Hey guys, with the release of Unraid 7.0.1 I've formated all my drives in ZFS and added a new ZFS mirror cache pool with two Sandisk Ultra 3D SSDs.
Since that, all my drives spin up twice a day exactly at the same times.
This happens every day and stresses my three HDDs unessescary and ruins one of the major benefits of Unraid for an silent and power efficient homeserver.

 

System Spes:

CPU: Intel Core i9 9900

RAM: 64GB DDR-4 3200 Kingston Fury Beast (2x32GB, JEDEC 2400)

Mainboard: Asrock H370M-ITX/ac (latest BIOS P4.30)

Additional NIC: Dual 2.5G Realtek RTL8125 (rev 5)

PSU: Corsair SF450

Case: Fractal Node 304

 

The disks spin up every day at 15:44 CEST and 18:59 CEST, the first one spins up all disks with the exclusion of disk1 (sdb) and the second one only disk2 (sdc) and the parity.

All shares for docker, vms and also the libvirt and docker.img and the system shares are configured to be on my NVME SSD, no shares with permanent acces are configured to be on my array.

 

I've tried all this things to solve the problem, with no sucess so far

• disbled/uninstalled all plugins
• disabled vm manger and docker
• unplugged ethernet cables
• created a new unraid flashdrive with a trial license, with minimal config without plugins etc. only docker, vms and my shares where configured
• BIOS reset with unchanged default settings
• checked my userscripts
• checked mover and autotrim
• downgraded from 7.0.1 to 7.0.0
• several server restarts
• changed the clock of BIOS and Unraid itself -3h

 

I really have no idea what triggers the disks to spin up, none of my actions stopped the spin ups.
Only the change of the system time to -3h in Unraid changed the behavior, with that my disks spin up 3 hours later than usual, but the exact time for the spin up stays the same. Changing the time in BIOS didn't changed anything.

 

I have an unraid offsite backup server with almost the same config and also all disks configured as ZFS and this server does not spin up the disks twice a day. The hardware is differnt, its an amd AM4 system, but besides that there is do difference.

The disks in my offsite backup server where formated to ZFS with Unraid 6.12.0 that integratet ZFS for the first time, now the version of this server is the latest 7.0.1 release, and I've no problems.

My current server with the spin up prolems where 7.0.1 when I've formated the disks to ZFS, maybe it's a bug in ZFS that persists in the newest Unraid version?  It is also strange, that even with a new Unraid flashdrive the problem persists, without any modified config files on the flashdrive and plugins etc.

 

Does anyone have a idea what can trigger the disks to spin up at the same times every day? I'am very curious what is causing this, there must be a process or something hardcoded anywhere, but I've no clue where to look further?

tower-diagnostics-20250401-2059.zip

Edited April 8, 20251 yr by Clepto
Added system specs

This is a very strange one.

38 minutes ago, Clepto said:

that even with a new Unraid flashdrive the problem persists, without any modified config files on the flashdrive and plugins etc.

Does this mean a flash drive using a stock Unraid install, with nothing from your current config?

 

 

Yes, i took a new flashdrive and created a new install with the Unraid USB Creator with the newest stable 7.0.1 release. There where no configs from the previous flashdrive imported.

 

After that I‘ve reassigned my drives with my parity marked as valid and created my shares and settings for dockers and vms new by hand.

So the problem occurs again with the new install, even without plugins and modified configs on the flashdrive.

Do you have any open connections to the server from say another PC on your network that's polling the NAS?

There are no active or open connections at the moment, i've tried to leave the server for hours without an ethernet cable and the disks also spun up.

 

I can't remeber the exact time when i've formatet my HDDs in ZFS, but maybe 15:44 and 18:59 are the exact times when they finished? Sounds weird, iknow ^^

But i can't explain that otherwise, seems like there is a tag in ZFS which triggers checks or something.

 

Some people here already complaint about many reads and writes on ZFS disks, i've read all the threads about it, but there wasn't an clear answer why this happens?

When my disks spin up daily, I also have some reads and writes (disk statistics counter), i've tried to trace that with the File Activity Plugin and also Open Files, but there were no entries. 

No idea what it could be, don't think I've seen a similar case before, and for sure it's not a general issue, even confirmed by your other server.

4 hours ago, Clepto said:

disks spin up every day at 15:44 CEST and 18:59 CEST

What do you get from command line with this?

cat /etc/cron.d/root

 

My crontab for root looks like this

# Generated docker monitoring schedule:
10 0 * * * /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/dockerupdate check &> /dev/null

# Generated system monitoring schedule:
*/1 * * * * /usr/local/emhttp/plugins/dynamix/scripts/monitor &> /dev/null

# Generated mover schedule:
40 3 1 * * /usr/local/sbin/mover start |& logger -t move

# Generated plugins version check schedule:
10 0 * * * /usr/local/emhttp/plugins/dynamix.plugin.manager/scripts/plugincheck &> /dev/null

# Generated Unraid OS update check schedule:
11 0 * * * /usr/local/emhttp/plugins/dynamix.plugin.manager/scripts/unraidcheck &> /dev/null


# CRON for CA background scanning of applications
44 * * * * php /usr/local/emhttp/plugins/community.applications/scripts/notices.php > /dev/null 2>&1

 

The other ones for cron.daily, hourly, weekly and monthly are empty.

i have the same issue while my zfs pool (with 3 HDDs) is waken up at the same time everyday. i dont know what process is doing it, but it is so frustrating.

