Drive activity

[thereverend]

I have a 12 drive 2u server running TrueNAS. It appears that all12 drives are Constantly running all the time and reading and Writing all the time and
never stop. I have 2 other servers that do not seem to do this. Can someone tell me why this is happening? I don't want the drives to run 24/7/365 when there is
no data on them yet. I think this might destroy the drives prematurely. I do understand that drives are always spinning but not always
writing and reading. But these are read/writing all the time it would appear from the activity lights on the drives.Thanks for the assistance in advance.

 

[ChrisRJ]

The class of drives that should be in there (rated for 12 or more drives in one enclosure) will have no problems with constant read/write activity.

 

[joeschmuck]

Constantly running all the time and reading and Writing all the time and
never stop.

Well they don't do much maintenance for a system that is idle (no data transferring over Ethernet), so not sure why you think it's reading and writing. Maybe you have something setup incorrectly, assuming the drives are actually active (more than just spinning), for example you are running SMART Short or Long Tests too frequently, Running a Scrub frequently, something like that. You didn't specify the drive models. Look at the SMART data for head loading and Spinup cycles, see if those are incrementing and if they are, how much over what period of time. One other option is the head loading on some drives will Park the heads frequently (sometimes after only a few seconds) and then the drive gets polled and the heads will load, even if there isn't a data read in progress, then the drive heads park shortly after. Also, unplug the Ethernet cable to see if that stops the activity, it might give you a clue as to what is causing your situation.

I think this might destroy the drives prematurely.

If the heads stay loaded and there is constant activity, that will likely not cause premature failure, but the heads unloading and loading, the drive motor spinning down and up again, those will likely cause premature failure.

 

[thereverend]

Well they don't do much maintenance for a system that is idle (no data transferring over Ethernet), so not sure why you think it's reading and writing. Maybe you have something setup incorrectly, assuming the drives are actually active (more than just spinning), for example you are running SMART Short or Long Tests too frequently, Running a Scrub frequently, something like that. You didn't specify the drive models. Look at the SMART data for head loading and Spinup cycles, see if those are incrementing and if they are, how much over what period of time. One other option is the head loading on some drives will Park the heads frequently (sometimes after only a few seconds) and then the drive gets polled and the heads will load, even if there isn't a data read in progress, then the drive heads park shortly after. Also, unplug the Ethernet cable to see if that stops the activity, it might give you a clue as to what is causing your situation.


If the heads stay loaded and there is constant activity, that will likely not cause premature failure, but the heads unloading and loading, the drive motor spinning down and up again, those will likely cause premature failure.

On this system I have removed ALL data to see if it had anything to do with data and there is nothing on the drives. The reason I think the drives are reading and writing is due to the activity lights on the drives themselves. when they are idle they have 1 light lit (power) when they are reading or writing they have the other light lit (Activity) and the activity light on the drive is flickering on all drives all the time. As far as the smart setting. I have not set that up yet on the unit. I just got this server so it's still quite new to me.

"If the heads stay loaded and there is constant activity, that will likely not cause premature failure, but the heads unloading and loading, the drive motor spinning down and up again, those will likely cause premature failure." --- Ok so it's better for the drives to spin, read and write 24/7 then to shut down and spin up again and again.. got it.. I have not changed any of the settings for the drives for power etc so I assume by default they just keep running.


This is a learning experience for sure. Thank you for the replies..

 

[joeschmuck]

Your hard drives very well may be parking the heads periodically, so you must look at the SMART data in order to figure that out. Data on the drive is not what I asked you to check, I said you need to disconnect the Ethernet cable, if the activity stops then you know that something external to your NAS is trying to access the NAS, it doesn't matter if there is data on it or not. For example Windows will go out and look for connected devices and poll them for status. Disconnect the Ethernet cable and find out. Look at your SMART data to find out if your drives are experiencing activity which may lead them to die prematurely.

