Google Cloud status: hosting issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Google Cloud 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 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	6 days ago
Township of Evan	Cloud Services	10 days ago
Catania	Cloud Services	19 days ago
Guayaquil	Web Tools	27 days ago
Mexico City	Cloud Services	1 month 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:

• 
  Donatello | AI, Tech & Wealth (@donatelloinvest) reported 5 minutes ago

  @googlecloud The useful part is not the word agent. It is reducing the gap between what a customer means and what catalog filters understand. If it can translate vague intent into better choices without trapping the customer in a chat, that is practical AI.

• 
  Rami (@ramiakeela) reported 6 minutes ago

  Today, @NeraSystems and @googlecloud are announcing partnership. Regulated orgs can now run @GeminiApp on their most sensitive data, client financials, patient records, proprietary datasets, without ever exposing it to the model. The LLM gets the query. It never gets the data.

• 
  DWE Post (@dwepost) reported 7 minutes ago

  @googleaccount Dear Google Team, I can’t access my Gmail account. I forgot my password, and the standard recovery options are not working for me. This is my primary account. Please help me recover it as soon as possible. Thank you. @googlechrome @googlecloud help me.

• 
  AndrewFreidmanEnthusiast (@DeportesEnjoyr) reported 9 minutes ago

  @MLB @googlecloud All the sudden he's hitting homers again. Cheating ****.

• 
  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.

• 
  Bally_AgenticAI (@bally_kehal) reported 12 minutes ago

  Google's Agent Identity gives each agent a SPIFFE ID and a token bound by mTLS to the runtime that issued it. Steal the credential and it won't replay anywhere else. Shared service accounts for agents were always a placeholder. @googlecloud #AgenticAI #IAM

• 
  Tomoaki Sekiya (@SnoopyLinkin) reported 13 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

• 
  Jakob (@jakob_btc) reported 14 minutes ago

  The privacy paradox of the AI era is that utility requires intimacy. To make an AI agent genuinely useful, you have to give it access to proprietary context: internal knowledge, emails, code, workflows. Generic data does not get you very far. So what happens to privacy when hiding the data is no longer realistic? My view is that the privacy stack is shifting from data redaction to verifiable processing. Three major shifts happening right now: 1. Confidential AI at Scale: We’re moving past relying on corporate promises. @googlecloud recently rolled out Confidential G4 VMs with NVIDIA Blackwell GPUs. The tech is mainstreaming. With hardware-based Trusted Execution Environments, your knowledge base is encrypted in memory while the model processes it. Even the cloud provider can’t peek 2. End-to-End Prompt Encryption: Infinite AI memory is a privacy nightmare. The next wave of tools will use localized, cryptographic session gates. Google again recently open-sourced Prompt Encryption SDKs to tackle this problem, establishing a secure channel that keeps data encrypted from the client all the way until it hits the secure chip, before the session vanishes 3. Local Middleware & Edge SLMs: Instead of sending everything blindly to centralized frontier models, enterprises are routing data through local semantic firewalls The bottom line: You will feed AI everything it needs to know, but the tech stack will make sure the vendor never actually "learns" or retains a single byte of it.

• 
  夏嘉莫察瓦绒 (@minyakonga) reported 15 minutes ago

  @googledevs @googlecloud @GoogleDeepMind wtf is that, when gemini 3.5 pro release?

• 
  Howard Beale (@Mad_as_Hel) reported 15 minutes ago

  @googlecloud Good, because they need help !

• 
  CEOInterviews.AI (@CEOinterview) reported 15 minutes ago

  The market spent this month betting AI capex will never earn its return. Alphabet's Wednesday earnings are the first hard rebuttal. Google Cloud operating income hit 8.8 billion dollars, triple the 2.8 billion of a year earlier, on 82 percent revenue growth. Back in April, Google Cloud CEO Thomas Kurian told Matthew Berman why he was not worried: in a capacity constrained market, owning your own chip keeps your unit economics attractive while everyone else pays up. Google is now selling that chip to Anthropic, OpenAI and Meta. Source: $GOOG @googlecloud, Matthew Berman @MatthewBerman

• 
  Gamhe Mitchell (@lilybyt_) reported 21 minutes ago

  I'm encountering some problems getting my billing account set up with @googlecloud Apparently you can only have a single individual billing address per country?

