You can run below command to monitor the number of grown defects on disks:-
select DiskID, SPUID, Grown_Defects, DiskModel, DiskSerial from _vt_scsi_defects where grown_defects > 0 order by grown_defects desc
The output of this script will list all the disks on Netezza server where number of grown defects is greater than 0.
Possibility of disk failure increases with increase in disk defects. Below you can find the list that shows the max disk defects at which disk will fail
Disk Type FRU number Disk Failure @ Max defects
ST31000640SS 42D0549 2000+
ST31000424SS 42D0778 4000+
ST31000424SS 49Y1939 4000+
ST1000NM0001 42D0778 4000+
It doesn’t mean that disk will not fail if disk defects is less than value specified. Disks can fail even when number of grown defects is as low as 100 also.
This script will at least help in tracking the grown defects on disk and in general if grown defects increase more than 1500, then it will be a good idea to have it manually failover during non-busy hours and get it replaced.
This script runs good on Netezza TwinFin on version 6.0 and above. Have not tested on any other server/versions.