Do not assume the system defaults will protect you, defaults will work for more NAS hard drives but if you are running some non-NAS consumer grade drives, those may not be doing what you expect. Also you should post your hard drive make/model for all your drives so someone can tell you if there is some quirk about them such as they may park the heads every 5 seconds automatically and it my not be adjustable.

Cheers

 

[ChrisRJ]

Regarding the timeout for parking heads, I recently learned that data center drives (in my case Seagate Exos) have a pretty short interval by default. IIRC mine had two or three minutes, which is a duration of inactivity that would be pretty unusual in data center. I changed to ten minutes to be on the safe side.

 

[thereverend]

I have SAS drives in the 12 BAY unit . I have unplugged the Ethernet cable and the drives do still have activity lights flickering

 

[joeschmuck]

I have SAS drives in the 12 BAY unit . I have unplugged the Ethernet cable and the drives do still have activity lights flickering

Check the SMART data as I have requested, flashing lights no not tell the entire story. Also you might provide some details as to your hardware configuration since other than hard drives, you haven't mentioned anything else. Things you should know, there is a SWAP file created by default on each drive for use, if you have enough RAM then that shouldn't be an issue. Also if this was a new install then there is likely the System Dataset located on the pool which does get used frequently. Do a search for "FreeNAS move system dataset" if you desire to move the system dataset to a separate drive, some people will move it to a SSD so the hard drives can sleep or just to take the burden off the hard drives. I personally do not do that, the system works well as designed. But that is another possibility why you feel the drive is writing a lot.

 

[truefriend-cz]

I have same problem as user "thereverend". Neverending write/read. In monitoring HDD is small data size.
UTB cable is disconected. System Dataset moved on boot-pool but not problem solved.

Any solution?

I have a Synology NAS nas QNAP NAS and these disks in Synology and QNAP NAS don't have this problem.
The disks are in RAID 1 (mirror) mode.
Fresh, clean install.

Due to this activity, I can't put the disks to sleep after xx seconds and diks keeps clicking or writing.

Tested on:

TrueNAS 12, 13 (Beta)
TrueNAS Scale 22.02.00, 22.02.01 (MASTER-20220223-015752), 22.12 (nightly)

 

[Redcoat]

This is normal for TrueNAS - a server software OS that is not expected to sleep drives - and where sleeping drives is anyway expected to shorten the life of the drives vs leaving them powered.

This is a much discussed subject here - try a search for "drive standby" or "sleep drives" and you will find some solutions tried by others as well as discussion on the topic.

 

[truefriend-cz]

I tried to connect a HDD with transparent glass (i used for physical monitoring) to the TrueNAS server to see what the headers are doing and it shows the same symptom (sound) as the HDD I have for data storage. The headers constantly park and travel to the stove. About once every 2-3 seconds. When I sent it to the manufacturer to be able to tell me how much the HDD wears out with this process, he wrote me that it would not damage the disc plates in any way, but what is more stressed than the normal diameter is the motor that controls the read / write head and in such a highly used way sees lifespan (read heads) of 1-2 years. If the header does not park but only "flies" in the space above the HDD, the standard life of the disk is 5 years.
If the disk is destroyed due to parking and read / write heads, the warranty also applies to this item and it is not a problem to replace the disk. But I think it's useless.
Would it please if this will be read by any of the developers to take a look at this issue in a little more detail?
The problem is that even if one buys NVMe M.2 cards and uses them as a cache, it also does not solve this problem and the disks "click" further. Paradoxically, the cache should protect the disks from frequent reads and writes, but this BUG wears them out itself, although it does, it caches data during normal reading and writing.

 

[joeschmuck]

but this BUG wears them out itself

This is not a bug and all of what you are talking about has been very well documented and the drives do not prematurely wear out if you are using a CMR NAS drive (Not SMR), most last well beyond the warranty period. You are right now just stating your opinion without factual data so if you provide the exact model of the drive in question and the full output of the command "smartctl -a /dev/xxx" where xxx= the drive and we can tell you if there are any red flags for that drive. This would be factual data.

