SOLVED ZFS state DEGRADED. Disks get removed spontanously

[Tarrega]

Hi everyone,
I am trying to set up my own FreeNas Server. I feel that I need to start by apologizing, since I noticed the firm recommendations against AMD here. So I am ashamed to say that I tried it anyway. I read this german guide where my hardware combination worked fine:
CPU: AMD Athlon 5350 with AES-Ni support
Motherboard: ASRock AM1H-ITX
RAM: 8GB Crucial Ballistix DDR3-1600, 1.5V
Case: CoolerMaster Elite 120 Advanced, Mini-ITX
FreeNas Drive: SanDisk Extreme SLC 16GB USB 3.0
Storage: 4x WD Red 4 TB
I used this hardware because of the low power consumption and the hardware AES support.
I use FreeNAS-9.3-STABLE-201506162331.

Here is my problem:
I created an encrypted (passphrase not yet set) RaidZ with 4x 4 TB WD Red disks.
The 4 disks were set to Standby after 60 mins and Advanced Power Management Level 1.
I created a CIFS share and started to copy data from my Windows 7 PC to it. During that, I get the Alarm:
"The volume DatenRaid1 (ZFS) state is DEGRADED: One or more devices has been removed by the administrator. Sufficient replicas exist for the pool to continue functioning in a degraded state."
I checked the Raid Volume Status page and found that one disk had the status "REMOVED". I could just click "Replace" and add it back into the raid array. The resilvering process started and finished and everything was healthy again.
Then 3 hours later the same thing happened with the same disk (ada3).
I set Standby to "always on" and the Advanded Power Management Feature of all disks off.
Then a few hours later 2 disks at the same time were removed from the array.
With that the array was unavailable. I couldn't add the disks back to the array. I rebooted the server. Then the pool was locked and I couldn't unlock it anymore.

I ran some short SMART tests on all 4 disks, and everything looked fine. I added the smart output of ada3 below.

What do you think this is? What can cause this removal of the disks?
Is it a disk, other hardware or a software problem? Can the power management cause this if a disk spins down?

Thanks for any help.

Code:

smartctl 6.3 2014-07-26 r3976 [FreeBSD 9.3-RELEASE-p16 amd64] (local build)
Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:  Western Digital Red (AF)
Device Model:  WDC WD40EFRX-68WT0N0
Serial Number:  WD-WCC4E5CP183H
LU WWN Device Id: 5 0014ee 20bc9bbf2
Firmware Version: 82.00A82
User Capacity:  4,000,787,030,016 bytes [4.00 TB]
Sector Sizes:  512 bytes logical, 4096 bytes physical
Rotation Rate:  5400 rpm
Device is:  In smartctl database [for details use: -P show]
ATA Version is:  ACS-2 (minor revision not indicated)
SATA Version is:  SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:  Sat Jul 11 14:49:33 2015 CEST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x00)   Offline data collection activity
           was never started.
           Auto Offline Data Collection: Disabled.
Self-test execution status:  (  0)   The previous self-test routine completed
           without error or no self-test has ever
           been run.
Total time to complete Offline
data collection:      (52680) seconds.
Offline data collection
capabilities:         (0x7b) SMART execute Offline immediate.
           Auto Offline data collection on/off support.
           Suspend Offline collection upon new
           command.
           Offline surface scan supported.
           Self-test supported.
           Conveyance Self-test supported.
           Selective Self-test supported.
SMART capabilities:  (0x0003)   Saves SMART data before entering
           power-saving mode.
           Supports SMART auto save timer.
Error logging capability:  (0x01)   Error logging supported.
           General Purpose Logging supported.
Short self-test routine
recommended polling time:     (  2) minutes.
Extended self-test routine
recommended polling time:     ( 527) minutes.
Conveyance self-test routine
recommended polling time:     (  5) minutes.
SCT capabilities:     (0x703d)   SCT Status supported.
           SCT Error Recovery Control supported.
           SCT Feature Control supported.
           SCT Data Table supported.

SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME  FLAG  VALUE WORST THRESH TYPE  UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate  0x002f  200  200  051  Pre-fail  Always  -  0
  3 Spin_Up_Time  0x0027  253  176  021  Pre-fail  Always  -  2875
  4 Start_Stop_Count  0x0032  100  100  000  Old_age  Always  -  14
  5 Reallocated_Sector_Ct  0x0033  200  200  140  Pre-fail  Always  -  0
  7 Seek_Error_Rate  0x002e  200  200  000  Old_age  Always  -  0
  9 Power_On_Hours  0x0032  100  100  000  Old_age  Always  -  64
 10 Spin_Retry_Count  0x0032  100  253  000  Old_age  Always  -  0
 11 Calibration_Retry_Count 0x0032  100  253  000  Old_age  Always  -  0
 12 Power_Cycle_Count  0x0032  100  100  000  Old_age  Always  -  14
192 Power-Off_Retract_Count 0x0032  200  200  000  Old_age  Always  -  5
193 Load_Cycle_Count  0x0032  200  200  000  Old_age  Always  -  41
194 Temperature_Celsius  0x0022  114  114  000  Old_age  Always  -  38
196 Reallocated_Event_Count 0x0032  200  200  000  Old_age  Always  -  0
197 Current_Pending_Sector  0x0032  200  200  000  Old_age  Always  -  0
198 Offline_Uncorrectable  0x0030  100  253  000  Old_age  Offline  -  0
199 UDMA_CRC_Error_Count  0x0032  200  200  000  Old_age  Always  -  0
200 Multi_Zone_Error_Rate  0x0008  100  253  000  Old_age  Offline  -  0

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
Num  Test_Description  Status  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline  Completed without error  00%  56  -
# 2  Short offline  Completed without error  00%  54  -

SMART Selective self-test log data structure revision number 1
 SPAN  MIN_LBA  MAX_LBA  CURRENT_TEST_STATUS
  1  0  0  Not_testing
  2  0  0  Not_testing
  3  0  0  Not_testing
  4  0  0  Not_testing
  5  0  0  Not_testing
Selective self-test flags (0x0):
  After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.

 

[Robert Trevellyan]

Is it a disk, other hardware or a software problem?

Not much information to go on, but I'd say 'other hardware' seems most likely. I don't see any issues in the SMART data for ada3. Do you trust your PSU?

Can the power management cause this if a disk spins down?

Seems unlikely, but I really don't know.

 

[Tarrega]

Do you trust your PSU?

Well it is a no-name external switching power supply with 15V instead of the demanded 19V. It has on output of 60W.
In the guide a demand of 36W was measured under load.
In the bios I checked the Voltage. The 12V and the 5V lines deviate only by about 0.1V.
Do you thing I should by a brand name 19V power supply with a higher output?

 

[Robert Trevellyan]

In the guide a demand of 36W was measured under load.

At startup? Drive spin-up is the most demanding time for the PSU. If your drives were actually powering down, you may have been stressing your under-specced PSU repeatedly.

Do you thing I should by a brand name 19V power supply with a higher output?

I think if the application demands 19V then that's what you should give it. Delivering 15V when 19V is expected seems like asking for trouble. I'm not familiar with the technicalities, but others have stated in these forums that under-volting a hard drive can cause permanent damage.

 

[Tarrega]

under-volting a hard drive can cause permanent damage.

Yes, I would agree there. But the disks get the power from the mainboard and you can see in the bios that the voltage does not deviate from the designed 5 and 12V, but I have to add that the bios values are of course not under load.

 

[Darren Myers]

But the disks get the power from the mainboard

Errrr what....? Do you have another known good, name brand powersupply to try?

 

[pirateghost]

Yes, I would agree there. But the disks get the power from the mainboard and you can see in the bios that the voltage does not deviate from the designed 5 and 12V, but I have to add that the bios values are of course not under load.

