IBM Support

A new EXN3000 disk shelf with 3TB drives added to an existing storage system might not be operable.

Flashes (Alerts)


Abstract

Some disks in a newly-added shelf fail to initialize correctly or initialize with incorrect default checksum type.

Content


Problem Description:

After adding a EXN3000 disk shelf populated with 3TB drives to a storage system, one or both of the following symptoms might be reported:


1. Some disks on the shelf fail to initialize with an 'error 5' code
2. All of the disks initialize, but some are initialized with the incorrect checksum type.
Symptom:

Symptom 1:

Example of EMS messages:
disk.init.failure.error:warning]: Disk 1a.24.1 failed initialization due to error 5.

disk.init.failure.error:warning]: Disk 1a.24.2 failed initialization due to error 5.

disk.init.failure.error:warning]: Disk 1a.24.3 failed initialization due to error 5.

Disks will also show up as Startup failed in the sysconfig-v output.

For example:
24.1 : IBM X308_HMARK03TSSM NA04 2855.8GB (5860533168 512B/sect) (Startup failed.)

24.2 : IBM X308_HMARK03TSSM NA04 2855.8GB (5860533168 512B/sect) (Startup failed.)

24.3 : IBM X308_HMARK03TSSM NA04 2855.8GB (5860533168 512B/sect) (Startup failed.)

Symptom 2:

New disks show up as Zoned Checksum spare disks, instead of Block Checksum spare disks.

CLI example showing Zoned Checksum:
node> aggr status –s

RAID Disk Device HA SHELF BAY CHAN Pool Type RPM Used (MB/blks) Phys (MB/blks)

--------- ------ ------------- ---- ---- ---- ----- --------

Spare disks for zoned and advanced_zoned checksum

spare 6d.01.15 6d 1 15 SA:A 0 BSAS 7200 1907479/3906516992 1907729/3907029168

CLI example showing Block Checksum:

Node> aggr status –s

RAID Disk Device HA SHELF BAY CHAN Pool Type RPM Used (MB/blks) Phys (MB/blks)

--------- ------ ------------- ---- ---- ---- ----- ---------

Spare disks for block checksum

spare 6d.01.15 6d 1 15 SA:A 0 BSAS 7200 1695466/3472315904 1695759/3472914816 (not zeroed)



Solution:

Solution for Symptom 1:

There are four options:


1. Upgrade the storage system to Data ONTAP 8.1.3 or later and run disk assign –c block for each drive.
2. Temporarily attach the disk shelf to an available system running Data ONTAP 8.1.3 or later. Once the shelf is recognized, run disk assign –c block for each drive, remove ownership of the disks, halt the system, remove the shelf, and add it to the target storage system.
3. Temporarily netboot the storage system on Data ONTAP 8.1.3 (or later) to maintenance mode and run disk assign –c block for each drive; then halt and reboot back to the resident Data ONTAP version.
4. Have the shelf and/or drives replaced using the 30 Day DOA process.

Solution for Symptom 2:

There is one option:

Run disk assign –c block for each drive.



Additional Information:

Example usage of the disk assign command:
node> disk assign –c block 6d.01.15

Wed Aug 28 14:10:08 GMT [node:raid.disk.inserted:info]: Disk 6d.01.15 Shelf 1 Bay 15 [IBM X306_WMANT02TSSM NA08] S/N [WD-WMAUR0121514] has been inserted into the system

[{"Product":{"code":"ST3VEL","label":"Network Attached Storage (NAS)-\u003EEXN3000 Storage Expansion unit (2857-003)"},"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Component":"Not Applicable","Platform":[{"code":"","label":"Data ONTAP"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
25 September 2022

UID

ssg1S1004431