7.0
Error recovery
7.1
Write error
If an error occurs during a write operation, it will be detected on the next revolution
by doing a read operation (commonly called a "write check"). To correct the error,
another write and write check operation must be done. If the write operation is not
successful after ten attempts have been made, a read operation should be
attempted on another track. This is done to determine if the media or the drive is
failing. If the error still persists, the disk should be considered defective and
discarded.
7.2
Read error
Most errors that occur will be "soft" errors. Soft errors are usually caused by the
following:
a.
Airborne contaminants passing between the read/write head and the disk. The
contaminants will generally be removed by the cartridge self-cleaning wiper.
b.
Random electrical noise which usually lasts for a few microseconds.
c.
Small defects in the written data and/or track not detected during the write
operation which may cause a soft error during a read.
The following procedure is recommended to recover from errors:
a.
Reread the track ten times or until such time as the data is recovered.
b.
If data is not recovered after using step "a" access the head to the adjacent
track in the same direction it was moved previously. Return to the desired
track.
c.
Repeat step "a".
d.
If data is not recovered, the error is not recoverable.
7.3
Seek error
Seek errors are detected by reading the ID field after the seek is completed. The ID
field contains the track address. If a seek error is detected, the host system should
issue a recalibrate operation (step out until the Track 00 line goes active) and seek
back to the original track.
RS Components
Issued September 1994
9419
Leaflet No. 9419 26/4/97 6:10 pm Page 31