I see no way for the motherboard to supply power to the hard drives. This is the most ridiculous thing I have read today....

 

[Tarrega]

Errrr what....? Do you have another known good, name brand powersupply to try?

I mean, the power goes through the mainboard, gets transformed there into different rails, like 5 and 12V.
No, I don't have another power supply. I would buy one, if someone with experience with that suggests it.

 

[Tarrega]

I see no way for the motherboard to supply power to the hard drives. This is the most ridiculous thing I have read today....

Have a look at my mainboard ASRock AM1H-ITX.
It has a connector for an external power supply. The power has to go through the mainboard. It has a power connector for hard disks on it.

 

[pirateghost]

I don't see any hard drive power

 

[Tarrega]

I don't see any hard drive power

It is between the ATX-Connector and the SATA Data Connectors.

 

[pirateghost]

The one marked panel1?

It doesn't change the fact that your power supply is still underpowered...

 

[Tarrega]

The one marked panel1?
It doesn't change the fact that your power supply is still underpowered...

No, the one marked "SATA_POW1". And, yes It may be underpowered, I am not arguing that.

 

[Ericloewe]

F#$%!/@ Technikaffe strikes again. Those morons should just stop spreading their useless bullshit.

I see no way for the motherboard to supply power to the hard drives. This is the most ridiculous thing I have read today....

Some boards (notably Supermicros C2xxx miniITX boards) can be supplied with just +12V/GND and have onboard circuitry to provide 5V and 3.3V, including a limited amount to drives.
Some boards also have the rather useless feature of providing power that can be shut off to drives (like my ASRock X99 WS - needless to say, it's totally unused).


SMART data looks ok, but those drives are nowhere close to "burned-in". For your sake, I recommend you test them properly before using them.

Also, encryption is not to be taken lightly it is designed to fail by locking you out. It will not easily fail into a "everyone can read this" scenario. Do not use encryption unless absolutely necessary. And do not use encryption before you have carefully and thoroughly read everything the manual and the forum stickies have to say about it. I repeat, it is not for the average user.

CPU: AMD Athlon 5350 with AES-Ni support
Motherboard: ASRock AM1H-ITX
RAM: 8GB Crucial Ballistix DDR3-1600, 1.5V
Case: CoolerMaster Elite 120 Advanced, Mini-ITX
FreeNas Drive: SanDisk Extreme SLC 16GB USB 3.0
Storage: 4x WD Red 4 TB

Seriously, it irritates me profoundly that those morons keep spreading these senseless builds. Then it's FreeNAS that gets a bad reputation.

 

[pirateghost]

I have a thin client converted to a firewall and web filter that has onboard power for a hard drive. But it is limited to only providing enough power for a laptop drive.

I wouldn't even trust a system like that to run my server. I'm appalled that any website would recommend such a horrible setup....

 

[Ericloewe]

I have a thin client converted to a firewall and web filter that has onboard power for a hard drive. But it is limited to only providing enough power for a laptop drive.

I wouldn't even trust a system like that to run my server. I'm appalled that any website would recommend such a horrible setup....

They've been recommending J1900 and ultra-low-end AMD stuff for FreeNAS repeatedly. They are uninformed idiots with just enough knowledge to make them dangerous - and they actively spread their lack of knowledge, screwing people over in the process.

 

[Tarrega]

but those drives are nowhere close to "burned-in". For your sake, I recommend you test them properly before using them.

Thanks for the advice, I will do that with this how-to.

encryption is not to be taken lightly it is designed to fail by locking you out. It will not easily fail into a "everyone can read this" scenario. Do not use encryption unless absolutely necessary.

Yes, I see that now. I think I will forget about this kind of encryption.

 

[gpsguy]

Buy a conventional power supply like a Seasonic G-450.

 

[Tarrega]

I bought a 90W 19V notebook power supply. Since then my Freenas Setup runs now for three months without a problem. Thanks for the help!