Fujitsu MBA3147RC Manuel D’Utilisation

Page de 439
Sense Data and Error Recovery Methods 
 
382 
C141-C013 
6.2.3  Error logging 
In order to collect information that is effective in maintenance, it is desirable for 
the INIT to accumulate (log) error information related to the SAS interface which 
it has detected itself (error frame detection, command completion wait time-out, 
etc.) and error information reported by the HDD. 
By specifying "1" in the PER bit in the "Read/Write Error Recovery parameter 
(Page 1) and the "Verify Error Recovery parameter (Page 7) of the MODE 
SELECT or MODE SELECT EXTENDED command, the INIT can know of the 
occurrence of errors which were recovered from successfully by the HDD's 
internal error recovery processing and their contents.  By logging that error 
information, the INIT can obtain data for more detailed analysis of the disk drive's 
operating condition. 
Concerning the contents of the error log, it is recommended that the data have a 
time stamp affixed so that they can be edited in a time series of the times the 
errors were detected, and that they include the following information. 
• 
The source which issued the command and the ID of the device which was 
the target of the command. 
• 
The CDB which was issued. 
• 
The receiving status, or, when the status was that the command could not be 
received, the code showing the type of error detected by the INIT and its 
content. 
• 
All bytes reported by the HDD. 
6.3   Disk Drive Error Recovery Processing 
This section explains concerning error recovery processing methods and 
procedures executed by the HDD for each type of disk drive related error.  The 
INIT can control the error recovery processing methods executed by the HDD 
using the MODE SELECT parameters. 
6.3.1  Error states and retry processing procedures 
(1)  Seek error 
Mechanical malfunctions of the disk drive during a seek operation and failure of 
serial addresses detected during verification of the ID area to match are included 
in this error. 
When a seek error occurs, the HDD first executes rezero seek, then tries 
repositioning at the target cylinder and restarts the original processing.  If the 
error is not recovered from, retry processing (rezero seek and repositioning) is 
performed up to the number of times specified in "Number of Retries during Seek 
Error" in the MODE SELECT parameters.