If you just want the drives to sleep because you want a silent system then while it may be possible to sleep the drives (I have not tried it on Scale) I do know that drives typically fail prematurely when they are constantly spun up causing power surges for the motor control circuity. The SMART output would confirm this is your case or not.

 

[truefriend-cz]

The drives are the new WD Red 6TB. I've already read this on the forum, but just that although there are so many posts on this topic, it's so weird that TrueNAS developers haven't solved it yet.
TrueNAS Scale is built on Debian. When I run pure Debian on the same server and install Samba (apt-get install samba) and configure smb.conf, etc.) it doesn't show this problem. So it will be something to do with TrueNAS.
I have already checked SMART and it does not show any problems. But I can get this information later.

I want to put the disks to sleep because I don't access the data on the disk all the time, but only occasionally. When I'm outside where I live, I go outside during the day and do other things than work with data on the server. If necessary, I will connect to it remotely via public IP, etc. And I would like to run services such as Plex, Web, etc. (which I can not now, because it would affect the monitoring of the problem). So most of the time, the work as such does not make up the data available for "shock actions". And at the same time I would like to create TrueNAS as an archive tool when I do not back up data to another NAS server but I would like to use a pair of disks in one server for live data and a couple of other disks using CRON from time to time and save data on them . Other NASs like Synology, QNAP, etc. can do this by default ... thanks to the fact that they work with hot-swap mode and SATA disks can be removed while the server is running. The problem is when those disks are unable to fall asleep. This then leaves thanks to the hotswap exchange very soon, because when the headers are just reading / writing, the disk will certainly be damaged very soon. If TrueNAS is published as Enterprise then I assume something like hot-swap removal and disk mounting should be the basis.

I think it's a BUG, because other NAS systems don't show such a problem, and also because even though I disconnect individual disks in the TrueNAS GUI with the "detach" command, they still work - the header still tries to park to read / write some data I don't know.

Unfortunately, I also failed to install any monitoring tools because TrueNAS Scale linux (Debian) is not allowed to install other applications (it has the "apt-get" command blocked and I also failed to install the tools from the manually saved DEB package). So I am not able to find out which file is currently being written or read.

 

[joeschmuck]

I think it's a BUG, because other NAS systems don't show such a problem

LOL, I'm not going to tell you that you are wrong but what you call a bug, someone else could call a feature. I for one and I know I speak for most here, prefer my drives to remain running. Think of it as a feature to ensure the drives maintain the highest level of readiness and able to respond with data immediately vice waiting for drives to spin up which would cause several seconds for a delay. Please keep in mind that this software is designed to support the business user as TrueNAS is sold as a product and we are just along for the free ride and we get the fun time to debug the software as a bonus.

I want to put the disks to sleep because I don't access the data on the disk all the time, but only occasionally.

That is understandable for a home NAS setup with very rare use, I could see that possibly being a benefit if you were spinning up the drive a few times a week at most. Another option you could do is use an older version of FreeNAS, maybe 11.3 and then use the script which sleeps the drives. Would this work for you? It's fairly recent and I think it would support your needs.

The other option I can see is to just use Debian or your favorite OS to create your own version of a home NAS. It sounds like you have the intelligence to do that and already have done that.

If you haven't already done so, submit a BUG report for the issue. The developers are good people and will respond as time permits.

I understand that you want the drive to spin down. Below is an email I get every day at 2AM, this is part of the email. It shows the status of my hard drives at a glance. Note that my drives have over 3.56 years of power on time and have only been spun up 134 times. Most of those times were due to me powering the system off for cleaning, some other event were moving and major power outages. Note the lack of errors. The drives still run quiet too so I'm very pleased with them lasting this long and I expect another 1.5 years out of them, that is a realistic expectation.

