Hello!

I would greatly appreciate some help to diagnose my cache drive suddenly showing as "Unmountable: Unsupported or no file system ". Unfortunately, it is a single drive in cache, and I had no idea that CA Backup and Restore was a thing, since I set this up years ago.

I first noticed a problem when Plex would not load my shares. Looking into the server, I saw that there was a SMART error flag for my cache drive, and that is was tagged as unmountable.

I ran an extended SMART self test, which passed.

Then, after reading a few forum posts, I tried the following:

Stop array and unassign both pool devices

Start array (click checkbox "I want to do this")

Stop array

Reassign both pool devices

Start array, post new diags.

Upon reading a suggestion to disable Docker and VM Manager, I did that as well.

I thought it best to start a new post with my diagnostics attached before I accidentally dug myself into a deeper hole.

I would greatly appreciate any guidance!

htpc-diagnostics-20260422-2340.zip

2 hours ago, Frankieoc82 said:

tried the following:

Stop array and unassign both pool devices

Start array (click checkbox "I want to do this")

Stop array

Reassign both pool devices

How could you do "both" when you only have one?

Do you know which file system was the pool?

Also post the output from

fdisk -l /dev/sdi

and

blkid /dev/sdi1

4 hours ago, trurl said:

How could you do "both" when you only have one?

Fair point. I should have said I CHECKED both. and disabled Docker. sorry for any confusion

Thank you @trurl and @JorgeB for your help!

The file system was btrfs. (typing "was" hurts a bit...)

Here are the outputs from the commands requested:

blkid /dev/sdi1 did not output anything

There are currently no partitions. Type sfdisk /dev/sdi, then type 2048 and hit enter and post the results, without typing anything else for now.

Ok, thank you very much for the guidance!

I typed sfdisk /dev/sdi then 2048. Here is the output:

Sorry, I'm not sure how, but I missed that the device does have a partition, but the partition doesn't have a btrfs signature. Hit CTRL + C to abort (that command doesn't create the partition on the actual device before it's written, so nothing was changed), and then type instead:

dd if=/dev/sdi bs=16M status=progress 2>/dev/null | hexdump -C | grep '_BHRfS_M'

If there's any output after you get the cursor back, post that, It will take a few minutes.

@JorgeB I ran the command above. It started running, printing a new line every couple of minutes. Unfortunately, when I came back, a few minutes later, the terminal window had closed by itself (which has happened a few times before).

Would it be detrimental to rerun that same command so i can post the output? Or should i pull some log that could show you the output?

Any way I can prevent that from happening again?

37 minutes ago, Frankieoc82 said:

terminal window had closed by itself (which has happened a few times before).

How are you accessing the command line?

2 hours ago, Frankieoc82 said:

Would it be detrimental to rerun that same command so i can post the output?

You can run it again, zero problem.

2 hours ago, trurl said:

How are you accessing the command line?

2 hours ago, trurl said:

How are you accessing the command line?

I usually access the server by local IP from a browser my laptop, then click the command line button from the GUI, which pops open a floating window. I can do it directly from the server though if that is preferable

36 minutes ago, JorgeB said:

You can run it again, zero problem.

Ok great! I will run it again when I get home

15 hours ago, JorgeB said:

You can run it again, zero problem.

Here is the full output from running the command you requested:

That lines up correctly with a btrfs filesystem on sector 64, which is the current partition starting sector, suggesting the main superblock at least is damaged, becuase no fs is bound

Let's see if it can be recovered from the backup superblock, post the output from:

btrfs-select-super -s 1 /dev/sdi1

7 hours ago, Frankieoc82 said:

Here is the full output from running the command you requested:

Thanks for checking this out. That prognosis sounds both good and not so good. I'm still hoping for a miracle though.

Here is the output:

13 hours ago, JorgeB said:

That lines up correctly with a btrfs filesystem on sector 64, which is the current partition starting sector, suggesting the main superblock at least is damaged, becuase no fs is bound

Let's see if it can be recovered from the backup superblock, post the output from:

btrfs-select-super -s 1 /dev/sdi1

Does this output look as expected?

using SB copy 1, bytenr 67108854

It does, does blkid /dev/sdi1 now show something?

If it does, try mounting the pool again, but it's better to reimport:

on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and number of slots

assign the pool device(s), leave the filesystem set to auto

start the array to import the pool

1 hour ago, JorgeB said:

It does, does blkid /dev/sdi1 now show something?

If it does, try mounting the pool again, but it's better to reimport:

on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and number of slots

assign the pool device(s), leave the filesystem set to auto

start the array to import the pool

IT WORKED!!!!!

YOU ARE A GENIUS!!! THANK YOU SOO MUCH!!

For the record and future reference:

Yes, blkid /dev/sdi1 did show the following now:

With the array stopped, I reimported and remounted the cache pool exactly as described, and started the array. The cache drive mounted!

Then I re-enabled Docker in settings and it all came back to life!

I can't thank you enough, @JorgeB !!

NOW, how do i prevent this from happening again? I think i am at my disk limit so i need to remove a data disk and rebalance before adding another drive for a 2 drive cache pool. I can do that if that is the recommended next step. Also , what is the best way to have automatic backups of cache and appdata?

I'm so grateful that this is such a supportive community. Thanks again!

Glad to hear. As for what caused it, it's difficult to say, that should not happen in normal usage. The main recommendation is to have good backups, for appdata you can use the appdata backup plugin, but it's important to back up anything else important on the pool, that can be with a script, snapshots, etc.

@JorgeB Unfortunately, it looks like I was celebrating a bit too early. The cache drive appears to still be causing some problems after the last steps fixed the unmountable drive issue.

The Cache drive is indeed recognized as a mountable drive now. I started the Docker service, and the dockers did all come back in the Docker tab.

Just after that, I went to plex to see if media was accessible. Some did load in the media library menu, but none would actually initiate. I then tried to access other dockers through their ip web GUIs and none would load.

I decided that before I dig the hole deeper and mess something up even more, I took your advice to install the Appdata Backup plugin and tried to manually run it. The backup failed twice, where a couple of dockers' appdata did get backed up successfully, but several other important dockers failed. The log shows walls of tar errors, like this:

Currently, all dockers are stopped, array is started, and I haven't done anything else (not even reboot) for fear of creating more errors.

I'd greatly appreciate your help while I continue troubleshooting this!

New diags, please.

New diagnostics

htpc-diagnostics-20260429-0658.zip

Edited April 29Apr 29 by Frankieoc82

The SSD dropped offline:

Apr 28 22:35:26 HTPC emhttpd: device /dev/sdi has size zero

Apr 28 22:35:26 HTPC emhttpd: offline: Samsung_SSD_860_EVO_250GB_S3YHNX0K835131V (sdi) 512 488397168

Replace its cables and try again

39 minutes ago, JorgeB said:

The SSD dropped offline:

Apr 28 22:35:26 HTPC emhttpd: device /dev/sdi has size zero

Apr 28 22:35:26 HTPC emhttpd: offline: Samsung_SSD_860_EVO_250GB_S3YHNX0K835131V (sdi) 512 488397168

Replace its cables and try again

I shut down, replaced the SATA cable with a new one and reseated the power connector, then rebooted and took new diagnostics, attached.

htpc-diagnostics-20260429-0823.zip