Fujitsu 80, Спецификации логического интерфейса SCSI

Страница: 279 / 310

5.3 Disk Drive Error Recovery Processing
C141-C010 5-23
5.3 Disk Drive Error Recovery Processing
This section explains concerning error recovery processing methods and procedures executed by the IDD
for each type of disk drive related error. The INIT can control the error recovery processing methods
executed by the IDD using the MODE SELECT parameters.
5.3.1
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 IDD 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.
If the error cannot be recovered from even when retry processing is executed the specified number of
times, the IDD terminates the command which is currently being executed with a CHECK CONDITION
status. The sense key in the sense data at this time is "HARDWARE ERROR [=4]" and the additional
sense code indicates "Mechanical positioning error [=15-01]" or "Positioning error detected by read of
medium [=15-02]."
(2)
Read error in uncorrectable data
This error is a failure to detect the Sync Byte pattern during reading of data or an ECC error for which
correction processing is impossible. If this error is detected, the IDD, after waiting one revolution of the
disk media, tries reading again. When the error is not recovered from, reading is retried repeatedly the
number of times specified in the MODE SELECT parameter ("Number of Retries during Read" or
"Number of Retries during Write").
In cases where correction is possible during execution of retry processing, the processing procedure is
according to "Correctable Data Read Error" in item (3). If errors cannot be recovered from even after
retry is executed the specified number of times, the IDD terminates the command that is currently being
executed with a CHECK CONDITION status. The sense key in the sense data at this time is "MEDIUM
ERROR [=3]" and the additional sense code indicates "Sync byte not found for data field [=13-00]" or
"Unrecovered read error [=11-01]."
In this error retry processing, rereading accompanied by head offset processing is performed.
(3)
Correctable data read error
This error is a correctable ECC error when data are being read. Processing methods when this error is
detected differ depending on the values specified in the EER flag and DCR flag in the MODE SELECT
parameters.
If immediate correction of errors is permitted (DCR = 0, EER = 1), the IDD immediately makes error
corrections in the data buffer and sends the corrected data to the INIT. Also, in the case of Verify
processing, at the point when it judges that correction is possible, it regards verification as being
successful and continues processing of subsequent data blocks.