap door. Building it in from day one just sidesteps the whole problem.

• 
  Scooter 5150 (@5150Scooter) reported 30 minutes ago

  @googlecloud Anyone else concerned about the "employees" signing a letter to urge US to slow development of AI? Sounds like it would give China advantage with AI.

• 
  abzal gylymbai (@abzalgylymbai) reported 32 minutes ago

  @googlecloud Recurring unexplained charges (~$83, repeating) on my Google Cloud Billing account despite no active projects showing in Console. Already paid multiple times, keeps reappearing. No response from support for days. Can someone please DM me / escalate this?

• 
  Michael Tierney (@Michael_WCD) reported 33 minutes ago

  @googlecloud Whats the protocol for passing context between agents in a network? Do you just pass the entire current state or some kind of delta?

• 
  Numero Uno🇳🇬 (@John_dgreatestt) reported 33 minutes ago

  @Concordium @Google @googlecloud Runtime identity solves a crucial enterprise security problem

• 
  Viraj Kokane (@virajkokane_) reported 39 minutes ago

  .@GoogleCloud @GoogleStartups @ragingdas We've been blocked for over 2 weeks trying to get a GPU quota approved for our production workloads. Every manual quota request is auto-rejected within seconds, with no explanation. Our assigned account manager has been unavailable for over a week, with no replacement. We've spoken to 3 different sales contacts, shared all requested details multiple times, and followed up repeatedly. Yet no meaningful update. Our platform has paying customers, and our render pipeline is effectively down because we can't provision the GPUs we need. We're not asking for special treatment - just a human review, transparency, and a point of contact who can help resolve this. We joined the Google for Startups program to build, not to spend weeks chasing responses. Can someone from the Google Cloud team please look into this?

• 
  Kat MacSpyro (@KatSpyroMacD) reported 40 minutes ago

  @googlecloud Unless you can #BringValkoBack I'm not interested. I genuinely don't think there's anything you can do to help me.

• 
  Mikey Pizza (@PizzaBoiPiazza) reported 41 minutes ago

  @MLBStats @googlecloud @Mets HANG THE BANNER

• 
  200divesL8r (@200divesL8r) reported 42 minutes ago

  @googlecloud Yeah, your product sucks bro!

• 
  Christian Nonis (@christiannonis) reported 44 minutes ago

  I am (unfortunately) dealing with @googlecloud and @awscloud and what I am noticing is that in months they are unable so solve any kind of problem, it’s like being bounced back by ai replies that are sold like assistance from humans, saying that they are “working on that” but nothing changes in months and also replies are all the same.. idk if they are experiencing a shortage in human labor or what’s going on but the experience has become worse than ever

• 
  Tomoaki Sekiya (@SnoopyLinkin) reported 46 minutes ago

  (English follows Japanese) GKEのNetwork Endpoint Group（NEG）事前プロビジョニングがプレビューで登場。NEGのServiceアノテーションにカスタムzonesパラメータを追加すると、対象ゾーンにノードがなくてもService作成時に空のNEGを作成でき、ワークロードのデプロイを待たずにバックエンドサービスへのNEGアタッチなどインフラ構築を自動化できます。 GKE's Network Endpoint Group (NEG) pre-provisioning is now in Preview. Adding a custom zones parameter to the NEG Service annotation creates an empty NEG at Service creation even before nodes exist in that zone, letting you automate infra steps like attaching a NEG to a backend service without waiting for workloads to deploy. #GoogleCloud #GoogleCloudAmbassador

• 
  wiinks (@wiinksdotcom) reported 46 minutes ago

  Been part of @GoogleStartups for almost a year, have not talked to anyone from Google, can't even use the credits becuase @googlecloud compliance flagged the account - just going to remove google oAuth login.

• 
  Mark Thompson (@Bettimoe) reported 46 minutes ago

  $GOOGL reported the most profitable quarter in Alphabet’s history with $112.1B in net income. But investors should not treat that as $112.1B in operating profit. Most of the headline number came from unrealized gains in Alphabet’s equity portfolio, likely driven largely by investments such as $SPCX and Anthropic. The market was right to look beyond that headline. What may be getting overlooked is Google Cloud’s $514B backlog and the fact that demand continues to outpace available capacity. That is why Alphabet is spending so aggressively—and why quarterly free cash flow fell to negative $5.9B. The investment gains created the record profit. The Cloud backlog could create the longer-term opportunity. Those are two very different things. #GOOGL #GoogleCloud #TechStocks

• 
  clive boulton (@iC) reported 46 minutes ago

  @jonerp @googlecloud Salesforce CRM ushered in the biz developer to customize becuz UX for online self-service has far more users. ERP is more complex and seems well suited for AI builders to repeat a similar capability. SAP's ABAP cloud & developer extensibility: on-stack custom logic built for AI?

• 
  Kat MacSpyro (@KatSpyroMacD) reported 50 minutes ago

  @googlecloud Unless you can #BringValkoBack I don't care. Your product can't help me in any way, shape, or form to fulfill my wants and/or needs.

• 
  ST (@ST_Wolfe_) reported 56 minutes ago

  1. The Engine is Built and Dyno-TestedThe actual software engine (the /verify flow on @googlecloud using cryptographic keys and @solana scoring) is fully built, bolted down, and running reliably. It works.

• 
  Henry Lorenzo (@ilhanerdal_) reported 57 minutes ago

  $GOOGL’s Q2 results produced the most profitable quarter in Alphabet’s history: $112.1B in net income—the company’s first quarter above $100B. But that headline needs context. Most of it came from unrealized gains in Alphabet’s equity portfolio, likely driven largely by its stakes in $SPCX and Anthropic. At the same time, $44.9B in CapEx pushed quarterly free cash flow to negative $5.9B. That is what the market focused on. The number I keep coming back to is Google Cloud’s $514B backlog—nearly 5x what it was a year ago. Demand is not the problem. Capacity is. Google still does not have enough compute and data-center capacity to serve that demand, so I believe management is right to keep investing aggressively. The buildout is pressuring free cash flow today. But if it unlocks even a meaningful portion of that backlog, investors may eventually view this spending very differently. The real question is not whether demand exists. It is how quickly $GOOGL can build enough capacity to capture it. #GOOGL #GoogleCloud #AI

• 
  nicholas ⛱ (@nnnnicholas) reported 58 minutes ago

  @_lopopolo @googlecloud @OpenAI woah they poached you? but you were no. 1! maybe you can tell the people responsible for gmail that gemini integration is terrible