VMware, Inc.
Methods and systems to troubleshoot and localize storage failures for a multitenant application run in a distributed computing system

Abstract:

Computational methods and systems that troubleshoot and localize failures in data storage of a multitenant application are executed in a multitenant database management system that comprises a control plane of control plane nodes and a data plane of independent and isolated data plane nodes. Each control plane node receives data requests and data queries. Each data plane node maintains data storage for tenants of the multitenant application and executes data storage and data query operations requested by the control plane nodes. Methods and system determine whether data plane nodes are network connected and whether the data cores of network connected data plane nodes are available to ingest data and respond to data queries. Data cores of disconnected data plane nodes and unavailable data cores are isolated from data ingestion and data queries. Remedial measures are executed to restore disconnected data plane nodes and unavailable data cores.