CREATING ENVIRONMENTAL SNAPSHOTS OF STORAGE DEVICE FAILURE EVENTS

摘要

A storage device failure in a computer storage system can be analyzed by the storage system by examining relevant information about the storage device and its environment. Information about the storage device is collected in real-time and stored; this is an on-going process such that some information is continuously available. The information can include information relating to the storage device, such as input/output related information, and information relating to a storage shelf where the storage device is located, such as a status of adjacent storage devices on the shelf. All of the relevant information is analyzed to determine a reason for the storage device failure. Optionally, additional information may be collected and analyzed by the storage system to help determine the reason for the storage device failure. The analysis and supporting information can be stored in a log and/or presented to a storage system administrator to view.

主权项

1. A method-comprising:
collecting information about a storage device; storing the collected information; polling shelf log data from a shelf controller of a storage shelf containing the storage device when the storage device fails to time-correlate an input/output (I/O) error with one or more errors of the shelf log data; analyzing the stored collected information and the shelf log data by a computer storage system to determine a reason for the storage device failure; in an event that the computer storage system cannot determine the reason for the storage device failure based on the stored collected information, the method further comprising:
gathering additional information about the storage device failure information; andanalyzing the stored collected information and the gathered additional information by the computer storage system to determine the reason for the storage device failure.