• 
  Antonio Gomes (@onchainantonio) reported 22 minutes ago

  @Googlecloud's recent post-quantum roadmap highlights just how complex this migration is. The most honest line in it is about hardware. Google says some physical replacement cycles will run past 2029. This is a company that owns its data centres, its silicon, its browser and its certificate authority. And it still cannot promise the hardware layer is done by the end of the decade. That is the whole migration in one sentence. Post-quantum is a systems programme, not an algorithm swap. Three things in the roadmap that make the point: → Key exchange was the easy part. Hybrid ML-KEM is already live on Google's endpoints. Encryption in transit moves fast because it only needs two parties to agree. → Signatures are the hard part. ML-DSA and SLH-DSA are too large for the existing WebPKI to carry, so Google is testing Merkle Tree Certificates with Chrome and Cloudflare. They did not swap the signature. They redesigned how trust is distributed. → Roots of trust are the slow part. Caliptra, TPM 2.0, OpenTitan for quantum-secure boot. Silicon has a shipping cycle. No software update shortens it. Now remove Google's advantages. No single owner of the stack. No control over the clients. Keys and signatures published permanently the day they were made. That is what a decentralised network is migrating from. It is why "we'll upgrade later" is a much bigger claim there than it sounds. The organisations with the most control are being the most careful about the timeline. Worth sitting with.

• 
  Alex Boudot (@the_alex_boudot) reported 23 minutes ago

  @NewsFromGoogle @googlecloud @Verizon Verizon users will meet Gemini as the outage that never happened.

• 
  Shailendra Varma (@_shailendra_var) reported 26 minutes ago

  🚨 URGENT REQUEST — Please Expedite My Project Reinstatement My Google Cloud project supporting Tailor Book has been suspended, and the entire app backend is currently down. Because of this, thousands of tailor businesses using the app are unable to manage their shops and daily work. My case has already been escalated to the internal team for project reinstatement, but the process is taking time while our users are completely dependent on the service. 🎫 Case ID: WRUOFYRHZ3ARFV4H6U2YBMRYEA @GoogleCloud @Firebase Please review my case and expedite the reinstatement process. This is a live production application, and every additional day of downtime is directly affecting real businesses. I understand that security reviews take time, but I urgently request your help in restoring the project as soon as possible. 🙏

• 
  Lokesh jha (@Lokesh1jha) reported 26 minutes ago

  @whyrohitwhy @googlecloud Even the gemini also did great work in deployment,I moved my FE from Vercel to Azure container App, Domain was on Cloudflare. It looked on all tabs and solved the issues. Later it got dizzy, may be free plan limits ,but it was really helpful. I needed that to be done overnight.

• 
  Michael Tierney (@Michael_WCD) reported 27 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?

• 
  durrablegoods (@JeffDurrbeck) reported 33 minutes ago

  @googlecloud nothing more annoying than MLB Al and google cloud the worst

• 
  Garrett Smith (@gsmithw91) reported 37 minutes ago

  @googlecloud I have been having a billing issue for 90 days now in which i was charge $3000 improperly. You guys have admitted it but I still haven't received my refund and have had a hard time getting any follow up at all regarding the case. CASE 70049182

• 
  Triman Kumar Saha (@T2872727) reported 44 minutes ago

  @gcloudpartners @googlecloud Urgent!!! I need help my Google Account got disabled by automated AI false positive error... this account is so much important to me and there's no human support.... please help Reference case id- 9-6635000040823

• 
  busch propaganda spreader 🇺🇸 (25-12) (@Nazar4Bedard) reported 54 minutes ago

  @StudiosLA90210 @MLBNetwork @googlecloud Your fanbase sucks and he was right cry about it

• 
  Garrett Smith (@gsmithw91) reported 55 minutes ago

  @googlecloud @ThomasOrTK What is the magic number of follow ups and requests someone has to do to get resolution for billing support? Its beyond 30 at this point.

• 
  The AI Leadership Institute, Nairobi (@aiacademy_nbo) reported 56 minutes ago

  @googlecloud An issue we encountered when trying to setup RAG Engine. "For new projects, using Spanner mode with RAG Engine in us-central1, us-east1, and us-east4 is restricted to only allow listed projects due to capacity limitation. Please switch to Serverless mode in us-central1..."

• 
  Vincent Defarge (@Woody2233) reported 1 hour ago

  @rseroter @googlecloud Identity and registration are a strong foundation. The next challenge is operational context: which service and business outcome can the agent affect, who owns it, and how do we observe or stop it? An agent registry becomes far more useful when it carries that context.

• 
  Mike Sowers (@mikesowerss) reported 1 hour ago

  @rseroter @googlecloud @GoogleAIStudio Cancelled ultra bc it’s the only ai I never even got close to hitting limits (lack of use), it’s entirely fractured across so many different ai sectors that aren’t intuitive at all to navigate.

• 
  Kat MacSpyro (@KatSpyroMacD) reported 1 hour ago

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

• 
  Numero Uno🇳🇬 (@John_dgreatestt) reported 1 hour ago

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

• 
  Sumeet (@sumeetweb) reported 1 hour ago

  @SunilPanda13 @googlecloud For outage? 🤧

• 
  Spinal (@GucciSpinal) reported 1 hour ago

  @HeyMila2008 @helloSQD @googlecloud Bad data can ruin an entire analytics stack

• 
  Parthasarathy Ranganathan (@Partha_Ranga) reported 1 hour ago

  AlphaEvolve now available on GoogleCloud, and as all the customer use-cases below suggest, it is pretty transformative.