22 hours ago, zunzun951218 said:

waken up at the same time everyday

On 4/8/2025 at 10:34 AM, trurl said:

What do you get from command line with this?

cat /etc/cron.d/root

My crontab looks like this

# Generated system monitoring schedule:

*/1 * * * * /usr/local/emhttp/plugins/dynamix/scripts/monitor &> /dev/null

# Generated mover schedule:

40 3 * * * /usr/local/sbin/mover start &> /dev/null

# Generated zfs scrub spinner schedule:

0 0 1 * * /usr/local/emhttp/plugins/dynamix/scripts/zfs_scrub start spinner &> /dev/null

# Generated zfs scrub ssd schedule:

0 0 1 * * /usr/local/emhttp/plugins/dynamix/scripts/zfs_scrub start ssd &> /dev/null

# CRON for CA background scanning of applications

45 * * * * php /usr/local/emhttp/plugins/community.applications/scripts/notices.php > /dev/null 2>&1

# Refresh Recycle Bin trash sizes every five minutes:

*/5 * * * * /usr/local/emhttp/plugins/recycle.bin/scripts/get_trashsizes &> /dev/null

But when i checked logs for disk spin up time, it were always at 00:45, 08:55, 16:50 and 21:25, so i dont think its related to cron job. btw, my log looks like this, so i don't know exactly what is being written or read.

--- 2025-05-22 00:45:22 Read:104 sectors Write:3200 sectors ---

Open files under /mnt/spinner:

A quick update on 23th and 24th. I tried these commands(via chatgpt recommendation), i logged off my browser and then the pool is not waken up anymore. I will be looking up for more detailed informations about effects of these command on the pool.

zfs set sync=disabled spinner

echo 60 > /sys/module/zfs/parameters/zfs_txg_timeout

echo 0 > /sys/module/zfs/parameters/zfs_prefetch_disable

zfs set atime=off spinner 

Edited May 24, 20251 yr by zunzun951218

I have a similar problem, except that my disks all wake up once a day at the same time. I have no idea when it started for me and trying to narrow down the problem for days/weeks and it must be related to ZFS.

Initially I tried to isolate the problem on my productive unraid server, but no chance, the disks all wake up once a day at the same time. In the LOG it looks like this:

May 27 12:51:43 unraid emhttpd: read SMART /dev/sdb
May 27 12:51:58 unraid emhttpd: read SMART /dev/sdc
May 27 13:06:44 unraid emhttpd: spinning down /dev/sdb
May 27 13:07:00 unraid emhttpd: spinning down /dev/sdc

Since I couldn't get any further, I wiped the entire server and reinstalled unraid without much configuration. I only mounted the disks unformatted as before. Result: Even a freshly installed server wakes up the disks once a day. The disks started up again at the same time a day later, although nothing was configured in the system.

Now I took a closer look at the disks and simply reformatted them all again. What can I say, the disks now start up every few minutes after formatting. During the last attempt I formatted one disk a little later than the other and this time difference can be seen in the LOGs.

root@unraid:~# zpool history hdd (sdc)
History for 'hdd':
2025-05-26.12:47:25 zpool create -f -o ashift=12 -O dnodesize=auto -O acltype=posixacl -O xattr=sa -O utf8only=on -m /mnt/hdd hdd /dev/sdc1
2025-05-26.12:47:25 zpool export -f hdd
2025-05-26.12:47:26 zpool import -f -m -N -o autoexpand=on -d /dev/sdc1 10996738503950216059 hdd
2025-05-26.12:47:26 zfs set -u mountpoint=/mnt/hdd hdd
2025-05-26.12:47:26 zpool set autotrim=off hdd
2025-05-26.12:47:26 zfs set compression=on hdd
2025-05-26.12:49:58 zpool export -f hdd
2025-05-26.12:51:32 zpool import -f -m -N -o autoexpand=on -d /dev/sdc1 10996738503950216059 hdd
2025-05-26.12:51:32 zfs set -u mountpoint=/mnt/hdd hdd
2025-05-26.12:51:32 zpool set autotrim=off hdd
2025-05-26.12:51:32 zfs set compression=on hdd
2025-05-27.15:23:41 zpool export -f hdd
2025-05-27.15:23:56 zpool import -f -m -N -o autoexpand=on -d /dev/sdc1 10996738503950216059 hdd
2025-05-27.15:23:57 zfs set -u mountpoint=/mnt/hdd hdd
2025-05-27.15:23:57 zpool set autotrim=off hdd
2025-05-27.15:23:57 zfs set compression=on hdd
2025-05-27.15:30:39 zpool export -f hdd
2025-05-27.15:32:10 zpool import -f -m -N -o autoexpand=on -d /dev/sdc1 10996738503950216059 hdd
2025-05-27.15:32:10 zfs set -u mountpoint=/mnt/hdd hdd
2025-05-27.15:32:10 zpool set autotrim=off hdd
2025-05-27.15:32:10 zfs set compression=on hdd

