Why does volume migrated from DS920 to DS3622 have two raid arrays using two paritions while newly created volume has one raid array using one partition.

[BootStrap]

I transferred a 4 disk volume from a DS-920 and added two disks to make up an SHR volume1 on storage pool 1 in a DS-3622 set up. I also created a second SHR array in volume 2 on storage pool 2. As you can see in the images. Volume 1 has two configured md devices (md2 and md3) using the same 6 disks in two different logical partitions (md2: sda5 sdb5 sdc5 sdd5 sdk5 sdm5 and md3: sda6 sdb6 sdc6 sdd6 sdk6 sdm6). Conversely, volume 2 which was freshly created has one md device using one logical partition on each of 6 disks (md5: sdg5 sdn5 sdh5 sdl5 sde5 sdf5).

 

On two occasions, volume 1 / storage pool 1 have degraded and reported a disk as crashed, while the HDD reports and SMART tests show it as healthy. This behavior is then followed by a report of parition failure on the disk.

 

Here are my questions:

 

1. Is it normal to have multiple md devices and partitions for a single volume? Why the difference between the volume 1 and volume 2 arrays?

2. Is the configuration of volume 1, the migrated volume, potentially the source of the inconsistent storage pool degradation while the disk is healthy?

Edited July 31, 2023 by BootStrap

[BootStrap]

Incidently, md4 and md6 are each the 1 SDD cache mounted on each of volumes 1 and 2. I don't really understand what md0 and md1 are doing.