Seagate (ST16000NM002G) - Инструкция по эксплуатации - Страница 20

Seagate Exos X16 SAS Product Manual, Rev. K

19 

www.seagate.com

Reliability specifications

Predictive failures

S.M.A.R.T. signals predictive failures when the drive is performing unacceptably for a period of time. The firmware keeps a running count of the

number of times the error rate for each attribute is unacceptable. To accomplish this, a counter is incremented each time the error rate is 
unacceptable and decremented (not to exceed zero) whenever the error rate is acceptable. If the counter continually increments such that it 
reaches the predictive threshold, a predictive failure is signaled. This counter is referred to as the Failure History Counter. There is a separate Failure

History Counter for each attribute.

5.2.4

Thermal monitor

Exos X16 drives implement a temperature warning system which:

1.

Signals the host if the temperature exceeds a value which would threaten the drive.

2.

Saves a S.M.A.R.T. data frame on the drive which exceeds the threatening temperature value.

A temperature sensor monitors the drive temperature and issues a warning over the interface when the temperature exceeds a set threshold. The

temperature is measured at power-up and then at ten-minute intervals after power-up.

The thermal monitor system generates a warning code of 01-0B01 when the temperature exceeds the specified limit in compliance with the SCSI 
standard.

This feature is controlled by the Enable Warning (EWasc) bit, and the reporting mechanism is controlled by the Method of Reporting Informational 
Exceptions field (MRIE) on the Informational Exceptions Control (IEC) mode page (1Ch).

5.2.5

Drive Self Test (DST)

Drive Self Test (DST ) is a technology designed to recognize drive fault conditions that qualify the drive as a failed unit. DST validates the

functionality of the drive at a system level. 

There are two test coverage options implemented in DST:

1.

Extended test

2.

Short test

The most thorough option is the extended test that performs various tests on the drive and scans every logical block address (LBA) of the drive. 
The short test is time-restricted and limited in length—it does not scan the entire media surface, but does some fundamental tests and scans 
portions of the media. 

If DST encounters an error during either of these tests, it reports a fault condition. If the drive fails the test, remove it from service and return it to

Seagate for service.

5.2.5.1

DST failure definition

The drive will present a “diagnostic failed” condition through the self-tests results value of the diagnostic log page if a functional failure is 
encountered during DST. The channel and servo parameters are not modified to test the drive more stringently, and the number of retries are not 
reduced. All retries and recovery processes are enabled during the test. If data is recoverable, no failure condition will be reported regardless of 
the number of retries required to recover the data.

The following conditions are considered DST failure conditions:
• Seek error after retries are exhausted
• Track-follow error after retries are exhausted
• Read error after retries are exhausted
• Write error after retries are exhausted

Recovered errors will not be reported as diagnostic failures.