root@unraid:~# zpool history ssd (sdb)
History for 'ssd':
2025-05-27.15:24:29 zpool create -f -o ashift=12 -O dnodesize=auto -O acltype=posixacl -O xattr=sa -O utf8only=on -m /mnt/ssd ssd /dev/sdb1
2025-05-27.15:24:30 zpool export -f ssd
2025-05-27.15:24:30 zpool import -f -m -N -o autoexpand=on -d /dev/sdb1 5500916159231983332 ssd
2025-05-27.15:24:30 zfs set -u mountpoint=/mnt/ssd ssd
2025-05-27.15:24:30 zpool set autotrim=on ssd
2025-05-27.15:24:30 zfs set compression=on ssd
2025-05-27.15:30:39 zpool export -f ssd
2025-05-27.15:32:10 zpool import -f -m -N -o autoexpand=on -d /dev/sdb1 5500916159231983332 ssd
2025-05-27.15:32:10 zfs set -u mountpoint=/mnt/ssd ssd
2025-05-27.15:32:11 zpool set autotrim=on ssd
2025-05-27.15:32:11 zfs set compression=on ssd

May 28 12:51:55 unraid emhttpd: read SMART /dev/sdc
May 28 13:06:57 unraid emhttpd: spinning down /dev/sdc
May 28 15:32:21 unraid emhttpd: read SMART /dev/sdb
May 28 15:47:22 unraid emhttpd: spinning down /dev/sdb

You can see from the history and the syslog that the disks always spin up a few minutes after formatting.

Could this be a bug? Completely crazy, as if something in ZFS tests or monitors something on the disks every 24 hours.

System
ODROID-H4 Ultra

48 GB Crucial RAM

Drives:

• Cache WD_Blue_SN570_2TB

• Hdd ST16000NM000J-2TW103 (sdc)

• Ssd WD_Blue_SA510_2.5_4TB (sdb)

On 5/24/2025 at 6:05 AM, zunzun951218 said:

A quick update on 23th and 24th. I tried these commands(via chatgpt recommendation), i logged off my browser and then the pool is not waken up anymore. I will be looking up for more detailed informations about effects of these command on the pool.

I would like to leave the sync or atime activated. So this is probably just a workaround.

Edited May 29, 20251 yr by BiNiCKNiCH

I'm experiencing the same issue.

The disk spins up once a day at the same time (13:45 GMT+9).

This occurs only on the ZFS pool and never happened on the existing XFS pool.

I am certain it is caused by ZFS.

Below are the actions I've already taken:

- Disabled all scheduled tasks

- Disabled ZFS autotrim

- Removed the ZFS master plugin

- Disabled all shares on that pool

- Disabled Docker

- Disabled VMs

- Disconnected the Ethernet cable

If you've resolved this issue or are experiencing the same symptoms, please share your experience.

On 12/5/2025 at 9:11 AM, Leash Your Hardware said:

我遇到了同样的问题。

该磁盘每天在同一时间（13:45 GMT+9）启动。

这仅在 ZFS 池上发生，且在现有的 XFS 池上从未发生过。

我确定这是由 ZFS 引起的。

以下是我已经采取的措施：

- 已禁用所有定时任务

- 已禁用 ZFS 自动修剪

移除了 ZFS 主插件

禁用该存储池上的所有共享

- 已禁用 Docker

- 已禁用 虚拟机

- 以太网线已断开连接

如果您已解决此问题或遇到相同症状，请分享您的经验。

Meet exactly same problem but at (14:20:46 GMT+8)
completely at a loss, i've uninstalled zfs master plugin, but still with Snapshot plugin installed(take snapshots for btrfs drivers)

Recommend following this bug report; this is being investigated. https://forums.unraid.net/bug-reports/stable-releases/v720-v723-there-is-a-bug-that-keeps-waking-up-most-of-my-hdds-everyday-at-the-same-time-v714-is-the-last-version-where-everything-works-fine-r4228/?&do=getNewComment&d=2&id=4228

On 12/25/2025 at 8:11 PM, JorgeB said:

Recommend following this bug report; this is being investigated. https://forums.unraid.net/bug-reports/stable-releases/v720-v723-there-is-a-bug-that-keeps-waking-up-most-of-my-hdds-everyday-at-the-same-time-v714-is-the-last-version-where-everything-works-fine-r4228/?&do=getNewComment&d=2&id=4228

This is the solution, thank you very much for your effort JorgeB!

Now everything works as intended and I finally know the reason for the spin ups and don't have to assume anymore that my server is cursed ^^