Gemini Partial Outage - August 25, 2026

<p> Incident began at <strong>2026-08-20 08:40</strong> and ended at <strong>2026-08-20 12:20</strong> <span>(all times are <strong>US/Pacific</strong>).</span></p><div class="cBIRi14aVDP__status-update-text"><h1>Preliminary Incident Report</h1> <p>We sincerely apologize for the disruption this incident caused to your business operations. Recognizing your reliance on Google Cloud, we express our sincere regrets for any operational impact experienced. Our engineering teams are actively addressing the underlying root cause to prevent future recurrences.</p> <p>Please note that the information provided herein reflects our current understanding as of the time of publication and remains subject to revision as the investigation progresses. A comprehensive Incident Report detailing preventive measures will be issued upon conclusion of our inquiry.</p> <p>If you have experienced impact outside of what is listed below, please reach out to Google Cloud Support using <a href="https://cloud.google.com/support">https://cloud.google.com/support</a>.</p> <h2>Date/Time of the Issue (All time US/Pacific)</h2> <ul> <li>Incident Start: 20 August 2026 08:00</li> <li>Incident End: 20 August 2026 10:22</li> <li>Duration: 2 hours, 22 minutes</li> </ul> <h2>Summary</h2> <p>On Thursday, 20 August 2026, multiple Google Cloud services encountered elevated latency, provisioning failures, increased error rates, and service degradations lasting for a duration of 2 hours and 22 minutes.</p> <h2>Preliminary Root Cause</h2> <p>The disruption originated during scheduled fiber optic maintenance, which unexpectedly compromised network capacity between data centers within the us-west1 region. Automated rerouting mechanisms failed to properly redistribute traffic to alternate capacity, resulting in network congestion as volumes exceeded available bandwidth in the affected area.</p> <p>This underlying network degradation subsequently impacted higher-level service components through request throttling, increased latency, and cascading retries. Consequently, both control plane operations and data plane requests failed to execute successfully across several dependent Google Cloud services, producing the observed latency and elevated error rates.</p> <h2>Remediation</h2> <p>Internal monitoring alerted Google engineers to the incident, confirming that multiple services across us-west1 were affected. To mitigate the immediate operational impact, engineers initiated traffic draining protocols, re-routing service workloads away from the degraded network infrastructure. Following the restoration of inter-campus network capacity and the resolution of the core issue, engineers systematically restored traffic to the us-west1 region and confirmed service normalization. Long-term engineering solutions are currently being finalized and assigned to address the root cause and strengthen system resilience regarding fiber maintenance procedures.</p> <h2>Description of Impact</h2> <p>On Thursday, August 20, between 08:00 and 10:22 US/Pacific, Google Cloud customers in the us-west1 region encountered elevated latency, provisioning failures, and increased error rates.</p> <ul> <li>Proportionate Impact / Error Rates: The precise scope of impact—including the proportion of affected projects and applications—and detailed error metrics will be published in the comprehensive Root Cause Analysis, as the reduced capacity caused request throttling, latency spikes, and increased retry volume.</li> <li>Scope Exclusion: Services and workloads operating in regions other than us-west1 remained fully operational and unaffected.</li> </ul> <p><strong>Affected Services and Features</strong></p> <p>The following services experienced elevated latencies and/or increased error rates, across their respective data planes and control planes:</p> <ul> <li>AlloyDB</li> <li>Apache Kafka</li> <li>Apigee Edge Public Cloud</li> <li>Apigee X</li> <li>Artifact Registry</li> <li>BigQuery & BigQuery Data Transfer Service</li> <li>Cloud Build</li> <li>Cloud Data Fusion</li> <li>Cloud Dataflow</li> <li>Cloud Filestore</li> <li>Cloud Key Management Service (KMS)</li> <li>Cloud Monitoring</li> <li>Cloud Run</li> <li>Cloud SQL</li> <li>Contact Center AI Platform</li> <li>Dataproc Metastore</li> <li>Google App Engine</li> <li>Google Cloud Bigtable</li> <li>Google Cloud Pub/Sub</li> <li>Google Cloud Storage (GCS)</li> <li>Google Compute Engine (GCE)</li> <li>Google Kubernetes Engine (GKE)</li> <li>Identity and Access Management (IAM)</li> <li>Managed Airflow (Cloud Composer) *Managed Service for Apache Spark (Dataproc) Persistent Disk</li> </ul> </div><hr><p>Affected products: AlloyDB for PostgreSQL, Apigee Edge Public Cloud, Apigee X, Artifact Registry, BigQuery Data Transfer Service, Cloud Build, Cloud Data Fusion, Cloud Filestore, Cloud Key Management Service, Cloud Monitoring, Cloud Run, Contact Center AI Platform, Dataproc Metastore, Google App Engine, Google BigQuery, Google Cloud Bigtable, Google Cloud Composer, Google Cloud Dataflow, Google Cloud Dataproc, Google Cloud Pub/Sub, Google Cloud SQL, Google Cloud Storage, Google Compute Engine, Google Kubernetes Engine, Identity and Access Management, Managed Service for Apache Kafka, Persistent Disk</p><p>Affected locations: Global, Oregon (us-west1)</p>