Spinning Rust Summary Report​

Device ID​

Serial Number​

Model Number​

HDD Capacity​

Rotational Rate​

SMART Status​

Temp​

Power-On Time​

Start/Stop Count​

Load Cycle Count​

Spin Retry Count​

Realloc Sectors​

Realloc Events​

Current Pending Sectors​

Offline Uncorr Sectors​

UltraDMA CRC Errors​

Seek Error Rate​

Multi Zone Error​

Last Test Age (days)​

Last Test Type​

/dev/ada0​

1JUMLBD​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

40*C​

31207​

134​

1506​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada1​

1JRSWLD​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

39*C​

31207​

134​

1505​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada2​

1JUMW4D​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

38*C​

31207​

134​

1502​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada3​

1GVD84B​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

38*C​

31207​

134​

1518​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

 

[truefriend-cz]

This is not a bug and all of what you are talking about has been very well documented and the drives do not prematurely wear out if you are using a CMR NAS drive (Not SMR), most last well beyond the warranty period. You are right now just stating your opinion without factual data so if you provide the exact model of the drive in question and the full output of the command "smartctl -a /dev/xxx" where xxx= the drive and we can tell you if there are any red flags for that drive. This would be factual data.

If you just want the drives to sleep because you want a silent system then while it may be possible to sleep the drives (I have not tried it on Scale) I do know that drives typically fail prematurely when they are constantly spun up causing power surges for the motor control circuity. The SMART output would confirm this is your case or not.

sda:

Code:

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


General SMART Values:
Offline data collection status:  (0x80) Offline data collection activity
                                        was never started.
                                        Auto Offline Data Collection: Enabled.
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:                (   87) seconds.
Offline data collection
capabilities:                    (0x5b) SMART execute Offline immediate.
                                        Auto Offline data collection on/off support.
                                        Suspend Offline collection upon new
                                        command.
                                        Offline surface scan supported.
                                        Self-test supported.
                                        No 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:        ( 693) minutes.
SCT capabilities:              (0x003d) 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     0x000b   100   100   016    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0004   132   132   054    Old_age   Offline      -       96
  3 Spin_Up_Time            0x0007   149   149   024    Pre-fail  Always       -       404 (Average 407)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       99
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   100   100   000    Old_age   Always       -       730
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       92
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       272
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       272
194 Temperature_Celsius     0x0002   153   153   000    Old_age   Always       -       39 (Min/Max 18/52)
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x000a   200   200   000    Old_age   Always       -       0


SMART Error Log Version: 1
No Errors Logged


SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]


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.

sdb:

Code:

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


General SMART Values:
Offline data collection status:  (0x80) Offline data collection activity
                                        was never started.
                                        Auto Offline Data Collection: Enabled.
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:                (   87) seconds.
Offline data collection
capabilities:                    (0x5b) SMART execute Offline immediate.
                                        Auto Offline data collection on/off support.
                                        Suspend Offline collection upon new
                                        command.
                                        Offline surface scan supported.
                                        Self-test supported.
                                        No 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:        ( 639) minutes.
SCT capabilities:              (0x003d) 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     0x000b   100   100   016    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0004   132   132   054    Old_age   Offline      -       96
  3 Spin_Up_Time            0x0007   148   148   024    Pre-fail  Always       -       406 (Average 407)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       99
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   100   100   000    Old_age   Always       -       730
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       92
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       272
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       272
194 Temperature_Celsius     0x0002   150   150   000    Old_age   Always       -       40 (Min/Max 17/52)
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x000a   200   200   000    Old_age   Always       -       0


SMART Error Log Version: 1
No Errors Logged


SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]


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.

 

[truefriend-cz]

LOL, I'm not going to tell you that you are wrong but what you call a bug, someone else could call a feature. I for one and I know I speak for most here, prefer my drives to remain running. Think of it as a feature to ensure the drives maintain the highest level of readiness and able to respond with data immediately vice waiting for drives to spin up which would cause several seconds for a delay. Please keep in mind that this software is designed to support the business user as TrueNAS is sold as a product and we are just along for the free ride and we get the fun time to debug the software as a bonus.


