February 28, 20251 yr Hello all. Spent the last 4 hours scouring the forum for the answer but have been unable to find an answer. My problem began when I bought a new ssd for my 2 drive zfs pool. Easy enough I removed the old drive and added the 2tb SSD Samsung Evo. That is when the downward spiral began. I was unable to mount the zfs pool I tried to clear the ssd but kept getting the same error invalid pool. I decided to do a newconfig so I can get rid of the pool. Once doing that I was unable to access my cache disk or unable to add back its a ZFS cache pool of one NVME drive. My main array is up but the cache is show in unassigned disks. I will add the diagnostic log. Appreciate any help. Thanks. Edited March 2, 20251 yr by HAVOC spelling
March 1, 20251 yr Community Expert The screenshot shows a mounted pool, which device(s) was the original pool, also post the output from: zpool status and zpool import
March 1, 20251 yr Author 1 hour ago, JorgeB said: zpool status root@Tower:~# zpool status pool: zfs state: ONLINE scan: resilvered 104G in 00:21:16 with 0 errors on Mon Jan 27 22:34:49 2025 config: NAME STATE READ WRITE CKSUM zfs ONLINE 0 0 0 sdc1 ONLINE 0 0 0 root@Tower:~# zpool import no pools available to import I reformatted my cache and restored from back up. Problem is this ZFS pool wont accept the SAMSUNG drive. It just hangs, when I remove it the array boots normally. Its odd. Edited March 1, 20251 yr by HAVOC
March 1, 20251 yr Community Expert Type zpool export zfs Stop the array, create a new pool with 1 slot, assign the sdc device, start array, post new diags.
March 1, 20251 yr Author I will give it a shot. My monthly parity sync has kicked off so I’ll wait for it to complete. I’ll post results when it’s done.
March 2, 20251 yr Author root@Tower:~# zpool export zfs cannot export 'zfs': pool is busy I removed the device restarted the array and then the command worked thank you!! When I try to add the samsung ssd the server is locked up and wont come online Mar 2 07:28:07 Tower emhttpd: mounting /mnt/zfs Mar 2 07:28:07 Tower emhttpd: shcmd (842512): mkdir -m 0666 -p /mnt/zfs Mar 2 07:28:07 Tower emhttpd: shcmd (842513): /usr/sbin/zpool import -f -m -N -o autoexpand=on -d /dev/sdc1 -d /dev/sdo1 9079094678971493684 zfs Mar 2 07:28:08 Tower emhttpd: zfs: zfs verify devices Mar 2 07:28:08 Tower emhttpd: /usr/sbin/zpool status -P zfs 2>&1 Mar 2 07:28:08 Tower emhttpd: pool: zfs Mar 2 07:28:08 Tower emhttpd: state: ONLINE Mar 2 07:28:08 Tower emhttpd: scan: resilvered 104G in 00:21:16 with 0 errors on Mon Jan 27 22:34:49 2025 Mar 2 07:28:08 Tower emhttpd: config: Mar 2 07:28:08 Tower emhttpd: NAME STATE READ WRITE CKSUM Mar 2 07:28:08 Tower emhttpd: zfs ONLINE 0 0 0 Mar 2 07:28:08 Tower emhttpd: /dev/sdc1 ONLINE 0 0 0 Mar 2 07:28:08 Tower emhttpd: errors: No known data errors Mar 2 07:28:08 Tower emhttpd: shcmd (842514): /usr/sbin/zfs set -u mountpoint=/mnt/zfs zfs Mar 2 07:28:08 Tower sysDrivers: SysDrivers Build Complete Mar 2 07:28:08 Tower emhttpd: shcmd (842515): /usr/sbin/zfs mount -o atime=off zfs Mar 2 07:28:09 Tower rc.nginx: Reloading Nginx server daemon... Mar 2 07:28:09 Tower emhttpd: shcmd (842516): /usr/sbin/zpool set autotrim=on zfs Mar 2 07:28:09 Tower emhttpd: shcmd (842517): /usr/sbin/zfs set compression=on zfs Mar 2 07:28:09 Tower rc.nginx: Checking configuration for correct syntax and then trying to open files referenced in configuration... Mar 2 07:28:09 Tower rc.nginx: /usr/sbin/nginx -t -c /etc/nginx/nginx.conf Mar 2 07:28:09 Tower rc.nginx: Reloading Nginx server daemon configuration... Mar 2 07:28:09 Tower emhttpd: shcmd (842518): /sbin/wipefs -af --lock /dev/sdo1 Mar 2 07:28:09 Tower emhttpd: shcmd (842519): /sbin/blkdiscard /dev/sdo1 Edited March 2, 20251 yr by HAVOC
March 2, 20251 yr Community Expert I cannot see why it's stuck, there are no errors, reboot and before starting the array post the output from zpool import
March 2, 20251 yr Author Its odd. Once I physically remove the drive the array it instantly comes on.
March 2, 20251 yr Author root@Tower:~# zpool import pool: zfs id: 9079094678971493684 state: ONLINE action: The pool can be imported using its name or numeric identifier. config: zfs ONLINE sdd1 ONLINE root@Tower:~#
March 2, 20251 yr Community Expert Solution Assuming the other device is still sdo, format it with the UD plugin first, any filesystem, it's just to create the partition, then with the array stopped type, if it's not sdo adjust the command zpool import zfs zpool attach -f zfs /dev/sdd1 /dev/sdo1 Then zpool status
March 2, 20251 yr Author Yup that was the trick sir. Thank you. root@Tower:~# zpool status pool: zfs state: ONLINE status: One or more devices is currently being resilvered. The pool will continue to function, possibly in a degraded state. action: Wait for the resilver to complete. scan: resilver in progress since Sun Mar 2 11:23:56 2025 153G / 153G scanned, 19.0G / 153G issued at 92.4M/s 19.0G resilvered, 12.48% done, 00:24:39 to go config: NAME STATE READ WRITE CKSUM zfs ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 sdc1 ONLINE 0 0 0 sdd1 ONLINE 0 0 0 (resilvering) errors: No known data errors
March 3, 20251 yr Community Expert Not sure why the GUI was hanging, but now you should just need to reimport the pool with both devices.
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.