Drive path keep changing after reboot

That's not the drive path. You're talking about the device name. (/dev/sda is a device name.)

Drive paths, as they are used in OMV, can be displayed under:

Storage, File Systems.
Click on the table Icon on the right and check the box for Mount Point.

You'll see drives mounted under their UUID's.

__________________________________________________________________

Why are your device names changing? If I were to guess, one of your many drives may have a bad cable or power connection. Maybe, if you have one of the cheaper HBA adapters, to attach additional drives beyond what the MOBO can handle, the card is not consistently reporting port numbers versus drive assignments to BIOS. (In that scenario, again if I were to guess, the drive(s) that are changing device names might be attached to that card.)

In any case, your scenario is one of the reasons why OMV mounts drives using UUID's in the mount point. UUID's don't change.

Quote from crashtest

That's not the drive path. You're talking about the device name. (/dev/sda is a device name.)

Drive absolute paths, as they are used in OMV, can be displayed under:

Storage, File Systems.
Click on the table Icon on the right and check the box for Mount Point.

You'll see drives mounted under their UUID's.

__________________________________________________________________

Why are your device names changing? If I were to guess, one of your many drives may have a bad cable or power connection. Maybe, if you have one of the cheaper HBA adapters, to attach additional drives beyond what the MOBO can handle, the card is not consistently reporting port numbers versus drive assignments to BIOS. (In that scenario, again if I were to guess, the drive(s) that are changing device names might be attached to that card.)

In any case, your scenario is one of the reasons why OMV mounts drives using UUID's in the mount point. UUID's don't change.

Display More

I mean, I could see reasons that the letters may change vs bad cabling, etc. Adding drives, etc. I've had mine change plenty of times after adding new drives, etc.

As you said though, not a concern if your system is set up properly.

BTW: I'm using a Dell HBA as well, in an Intel commercial server. It's a Dell Perc H200 that I flashed to JBOD. I haven't had any problems with it.

Quote from shawnhalu

oh so it doesn't matter?

It matters. If drive device names are changing frequently (after a reboot), there may be an impact on setting up drives and related tasks.. I'm not sure if SNAPRAID converts to UUID, after the initial configuration, but it would make sense to use the mounted (UUID) path. (When you mentioned "parity" and "data" I'm assuming SNAPRAID.)

If that's the case, are you having problems with SNAPRAID when you run a scrub?

Quote from shawnhalu

yes I am using hba to connect all the drive. it is a dell hba bought from "the art of server"

Well, I'm guessing it's a used card. Did it come inside of a used server? If not, if it was bought separately, was the card handled properly? (Did it come in an anti-stat bag?) Have you ever heard of ESD? (Electrostatic Discharge Damage.) ESD damage can cause maddening intermittent problems.

Easy to do items to check off the list of possibilities are:
-Reseat the HBA.

-Reseat SATA/SAS cables (both ends) and power connections.

- Look at SMART stat's to see if one of the drives are going south. (Any one drive can cause problems, especially with reordered device names.)

Look at the following:
SMART 5 – Reallocated_Sector_Count.

SMART 187 – Reported_Uncorrectable_Errors.

SMART 188 – Command_Timeout.

SMART 197 – Current_Pending_Sector_Count.

SMART 198 – Offline_Uncorrectable.

And the following:

SMART 199 - UltraDMA CRC errors

(Usually hardware or cable related)

___________________________________________________________________________

I'm curious as to, whether or not, the drives that are changing are attached to the HBA.
Storage, Disks will provide you with drive serial numbers versus device names. See if the changing drive devices are attached to the HBA.

Realize that I'm going out on a limb with these guesses. (They were made with very up front little information.)

Quote from Dr1ver

I have the same problem. Only drives connected to the motherboard in SATA3

And what version of OMV are you using?
_____________________________________________________________________

(Setting aside reseating an addon HBA,) some of the same checks would still apply:

Easy to do items to check off the list of possibilities are:

-Reseat SATA cables (both ends) and power connections.

- Look at SMART stat's to see if one of the drives are going south. (Any one drive can cause problems, especially with reordered device names.)

Look at the following:

SMART 5 – Reallocated_Sector_Count.

SMART 187 – Reported_Uncorrectable_Errors.

SMART 188 – Command_Timeout.

SMART 197 – Current_Pending_Sector_Count.

SMART 198 – Offline_Uncorrectable.

And the following:

SMART 199 - UltraDMA CRC errors

(Usually 199's are hardware interface or cable related)