Skip to content

Manually failover a disk in Netezza

1) Check for the disk to be failed over (hwid 1355) and make sure it have a healthy mirror available.

MyServer:/export/home/nz->nzds -id 1355
Data Slice Status SPU Partition Size (GiB) % Used Supporting Disks
———- ——- —- ——— ———- —— —————-
23 Healthy 1421 7 356 79.92 1138,1355
24 Healthy 1421 6 356 79.84 1138,1355

 

2) Identify the SPA in which this disk is located.

MyServer:/export/home/nz->nzhw -type Disk | grep 1355
Disk 1355 spa1.diskEncl4.disk6 Active Ok

 

3) Make sure we have a spare available in the SPA identified in above step.

MyServer:/export/home/nz->nzhw -type Disk | grep Spare
Disk 1582 spa3.diskEncl1.disk12 Spare Ok
Disk 1586 spa4.diskEncl1.disk8 Spare Ok
Disk 1621 spa2.diskEncl3.disk10 Spare Ok
Disk 1622 spa2.diskEncl1.disk10 Spare Ok
Disk 1623 spa1.diskEncl4.disk7 Spare Ok
Disk 1624 spa4.diskEncl4.disk3 Spare Ok
Disk 1625 spa3.diskEncl3.disk7 Spare Ok
Disk 1626 spa1.diskEncl4.disk4 Spare Ok

 

4) Just check the system usage before you manually failover a disk.

MyServer:/export/home/nz->nz_responders

Cur Time Plan # Snippet Time S/P State Busy Dataslices … SQL Username/Database
======== ======= ========= ========= ======= ==== ==================== ======================================== ====================
07:33:45

 

5) Manually issue a failover command

MyServer:/export/home/nz->nzhw failover -id 1355
Are you sure you want to proceed (y|n)? [n] y

 

6) Now check for hardware issue and you will now see one fialed disk and other in assigning state (WHich will disappear from nzhw -issue after regen is complete).

MyServer:/export/home/nz->nzhw -issues
Description HW ID Location Role State
———– —– ——————– ——— —–
Disk 1355 spa1.diskEncl4.disk6 Failed Ok
Disk 1623 spa1.diskEncl4.disk7 Assigning Ok

 

7) Check for the regen status.

MyServer:/export/home/nz->nzds -regenstatus
Data Slice Status SPU Partition Size (GiB) % Used Supporting Disks Start Time % Done
———- ——— —- ——— ———- —— —————- ———- ——-
23 Repairing 1421 7 356 79.92 1138,1623 0 0.00
24 Repairing 1421 6 356 79.84 1138,1623 0 0.00

 

8) Now you will see thestatus is changed from assigning to assigned. It can take few seconds to couple of minutes for this state change depending on system usage, data slice used space.
[
MyServer:/export/home/nz->nzhw -issues
Description HW ID Location Role State
———– —– ——————– ——– —–
Disk 1355 spa1.diskEncl4.disk6 Failed Ok
Disk 1623 spa1.diskEncl4.disk7 Assigned Ok

 

9) You can see the regen status is also progressing now.

MyServer:/export/home/nz->nzds -regenstatus
Data Slice Status SPU Partition Size (GiB) % Used Supporting Disks Start Time % Done
———- ——— —- ——— ———- —— —————- ——————- ——-
23 Repairing 1421 7 356 79.92 1138,1623 2015-06-29 07:42:05 5.29
24 Repairing 1421 6 356 79.85 1138,1623 2015-06-29 07:42:05 0.00