Greetings, all
I have been attempting to get a decommissioned Oracle Database Appliance (model X8-2L) reconfigured for use as a temporary landing zone for some unstructured data that we’re moving around in our environment.
I’m using a single host, dual connected to a DE3-24C ZFS storage shelf from a single ORCL-EXT-SAS3 (LSI 3008 7110119) HBA. Host OS is Windows Server 2022 Standard w/ Hyper V.
As expected, the host OS doesn’t see anything as far as the drives inside the array, but StorCLI64 does see the controllers, the array, and the drives themselves.
Here’s what I’ve done so far:
1. Made sure that the firmware for the HBA is in IT mode.
2. Updated the HBA firmware to latest revision from Broadcom
3. Disabled UEFI on the HBA in the system BIOS
4. Updated the Oracle system firmware for the host itself to the latest available version.
5. Installed TrueNAS Core as a Hyper V VM and tested successfully without attempting HBA passthrough.
6. Disabled HBA (Disable-PnpDevice) successfully. HBA’s no longer even appear in Device Manager
7. Dismounted the HBA from the host OS (Dismount-VMHostAssignableDevice) successfully
8. Assigned the HBA to the virtual machine (Add-VMAssignableDevice) successfully
9. Added the additional SCSI controller to the VM hardware with no drives attached.
TrueNAS boots up to a point, and fails with the dreaded “KDB: enter: panic:” error
I really thought I had this problem licked. I’ve done extensive googling on this and it seems like I’ve run through all of the possible gotchas, with no success. Any guidance ya’ll could provide on this would be greatly appreciated