ESXi 6.5 standalone: Raid5 degraded.... or not?

[bellocarico]

I had a disk crash on my DATA raid5 made by 3x 8TB SATA HDD. The controller is an Adapted 6805T and ESXi has been installed with a custom ISO with vendor specific .vib. It worked fine for years...

Due to a supplier issue the disk was replaced with a different brand (old brand Toshiba, new brand is Seagate), still I read up in Internet and was also reassured by the vendor you can mix and match different HDD make, so I have installed it. Initialised the disk using the controller BIOS, added to the RAID5 and the re-sync immediately started. At that point I restarted the box to allow for ESXi to come back online.

Now... I personally don't think this has anything to do with the different HDD brand here but my issue is: after about 24 hours the RAID5 is still reported as degraded:

However... using arcconf I get a different message:

Logical Device number 2
   Logical Device name                      : DATA
   Block Size of member drives              : 512 Bytes
   RAID level                               : 5
   Unique Identifier                        : D63F93AD
   Status of Logical Device                 : Optimal
   Additional details                       : Initialized with Build/Clear
   Size                                     : 15257590 MB
   Parity space                             : 7628800 MB
   Stripe-unit size                         : 256 KB
   Interface Type                           : Serial ATA
   Device Type                              : HDD
   Read-cache setting                       : Enabled
   Read-cache status                        : On
   Write-cache setting                      : Disabled
   Write-cache status                       : Off
   Partitioned                              : Yes
   Protected by Hot-Spare                   : No
   Bootable                                 : No
   Failed stripes                           : No
   Power settings                           : Disabled
   --------------------------------------------------------
   Logical Device segment information
   --------------------------------------------------------
   Segment 0                                : Present (7630885MB, SATA, HDD, Connector:1, Device:0)             WKD068VK
   Segment 1                                : Present (7630885MB, SATA, HDD, Connector:1, Device:1)         27M2K0VBFP9E
   Segment 2                                : Present (7630885MB, SATA, HDD, Connector:1, Device:2)         27M2K0VLFP9E

I suppose rebooting the box would be a thing to try but I'm worried about the re-sync to start over. Who to believe here?

[IRIX201110141 · Accepted solution]

Maybe the state is corrected when resetting IPMI/CIM on the ESXi.

Regards,
Joerg

[bellocarico]

Hum.... for me to understand: why would IPMI play any part here?

This is a driver running on ESXi querying directly the pci-e raid controller, e.g. if I log into IPMI it doesn't provide me any information on cards installed or similar...

I had disks failing in the past, the second Logical disk you see in the first image was degraded last week as I incidentally touched the power cable of one of the two disks. I spotted the mistake, plugged back and yellow turned into green within 6 hours. It seems like this issue is linked to the logical array.

[bellocarico]

I think I got what you mean now.... you were referring to the CIM service!

That fixed, thank you 🙂