Google Cloud Outage History

The Chronicle Security WORKSPACE_ACTIVITY data ingestion experienced delays in the Asia Southeast1 and Asia South1 regions, impacting Chronicle Security customers relying on ingestion pipelines. The issue began on Jan 31, 2025 12:53 AM UTC and was resolved by Jan 31, 2025 2:20 PM UTC. The outage lasted about 13 hours and 27 minutes.

Google Cloud issued several status updates during the incident as engineers investigated and worked to restore ingestions in the affected regions. Some user reports indicated continued ingestion delays during the incident, which were resolved as regional services returned to normal.

Google Cloud Translation experienced elevated latency and error rates, which may have caused slower translations or intermittent service unavailability for users. The issue affected the Global Cloud Translation component and mitigation work was underway with no ETA for a fix. The outage began on 12:46 AM UTC and was resolved by 5:00 AM UTC.

Google Cloud Translation and AutoML Translation experienced degraded performance beginning Jan 28, 2025 6:35 PM UTC, with users likely seeing slower translations and intermittent errors. The issue started with AutoML Translation and later showed elevated latency and error rates across Cloud Translation as mitigation steps were implemented. The impact was global, affecting translation requests across these two components.

The AppSheet service within Google Cloud experienced an outage affecting US East 4 (us-east4) and Europe West 4 (europe-west4). Users could not load apps via the editor or app load page, and some pages appeared blank. The incident began at 6:45 PM UTC and lasted about 1 hour and 15 minutes, with restoration by 8:00 PM UTC. A notable volume of user reports accompanied the outage as affected customers reported editor/app-load failures.

Google Cloud issued two status updates during the incident. Mitigation work was underway with no ETA. The affected regions returned to normal by 8:00 PM UTC as services stabilized across the impacted components.

The Chronicle Security feature within Google Cloud experienced elevated timeout errors across US regions, affecting Chronicle Security views, APIs related to rules and detections, and some UI pages. Users in the United States faced degraded performance as Chronicle Security components across multi-regions and US regions were impacted. The outage began on Jan 15, 2025 8:30 PM UTC and was resolved by Jan 15, 2025 11:45 PM UTC, lasting about 3 hours and 15 minutes.

The Google Cloud Status page recorded three updates during the incident as engineers worked to restore normal service: first noting the issue at Jan 15, 2025 10:05 PM UTC; a follow-up at Jan 15, 2025 10:10 PM UTC confirming ongoing investigation; and resolution at Jan 15, 2025 11:45 PM UTC with services restored.

The Vertex Gemini API experienced elevated errors beginning at 7:36 PM UTC (11:36 AM US/Pacific) for Gemini 1.5 Flash 002 model. Customers making Vertex Gemini API calls saw 5XX or 429 responses, resulting in degraded API operations. The impact spanned multiple regions including Europe (regions) / Vertex Gemini API, Asia Pacific (regions) / Vertex Gemini API, europe-west2, europe-west3, asia-northeast1, asia-south1, and australia-southeast1.

Google Cloud issued 2 status updates during the incident. The status was moved to a degraded state on Jan 10, 2025 8:40 PM UTC, with engineering teams continuing to investigate, and the issue was resolved by Jan 13, 2025 2:20 PM UTC as Vertex Gemini API returned to normal operation. The outage lasted roughly 2 days, 18 hours, and 44 minutes.

Google Cloud Chronicle Security experienced data ingestion feed errors and ingestion delays affecting SecOps. The incident began at 12:20 AM UTC and lasted about 8 hours and 30 minutes, with Chronicle Security ingestion degraded across multiple regions. The issue was observed as DNS-related problems and ingestion delays during the incident, according to Google Cloud status updates. The problem was resolved by 8:50 AM UTC.

An outage affected Chronicle Security within Google Cloud, causing degraded access for some users. Specifically, some non-admin users were shown a 403 Forbidden error when logging in to the Cases view, impacting Chronicle Security operations across 20 regional and multi-regional components worldwide.

The incident began at 6:55 PM UTC and was resolved by 7:45 PM UTC, lasting about 50 minutes. Google Cloud issued 2 status updates during the event, describing mitigation work and the eventual restoration of access as Chronicle Security services returned to normal.

Google Cloud Pub/Sub experienced a regional outage that blocked publish operations in multiple regions and prevented subscribing in several regions, disrupting messaging workflows for affected customers. The issue affected Europe, the Americas, and Asia Pacific, with global Pub/Sub status showing degradation. The outage began at 4:00 PM UTC and was largely resolved by 4:15 PM UTC.

Google Cloud's status updates documented the progression: an initial report of publish issues across multiple regions, followed by updates noting subscribe issues, and a partial recovery as regions returned to normal. The provider issued three official status updates during the incident, and there was no workaround available at the time. The disruption represented a wide geographic impact across dozens of Pub/Sub regions, reflecting a substantial incident for messaging workloads.

Some Apigee and Apigee Edge Public Cloud customers were unable to log in to integrated developer portals using SAML. The disruption began at 8:51 PM UTC and affected users across global regions, including the Americas, Europe, and Asia Pacific, with intermittent login failures reported. The incident prompted ongoing investigation by Google Cloud engineers and multiple status updates as the team worked to restore access. The provider issued 5 status updates over the incident course.