Google Cloud MINOR
us-central1: Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction are experiencing issu...
February 8, 2023 · 06:23 PM UTC – 07:22 PM UTC · Duration: 59min
Affected Services
Vertex AI Matching EngineVertex AI Online PredictionVertex AI Batch PredictionCloud Machine Learning
Timeline
07:22 PM
The issue with Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction has been resolved for all affected users as of Wednesday, 2023-02-08 11:09 US/Pacific.
We thank you for your patience while we worked on resolving the issue.
07:01 PM
Summary: us-central1: Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction are experiencing issues with cluster deployments
Description: Mitigation work is currently underway by our engineering team.
The mitigation is expected to complete by Wednesday, 2023-02-08 12:00 US/Pacific.
We will provide more information by Wednesday, 2023-02-08 12:10 US/Pacific.
Diagnosis: Customers attempting to deploy Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction clusters will receive an error message: "Machine type temporarily unavailable, please deploy with a different machine type or retry"
Workaround: Customers can use unaffected regions to deploy the clusters where feasible.
06:55 PM
Summary: us-central1: Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction are experiencing issues with cluster deployments
Description: We are experiencing an issue with Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction.
Our engineering team continues to investigate the issue.
We will provide an update by Wednesday, 2023-02-08 11:40 US/Pacific with current details.
We apologize to all who are affected by the disruption.
Diagnosis: Customers attempting to deploy Vertex AI Batch Prediction, Vertex AI Matching Engine, Vertex AI Online Prediction clusters will receive an error message: "Machine type temporarily unavailable, please deploy with a different machine type or retry"
Workaround: Customers can use unaffected regions to deploy the clusters where feasible.