Incorrect Enterprise/High Scale latency metrics in the Cloud Filestore

The issue with Cloud Filestore has been resolved for all affected projects as of Thursday, 2022-12-29 04:26 US/Pacific. If customers are still experiencing issues please raise a support case. We thank you for your patience while we worked on resolving the issue.

Summary: Incorrect Enterprise/High Scale latency metrics in the Cloud Filestore Description: We are experiencing an issue with Cloud Filestore beginning at Wednesday, 2022-11-09 for new instances, and existing instances starting Thursday 2022-12-15. This issue will be corrected in the next software release in January. Customers can request fixes for specific instances prior to that by reaching out to Google support. We will provide an update by Thursday, 2022-12-29 10:30 US/Pacific with current details. Diagnosis: The enterprise and high scale Cloud Filestore instances are misreporting average read and write latency metrics. Workaround: While querying latency data, users need to 'multiply the value by 1000' to obtain the correct units.

Summary: Incorrect Enterprise/High Scale latency metrics in the Cloud Filestore Description: We are experiencing an issue with Cloud Filestore beginning at Thursday, 2022-12-22 00:00 US/Pacific. Our engineering has identified a fix and are in the process of creating minimum disruption roll out plan. We will provide an update by Thursday, 2022-12-22 10:30 US/Pacific with current details. Diagnosis: The enterprise and high scale Cloud Filestore instances are misreporting average read and write latency metrics. Workaround: While querying latency data, users need to 'multiply the value by 1000' to obtain the correct units.

Summary: Incorrect Enterprise/High Scale latency metrics in the Cloud Filestore Description: We are experiencing an issue with Cloud Filestore. Our engineering team continues to investigate the issue. We will provide an update by Thursday, 2022-12-22 08:00 US/Pacific with current details. Diagnosis: The enterprise and high scale Cloud Filestore instances are misreporting average read and write latency metrics. Workaround: While querying latency data, users need to 'multiply the value by 1000' to obtain the correct units.