Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

Cache pool no longer available

Featured Replies

Yesterday i installed the new Folderview Plus and after setup and creating a new folder the UI suddenly was stuck.

After a force reload in the browser I also rebooted the server but was unable to start the array as the cache pool (2 Samsung 990 EVO Plus) was giving me a config failed message when i tried to start the array.

When checking the forums (this thread for example) i removed the cache pool and recreate it, but now it says 'Unmountable: wrong or no file system'.

Hope no data is lost from the NVMe drives, but how can i get to it when the pool cannot be mounted?

blkid output gives me this

/dev/sdb1: LABEL="data" UUID="8685187093077842655" UUID_SUB="10709511807475224983" BLOCK_SIZE="4096" TYPE="zfs_member" PARTUUID="e0008686-01"

/dev/sda1: LABEL_FATBOOT="UNRAID" LABEL="UNRAID" UUID="B485-160D" BLOCK_SIZE="512" TYPE="vfat"

/dev/loop1: BLOCK_SIZE="131072" TYPE="squashfs"

/dev/loop0: BLOCK_SIZE="131072" TYPE="squashfs"

Diagnostics are attached.

Also i was already able to make an image with the dd command

dd if=/dev/nvme1n1 of=/mnt/disks/data1-2tb/nvme0.img bs=1M status=progress

dd if=/dev/nvme1n1 of=/mnt/disks/data2-2tb/nvme1.img bs=1M status=progress

and that for both pools

zdb -l /dev/nvme0n1 gives me no labels

zdb -l /dev/nvme1n1 gives the following output

------------------------------------

LABEL 0

------------------------------------

version: 5000

name: 'cache'

state: 1

txg: 3491005

pool_guid: 10896664973895392852

errata: 0

hostname: 'Serenity'

top_guid: 6220366269392501499

guid: 13938023003294794310

vdev_children: 1

vdev_tree:

type: 'mirror'

id: 0

guid: 6220366269392501499

whole_disk: 0

metaslab_array: 132

metaslab_shift: 33

ashift: 12

asize: 1000198897664

is_log: 0

create_txg: 4

children[0]:

type: 'disk'

id: 0

guid: 13938023003294794310

path: '/dev/nvme0n1p1'

whole_disk: 0

DTL: 395

create_txg: 4

children[1]:

type: 'disk'

id: 1

guid: 12792291676762730391

path: '/dev/nvme1n1p1'

whole_disk: 0

DTL: 519

create_txg: 4

features_for_read:

com.delphix:hole_birth

com.delphix:embedded_data

com.klarasystems:vdev_zaps_v2

labels = 0 1

failed to unpack label 2

failed to unpack label 3

How can i continue to recover data from the nvme1.img file and format the cache pool again on the server and then copy the data back over.

Guess trying to recover the ZFS pool itself is not an immediate option, correct?

serenity-diagnostics-20260330-1806.zip

  • Community Expert

Post the output from zpool import and also

fdisk -l /dev/nvme1n1

fdisk -l /dev/nvme0n1

  • Author

zpool import

no pools available to import

which makes sense since it currently shows 'Unmountable: wrong or no file system'

fdisk -l /dev/nvme1n1

Disk /dev/nvme1n1: 931.51 GiB, 1000204886016 bytes, 1953525168 sectors

Disk model: Samsung SSD 990 EVO Plus 1TB            

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

fdisk -l /dev/nvme0n1

Disk /dev/nvme0n1: 931.51 GiB, 1000204886016 bytes, 1953525168 sectors

Disk model: Samsung SSD 990 EVO Plus 1TB            

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

  • Community Expert

Both devices are missing the partition. Was the pool created in Unraid?

  • Author

Yes the pool was created in the Unraid UI itself

  • Community Expert

Type sfdisk /dev/nvme0n1, then type 2048 and hit enter and post the output; don't type anything else for now

  • Author

Got this as output and have not typed anything else yet.

Created a new DOS (MBR) disklabel with disk identifier 0xdcb90663.

Created a new partition 1 of type 'Linux' and of size 931.5 GiB.

Partition #1 contains a zfs_member signature.

Do you want to remove the signature? [Y]es/[N]o:

  • Community Expert

Type N to keep the signature, then type write and enter

Repeat the same for the other device; if a signature is found, keep it as well, then once both partitions have been recreated, try zpool import again

  • Author

Thanks so far @JorgeB ,

Second one just gives me this

Created a new DOS (MBR) disklabel with disk identifier 0xa4782ddd.

Created a new partition 1 of type 'Linux' and of size 931.5 GiB.

/dev/nvme1n1p1 :         2048   1953525167 (931.5G) Linux

Should I also just type write here?

Also what is the preferred command for zpool import, have little experience using ZFS commands through the terminal.

  • Community Expert
30 minutes ago, Stiibun said:

Should I also just type write here?

Note if it doesn't find the zfs signature

31 minutes ago, Stiibun said:

Also what is the preferred command for zpool import

Just open a terminal window and type that command, then paste here the output, do it now, with just the partition restored on one pool device

  • Author

Will come back on this later today and post the output of the zpool import command, at work now. One has to work to pay for this hobby 😉

For the second NVMe it indeed didn't find a zfs signature.

  • Author

@JorgeB , output of the zpool import command is as follows;

  pool: cache

    id: 10896664973895392852

state: DEGRADED

status: One or more devices contains corrupted data.

action: The pool can be imported despite missing or damaged devices.  The

fault tolerance of the pool may be compromised if imported.

   see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J

config:

cache          DEGRADED

  mirror-0     DEGRADED

    nvme0n1p1  ONLINE

    nvme1n1p1  UNAVAIL

  • Community Expert

You should be able to import the pool degraded using the device that had the partition restored, then add the other device to recreate the mirror.

