Howdy folks, this has happened to me a few times, never the cache drives though so now im a little more concerned and want to be careful. ive noticed on rare occasions, typically after an update of some kind i will boot up unraid and my disks will not be properly assigned to my array or my unassigned device will be missing, but in its place a new drive will display. it will share a common name but have what looks like a new serial number attached to its name. i managed to rename my plex SSD partition mount point to match what my docker was configured to use and that fixed that issue. i then pointed my VM to its "new device" to attached to its hard drive file and now its working but how do i recover the 2 cache disks? starting the array they become disabled and emulated.

 

darktower-diagnostics-20220805-2026.zip

03:00.0 Serial Attached SCSI controller [0107]: Hewlett-Packard Company Smart Array Gen9 Controllers [103c:3239] (rev 01)
	Subsystem: Hewlett-Packard Company H240 [103c:21c7]
	Kernel driver in use: hpsa
	Kernel modules: hpsa
04:00.0 Serial Attached SCSI controller [0107]: Hewlett-Packard Company Smart Array Gen9 Controllers [103c:3239] (rev 01)
	Subsystem: Hewlett-Packard Company H240 [103c:21c7]
	Kernel driver in use: hpsa
	Kernel modules: hpsa

Are these RAID controllers? Inconsistent disk identification is one of many reasons RAID controllers are NOT recommended.

 

12 minutes ago, Clayton said:

starting the array they become disabled and emulated.

They are not emulated because the are not part of the parity array.

 

 

they might be raid controllers but i have them in JBOD mode. My LSI controller died on me a year or so ago and these were half the cost. they have worked great minus this issue which did seem to come up only after they were installed. you might be onto something about the cause being the controller card.

 

When i hover over the device, it states its disabled and contents emulated.

 

would you have any suggestion on how i might be able to recover the cache drives?

1 hour ago, Clayton said:

my disks will not be properly assigned to my array

What did you do on those occassions?

20 minutes ago, trurl said:

What did you do on those occassions?

Nothing out of the normal. Usually an update or I physically moved the server so I shut it down for a little while. 

Did the drive identifications fix themselves so that Unraid recognized them again?

20 minutes ago, trurl said:

Did the drive identifications fix themselves so that Unraid recognized them again?

in the past i was able to fix the unassigned or raid disk by moving on with the new identification but i haven't done anything for the cache drives this time around. im hoping there is something that can be done to re-initiate or something i can do to rebuild them. not sure i you have any suggestions? i dont have a way to change the drive identification that i know of. this happens at random or during an update of some kind. ive rebooted a few times since i started this thread. i confirmed the card is running in HBA mode. 

49 minutes ago, Clayton said:

moving on with the new identification

How exactly did you do that?

Just now, trurl said:

How exactly did you do that?

New Config?

38 minutes ago, trurl said:

New Config?

For example it was an unassigned drive. I renamed a mount partition and or repointed my docker or vm references. The drive and it's data is intact, just need to repoint to it. Idk about cache drives. Since my dockers and vms are running right now, does that mean my cache drive can be restored to the pool?

Edited August 6, 20223 yr by Clayton

7 hours ago, Clayton said:

they might be raid controllers but i have them in JBOD mode.

Problem is still controller related.

2 hours ago, JorgeB said:

Problem is still controller related.

its an HP H240 card running in HBA mode, a common card i see mentioned in the unraid forums but most mentions i have found is drives not being seen. I think its due to running older firmware. ill look into possibly replacing the card if that's the cause of my instability.

 

my main focus at this point is how i can recover the cache drives. they look to be in an emulation state so i would assume that means i can recover them. 

Edited August 6, 20223 yr by Clayton

25 minutes ago, Clayton said:

my main focus at this point is how i can recover the cache drives.

Unassign all pool members, start array, stop array, re-assign all pool members, start array and you're done.

On 8/6/2022 at 7:10 AM, JorgeB said:

Unassign all pool members, start array, stop array, re-assign all pool members, start array and you're done.

btw, pool is restored after removing the pool members and adding them back in. all is well, thank you for your help!

On 8/6/2022 at 4:10 AM, JorgeB said:

Unassign all pool members, start array, stop array, re-assign all pool members, start array and you're done.

Sorry to boot this back up as an issue. I am having issues with my Broadcom LSI 9305-24i. I recently went to a new PC case with 24 drive bays. The transition worked without issue while using the new LSI card.

Today as I try and update to 7.2.2>7.2.3. I boot in and my Array shows 2 missing devices(It selects the 2 drives missing without me needing to). One Parity and one Disk. I do not know what the serial numbers were for them as it is not telling me they are missing.. like what happens when I normally remove a drive. Would the method you mentioned work?

"Unassign all pool members, start array, stop array, re-assign all pool members, start array and you're done."
Unassign all pool member: You are just choosing no device on all parts of the array?
Then just adding the array back? Or does your method only work with pool devices and not array devices?

27 minutes ago, jtaz16 said:

I boot in and my Array shows 2 missing devices

They are not missing; they are disabled. Start the array and post the diagnostics.