Assigned failed disk to node will automatically rebuild the raid

[SVHO]

Hello,

We had a failed disk on node4.  During that time, node4 didn't have a spare.  Support came to replace the failed disk drive.  We then moved the newly replaced (assigned) drive to node4 using the GUI.  So would the raid be automatically rebuilt after this?

cmcnas::*> storage disk show -container-type spare
Usable Disk Container Container
Disk Size Shelf Bay Type Type Name Owner
---------------- ---------- ----- --- ------- ----------- --------- --------

Info: This cluster has partitioned disks. To get a complete list of spare disk capacity use "storage aggregate show-spare-disks".
1.10.3 8.89TB 10 3 FSAS spare Pool0 cmcnas-03
1.10.24 8.89TB 10 24 FSAS spare Pool0 cmcnas-04
1.10.46 8.89TB 10 46 FSAS spare Pool0 cmcnas-03
1.10.48 8.89TB 10 48 FSAS spare Pool0 cmcnas-03
1.10.50 8.89TB 10 50 FSAS spare Pool0 cmcnas-03
1.10.51 8.89TB 10 51 FSAS spare Pool0 cmcnas-03
1.10.52 8.89TB 10 52 FSAS spare Pool0 cmcnas-03
1.10.54 8.89TB 10 54 FSAS spare Pool0 cmcnas-03
1.10.55 8.89TB 10 55 FSAS spare Pool0 cmcnas-03
1.10.57 8.89TB 10 57 FSAS spare Pool0 cmcnas-03
1.10.58 8.89TB 10 58 FSAS spare Pool0 cmcnas-03
11 entries were displayed.

cmcnas::*> storage disk show -container-type broken
Usable Disk Container Container
Disk Size Shelf Bay Type Type Name Owner
---------------- ---------- ----- --- ------- ----------- --------- --------
1.10.25 8.89TB 10 25 FSAS broken - cmcnas-04

TT

[SpindleNinja · Accepted solution]

Generally yes.  If the node has no spares and an aggr/rg with a failed drive,  it will grab what it can to replace and rebuild the failed drive.    ONTAP will start to rebuild when it sees the new spare drive (Providing it's an acceptable spare) 

Here's a KB on how ONTAP picks disks https://kb.netapp.com/on-prem/ontap/OHW/OHW-KBs/How_does_ONTAP_select_spares_for_aggregate_creation_aggregate_addition_and_failed_disk_replacement

[SVHO]

Doesn't look like its rebuilding.  We moved another spare into the node.  It has 2 spares on the node now.  The GUI still shows its broken.

[SVHO]

We finally talked to support.  They said the cluster is healthy.  The newly drive that was replaced is considered a bad drive ("broken").