If the pool is still assigned, first go to 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 1 slot

assign just nvme0n1, leave the filesystem set to auto

start the array to import the pool

Pool should import degraded, if it doesn't post new diags; if it does, stop the array, assign the other device to the disabled slot, start array to resilver.

  • Author

Well that seemed to have ended in a panic after following your instructions, so removed pool after stopping array. Created a new pool with the same name and 1 slot and assigned nvme01n1 to it and file system on auto.

After starting the array I see the following happening (terminal window was still opened)

Message from syslogd@Serenity at Mar 31 18:53:54 ...

kernel:VERIFY0(zap_lookup(spa->spa_meta_objset, spa->spa_feat_enabled_txg_obj, feature->fi_guid, sizeof (uint64_t), 1, res)) failed (52)

Message from syslogd@Serenity at Mar 31 18:53:54 ...

kernel:PANIC at zfeature.c:294:feature_get_enabled_txg()

  • Author

OK this took some time but the pool with 1 slot is now back.

  • Author

Now I see this after the pool was imported, but in the UI it's not immediately clear that the pool is degraded?

serenity_cache_pool.png

Edited by Stiibun

  • Author

@JorgeB , server keeps at Array Starting after adding the pool.

And taking diagnostics doesn't seem to work either as it hangs at

smartctl -x  '/dev/sda' 2>/dev/null|todos >'/serenity-diagnostics-20260401-0756/smart/Samsung_Flash_Drive_0355025010003213-0-0-2025-12-18 flash (sda).txt'
/usr/sbin/zpool status 2>/dev/null|todos >>'/serenity-diagnostics-20260401-0756/system/zfs-info.txt'
  • Community Expert

The pool is possibly panininc on import.

Instead of starting the array, see if you can import the pool manually read-only

zpool import -o readonly=on cache

If that works, look for the data in /mnt/cache and copy it somewhere else, then redo the pool.

If the pool still fails to import read-only, see if you can get the output from this tail -n 100 /proc/spl/kstat/zfs/dbgmsg

  • Author

@JorgeB , it panics again when executing the import manually

zfs dbgmsg logs the following;

timestamp message

1775041272 ffff88811c58b300 metaslab.c:1703:spa_set_allocator(): spa allocator: dynamic

1775041272 ffff88811c58b300 spa.c:6921:spa_tryimport(): spa_tryimport: importing cache

1775041272 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load($import-ffff88811c58b300-cache, config trusted): LOADING

1775041272 ffff8881a77fa200 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': probe done, cant_read=0 cant_write=1

1775041273 ffff88810696c400 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme1n1p1': open error=2 timeout=1007543090/1000000000

1775041273 ffff88811c58b300 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': best uberblock found for spa $import-ffff88811c58b300-cache. txg 3491005

1775041273 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load($import-ffff88811c58b300-cache, config untrusted): using uberblock with txg=3491005

1775041273 ffff8881a77fa200 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': probe done, cant_read=0 cant_write=1

1775041274 ffff88819b3e0000 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme1n1p1': open error=2 timeout=1008705958/1000000000

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading checkpoint txg

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading indirect vdev metadata

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Checking feature flags

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading special MOS directories

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading properties

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading AUX vdevs

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading vdev metadata

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading dedup tables

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading BRT

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Verifying Log Devices

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Verifying pool data

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Calculating deflated space

1775041274 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Starting import

1775041274 ffff88811c58b300 spa.c:9320:spa_async_request(): spa=$import-ffff88811c58b300-cache async request task=2048

1775041274 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load($import-ffff88811c58b300-cache, config trusted): LOADED

1775041274 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load($import-ffff88811c58b300-cache, config trusted): UNLOADING

1775041274 ffff88811c58b300 metaslab.c:1703:spa_set_allocator(): spa allocator: dynamic

1775041274 ffff88811c58b300 spa.c:6766:spa_import(): spa_import: importing cache

1775041274 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load(cache, config trusted): LOADING

1775041274 ffff88819d183300 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': probe done, cant_read=0 cant_write=1

1775041275 ffff888118a58000 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme1n1p1': open error=2 timeout=1004162104/1000000000

1775041275 ffff88811c58b300 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': best uberblock found for spa cache. txg 3491005

1775041275 ffff88811c58b300 spa_misc.c:430:spa_load_note(): spa_load(cache, config untrusted): using uberblock with txg=3491005

1775041275 ffff88819d183300 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme0n1p1': probe done, cant_read=0 cant_write=1

1775041276 ffff8881027a6600 vdev.c:184:vdev_dbgmsg(): disk vdev '/dev/nvme1n1p1': open error=2 timeout=1000928757/1000000000

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading checkpoint txg

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading indirect vdev metadata

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Checking feature flags

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading special MOS directories

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading properties

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading AUX vdevs

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading vdev metadata

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading dedup tables

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Loading BRT

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Verifying Log Devices

1775041276 ffff88811c58b300 spa_misc.c:2405:spa_import_progress_set_notes_impl(): 'cache' Verifying pool data

  • Community Expert

Pool may have other issues besides the missing device. You can try this to see if it can recover:

zpool import -o readonly=on -f cache

If that still fails:

zpool import -o readonly=on -F -f cache

and as a last resort:

zpool import -o readonly=on -X -f cache

  • Author

@JorgeB , to conclude this thread none of the commands you provided a solution I each time got a ZFS panic warning and needed to reboot each time. Also tried with a Ubuntu Live USB stick but to no avail.

So I refreshed my USB stick as I wanted to start a fresh but now I have troubles with the registration. Will create a new post for this.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.