Is Google Cloud down or is it just me? Quick checks
- 1
Read the live status above.
Google Cloud lists every incident with the products and regions affected. Most hit one region or zone, not the whole platform.
- 2
Check your region.
If the incident names us-central1 and you run in europe-west1, it is probably not your problem.
- 3
Check quotas and billing.
Exceeded quotas and a disabled billing account return errors that look exactly like an outage.
- 4
Check permissions.
A removed IAM role, an expired service account key or an organization policy change can block you overnight.
- 5
Use the CLI.
Run gcloud projects list. If it answers while the Console does not, the problem is the Console, not your resources.
Google Cloud error messages and what they mean
| Error | What it means | What to do |
|---|---|---|
| 503 Service Unavailable / UNAVAILABLE | A Google Cloud service could not handle the request | Retry with backoff; check the product and region on Service Health |
| 429 RESOURCE_EXHAUSTED / Quota exceeded | You hit a rate limit or quota | Not an outage: check IAM & Admin, Quotas, or request an increase |
| 403 PERMISSION_DENIED | The account lacks a role or an organization policy blocks it | Not an outage: check IAM roles and org policies |
| ZONE_RESOURCE_POOL_EXHAUSTED | The zone has no capacity for that machine type right now | Try another zone or machine type; this is capacity, not an outage |
| The billing account for the owning project is disabled | Billing was suspended or unlinked | Not an outage: fix billing in the Console |
For cloud engineers during a Google Cloud outage
Confirm the scope first. Match the products and regions on Service Health with what you run before starting failover.
Fail over between regions, not zones, for regional incidents. Multi-zone deployments survive zone failures; only multi-region designs survive a regional one.
Hold deployments. Pause pipelines and autoscaling changes until the incident is resolved, so you are not debugging two things at once.
Watch your own service accounts. IAM and authentication incidents can break workloads that run fine otherwise. Check error logs for PERMISSION_DENIED spikes.
Why does Google Cloud go down?
- Global configuration changes
- In June 2025 a single policy change with blank fields crashed the system that checks every API request, causing 503 errors across Google Cloud worldwide.
- Regional and zonal failures
- Most incidents are limited to one region or zone, caused by network, power or cooling problems in a data centre.
- Shared dependencies
- Products such as Cloud Run, Cloud SQL and BigQuery share underlying systems, so one failure often lists many products at once.
- Capacity limits
- Popular machine types and GPUs can run out in a zone during demand spikes.
Google Cloud outage history
The biggest Google Cloud incidents and what caused them. Durations are approximate and come from public reporting and official reviews.
AUG 20, 2026
Timeouts, errors and elevated latency across many products in the us-west1 (Oregon) region for 3 hours 40 minutes
Google posted a multi-product outage for the region. Source: Google Cloud Service Health
JUL 15, 2026
Google Cloud VMware Engine, NetApp Volumes and Bare Metal Solution were down in zone europe-west4-a (Netherlands)
A cooling failure in the zone. Source: Google Cloud Service Health
JUN 12, 2025
Google Cloud and Google Workspace products returned 503 errors to API requests worldwide for about three hours
A policy change with blank fields crashed Service Control, the system that checks quota and policy for every API request. Source: Google Cloud incident report
How this page works
Every minute our servers request Google Cloud's main endpoints, resolve console.cloud.google.com through Cloudflare, Google and google's own nameservers, and read status.cloud.google.com. We combine that with one-tap user reports, filtered for duplicates and abuse: a spike needs at least 12 reports in 15 minutes from 8 or more different networks.
Built to avoid false alarms: a failed request is retried before it counts, a single bad check never says "down" on its own, and a check that fails for reasons on our side is shown as inconclusive rather than as an outage. Limits: we check from one location, so a regional problem may first show up in user reports. Reports store only the problem type, a country code and anonymous codes that change daily; never your IP address. Protego is independent and not affiliated with Google Cloud. Last reviewed .
Frequently asked questions
Is Google Cloud down today or is it just me?
Compare two signals on this page. Our live check shows whether Google Cloud answers from our servers, and the user-reports chart shows whether other people are reporting problems right now. If our checks fail or reports spike, it is likely Google Cloud. If we reach it but you cannot, check your quotas, permissions, billing and region.
Is GCP down right now?
The status box at the top of this page answers that live. We check the Cloud Console and the Compute Engine, Cloud Storage, IAM and Cloud Run APIs every minute, read Google Cloud Service Health and count user reports.
Why does Service Health show an incident but everything works for me?
Most Google Cloud incidents affect specific products in one region or zone. If you do not use that product in that location, you are not affected.
Is the Cloud Console down or just my project?
If the Console loads for other projects, or gcloud works, the problem is your project: billing, permissions or an organization policy. If the Console itself does not load, check the status above.
Is Gmail down because Google Cloud is down?
Sometimes. Google Workspace shares some systems with Google Cloud, as in June 2025, but most Google Cloud incidents do not affect Gmail. See our Google status page.
Where does Google post Google Cloud outages?
On Google Cloud Service Health at status.cloud.google.com, with the affected products and regions for each incident. Signed-in customers also see project-specific events in Personalized Service Health. We show the public feed live on this page.
FOR CLOUD BEGINNERS
How Google Cloud is built
Our free Google Cloud Digital Leader course explains regions, zones, identity and reliability in plain language.
Start the free courseIS IT DOWN?
Check other services
Is Apple down?
iCloud, Apple Account, App Store and Music
Is Google down?
Search, Gmail, sign-in, Drive and Meet
Is ChatGPT down?
ChatGPT, OpenAI sign-in and the API
Is Facebook down?
Website, app, login, Messenger and DNS
Is AWS down?
Amazon Web Services regions and services
Is Microsoft 365 down?
Outlook, Teams and sign-in
Is Azure down?
Azure portal and cloud services
Is Amazon down?
Amazon.com shopping and checkout
Is LinkedIn down?
Website, app, messaging and feed
Is Reddit down?
Website, app and comments
Is GitHub down?
Git, Actions, API and Copilot
Is Cloudflare down?
CDN, DNS, 1.1.1.1, dashboard and API
Is Akamai down?
CDN, edge network and Control Center
Is Fastly down?
CDN, control panel and API
Is Okta down?
Work sign-in, SSO and Okta Verify
Is Stripe down?
Payments API, Checkout and Dashboard
Is Twilio down?
SMS, verification codes and API
Is Docker Hub down?
Registry, docker pull and Hub
Is GitLab down?
GitLab.com, CI/CD and API
Check any website
DNS, connection, SSL and firewall checks for any address
Related free tools