Seagate Cheetah ST3146854, User Manual

Page: 53 / 108

Cheetah 15K.4 SCSI Product Manual, Rev. D 43
[1] These values are subject to change.
Setting these retry counts to a value below the default setting could result in an increased unrecovered
error rate which may exceed the value given in this product manual. A setting of zero (0) will result in the
drive not performing error recovery.
For example, suppose the Read/Write Recovery page has the RC bit set to 0, read retry count set to 4,
and the recovery time limit field (Mode Sense page 01, bytes 10 and 11) set to FF FF hex (maximum). A
four LBA Read command is allowed to take up to 271.32
msec recovery time for each of the four LBAs in
the command. If the recovery time limit is set to 00 C8 hex (200 msec decimal) a four LBA read command
is allowed to take up to 200 msec for all error recovery within that command. The use of the Recovery
Time Limit field allows finer granularity on control of the time spent in error recovery. The recovery time
limit only starts counting when the drive is executing error recovery and it restarts on each command.
Therefore, each command’s total recovery time is subject to the recovery time limit. Note: A recovery time
limit of 0 will use the drive’s default value of FF FF. Minimum recovery time limit is achieved by setting the
Recovery Time Limit field to 00 01.
7.3 SCSI systems errors
Information on the reporting of operational errors or faults across the interface is given in the SCSI Interface
Product Manual. Message Protocol System is described in the SCSI Interface Product Manual. Several of the
messages are used in the SCSI systems error management system. The Request Sense command returns
information to the host about numerous kinds of errors or faults. The Receive Diagnostic Results reports the
results of diagnostic operations performed by the drive.
Status returned by the drive to the Initiator is described in the SCSI Interface Product Manual. Status reporting
plays a role in the SCSI systems error management and its use in that respect is described in sections where
the various commands are discussed.
7.4 Background Media Scan
Background Media Scan (BGMS) is a self-initiated media scan. It performs sequential reads across the entire
pack of the media while the drive is idle. In RAID arrays, BGMS allows hot spare drives to be scanned for
defects prior to being put into service by the host system. On regular duty drives, if the host system makes use
Table 15: Read and write retry count maximum recovery times [1]
Read retry
count [1] Maximum recovery time per LBA
(cumulative, msec) Write retry
count Maximum recovery time per LBA
(cumulative, msec)
0 51.9 0 23.9
1 59.9 1 35.9
2 203.5 2 55.9
3 235.4 3 67.8
4 271.3 4 119.7
5 283.3 5 (default) 147.6
6 315.2
7 395.0
8 454.9
9 486.8
10 522.7
11 (default) 1,264.8