Cloud Storage holds the objects behind data lakes, container images (Artifact Registry sits on it), static sites, and ML training data. Full GCS outages are rare; incidents usually appear as elevated error rates or latency on reads/writes in one region or multi-region - enough to break pipelines and image pulls that assume storage is always there.
Elevated 5xx error rates for object reads/writes in one region or multi-region
Increased time-to-first-byte that times out impatient clients and CI jobs
Image pulls and artifact downloads failing because the underlying bucket path is degraded
Downstream failures: BigQuery loads, Dataflow jobs, and backups erroring because GCS is degraded
Recent Cloud Storage incidents
We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region.
August 20, 2026 · 3h 40m
We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region.
The live status above reflects our latest check. GCS failures are usually partial - elevated error rates rather than a hard outage - so your error budgets and retry dashboards often fire before status pages flip.
What was the big Cloud Storage outage?+
On March 12, 2019, degradation in an internal blob service pushed elevated error rates across Cloud Storage and Gmail/Photos attachments for roughly four hours. Our post-mortem covers what broke and what it teaches about internal dependencies.
Does Cloud Storage downtime earn SLA credits?+
Yes - multi-region standard storage targets 99.95% monthly availability (99.9% regional). Below target, tiered credits apply, but Google requires you to request them within 30 days, and total credits cap at 50% of that month’s bill for the service.
Get alerted when Cloud Storage breaks
Instant email the moment we detect an incident - affected services and regions included.
gcp