Hi
Just wondering if anyone can clue me in on why the scrub time is so long on this? (16 days) Have I got something incorrectly configured or is that just what it is?
Ryzen 3600
X570 Crosshair
64 gig of ECC
6 x Seagate Exos X18 18TB
Thanks for any help, this is all new to me.
What is the pool layout? Raid-Z(1,2,3), Mirrors, etc. How full is the pool? How much data? More data would mean longer scrub times. Is that an estimated scrub time or the actual reported time from a previous scrub? Was there a lot of other access while the scrub was happening?
I think I am at about a day (22 hours 42 minutes 58 seconds) for aprox 13.88TiB on a single drive pool. Just started a scrub a little bit ago and the GUI is estimating it about a 1.5 days.
1 x RAIDZ2 | 6 wide | 16.37 TiB Used: 26.33 TiB Available: 38.95 TiB 40.3% Full. The NAS isnât doing anything else atm, not running anything on it locally, not copying to or from it etc. Itâs basically a big external HDD for how it gets used. I never had scrubs set up in the past but while helping a friend set up his recently I watched an updated guide that recommended setting it up monthly so this is the 1st scrub ever. Itâs dropped down to a mere 10 days and 11 hours now in the GUI.
Lots of small files and or fragmentation? It is a little bit high but nothing outrageous.
How long did your scrubs normally take?
Itâs all video files, though there is trickplay files for every video from jellyfin so maybe that? This is the 1st time I have scrubbed it since itâs been built, didnât know it was a thing.
Is the blocksize 128k? I used to get scrubs that took over a day for a similar set up with 16TB drives, but I had a much larger block size.
Where would I find that to check?
16 days to scrub 16TB is not normal.
Is there anything running in the background that would pull the drives away from a maximum throughput scrub?
What is the speed of the scrub?
zpool status mypool
zfs get -t filesystem,volume -r recordsize mypool
At which stage? Time estimates are initially way off the mark⌠and this may improve only slowly.
No chance that you have a Smart Long test running at the same time or anything else hitting the pool in terms of reads/writes?
ZFS Status gives me -
Last login: Tue Jun 30 16:41:44 ACST 2026 on pts/0
root@truenas[~]# zpool status
pool: NERV
state: ONLINE
status: One or more devices has experienced an unrecoverable error. An
attempt was made to correct the error. Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
using âzpool clearâ or replace the device with âzpool replaceâ.
see: Message ID: ZFS-8000-9P â OpenZFS documentation
scan: scrub in progress since Mon Jun 29 22:36:20 2026
4.82T / 39.5T scanned at 76.9M/s, 2.77T / 39.5T issued at 44.2M/s
0B repaired, 7.00% done, 10 days 02:21:00 to go
config:
NAME STATE READ WRITE CKSUM
NERV ONLINE 0 0 0
raidz2-0 ONLINE 0 0 0
sda2 ONLINE 0 0 0
sdb2 ONLINE 0 0 0
sdc2 ONLINE 0 0 0
sdd2 ONLINE 0 0 2
sdf2 ONLINE 0 0 0
sdg2 ONLINE 0 0 0
errors: No known data errors
pool: boot-pool
state: ONLINE
status: One or more features are enabled on the pool despite not being
requested by the âcompatibilityâ property.
action: Consider setting âcompatibilityâ to an appropriate value, or
adding needed features to the relevant file in
/etc/zfs/compatibility.d or /usr/share/zfs/compatibility.d.
scan: scrub repaired 0B in 00:00:33 with 0 errors on Sat Jun 27 03:45:34 2026
config:
NAME STATE READ WRITE CKSUM
boot-pool ONLINE 0 0 0
ata-Samsung_SSD_870_EVO_250GB_S6P2NL0TA03833T-part2 ONLINE 0 0 0
errors: No known data errors
Drive usage seems pretty weird but I donât know all that much.
Do you have any tasks shown on the Data Protection tab and Periodic S.M.A.R.T. Tests?
Check your Reporting, Disks tab over a days time. Is the graph the same for the entire period of your scrub? How are the disk temps?
I did a reboot and started a scrub. This is my graph for the boot drive and the data pool drive. This is what I would expect to see. System is idle with no Apps nor VMs. It just serves as SMB backup.
Indeed, the periodic spikes are strange.
But note that it has dropped from 16 to 10 days in one day, and will likely keep dropping.
That really looks like something else is running in the background & the scrub is only happening in short bursts.
Whats the output of smartctl -a /dev/sd(disknumberhere)? Anything in the output that indicates a smart test is running?
And with a 6am wandering past the room check in, yay itâs buzzing along now.
Next question though, what do I do at the end to clean up the pool? Any way to get a file list of the broken files?
You should be able to see it at the end with zpool status -v NERV Why do you think there are broken files? Are you showing a lot more errors across the drives?
Shouldnât you be checking for broken disks even before broken files? Degraded array and allâŚ
If it were just the two checksum errors on one drive, not increasing further, youâd do zpool clear NERV after checking drive health (parsing report after a long SMART test) and checking that cables are OK.
With a raidz2 your files are probably fine (software side), but you know face more investigation to find out why the pool went to âdegradedâ (likely hardware issue).