That is understandable for a home NAS setup with very rare use, I could see that possibly being a benefit if you were spinning up the drive a few times a week at most. Another option you could do is use an older version of FreeNAS, maybe 11.3 and then use the script which sleeps the drives. Would this work for you? It's fairly recent and I think it would support your needs.

The other option I can see is to just use Debian or your favorite OS to create your own version of a home NAS. It sounds like you have the intelligence to do that and already have done that.

If you haven't already done so, submit a BUG report for the issue. The developers are good people and will respond as time permits.

I understand that you want the drive to spin down. Below is an email I get every day at 2AM, this is part of the email. It shows the status of my hard drives at a glance. Note that my drives have over 3.56 years of power on time and have only been spun up 134 times. Most of those times were due to me powering the system off for cleaning, some other event were moving and major power outages. Note the lack of errors. The drives still run quiet too so I'm very pleased with them lasting this long and I expect another 1.5 years out of them, that is a realistic expectation.

Spinning Rust Summary Report​

Device ID​

Serial Number​

Model Number​

HDD Capacity​

Rotational Rate​

SMART Status​

Temp​

Power-On Time​

Start/Stop Count​

Load Cycle Count​

Spin Retry Count​

Realloc Sectors​

Realloc Events​

Current Pending Sectors​

Offline Uncorr Sectors​

UltraDMA CRC Errors​

Seek Error Rate​

Multi Zone Error​

Last Test Age (days)​

Last Test Type​

/dev/ada0​

1JUMLBD​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

40*C​

31207​

134​

1506​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada1​

1JRSWLD​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

39*C​

31207​

134​

1505​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada2​

1JUMW4D​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

38*C​

31207​

134​

1502​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

/dev/ada3​

1GVD84B​

HGST HDN726060ALE614​

[6.00TB]​

7200​

PASSED​

38*C​

31207​

134​

1518​

0​

0​

0​

0​

0​

0​

0​

0​

0​

Short​

The problem I described is not even a hint that I would decide whether or not to disc.
TrueNAS includes a feature that should allow Always operation, or turn off disk xx seconds after xx seconds. This feature cannot be used because ...

The problem I mentioned is that TrueNAS constantly parks / reads / writes the read / write head. And if that's how you write "normal" then I would also require a function so that it is possible to turn off this normal thing. I don't know of a single person who wants this "function". And I'm one of them.

I think that better than speculating on the individual hours of disk wear, you would (if you have the power) urge a solution to this problem. Whether in the Jira system or with your development colleagues.

And I would like you not to choose from the text I am writing to you. I'm not doing this to you either, and we're probably not kids or Russia and Ukraine to compete with each other. It is clear that if you select certain sentences, the context of the messages will also change. So please don't act like that. Thank you.

I would welcome a way to solve the problem. For example, to recommend me tools, commands with which I can find out which file or what data the read head reads / writes, or which application accesses the files. In my opinion, this will help to find out where the problem might be. If that's your interest at all.

Thank you

 

[joeschmuck]

@truefriend-cz
You cut off part of the output, the part that states the hard drive make/model and is important to me, I do not need the serial number.

The problem I mentioned is that TrueNAS constantly parks / reads / writes the read / write head.

If that is the problem you are reporting then I would agree that it is a problem. I was thinking you wanted the drives to sleep for long periods of time. You need to ensure that the drives are setup as follows: HDD Standby: ALWAYS ON, Advanced Power Management: DISABLED, Acoustic Level: DISABLED, S.M.A.R.T.: ENABLED/TRUE and this should eliminate the drives from sleeping, UNLESS the drive is a laptop type drive where it will put itself to sleep constantly. I have my test NAS setup with old laptop hard drives. They work for testing but they park after a few seconds on non-use. It makes some of my testing slow but they do what I need and they were scrap parts. And you can play with these settings but look at the SMART report ID's 192 and 193 to see if the heads are unloading still.

As for your drives, the data provided indicated that your drives were powered off 272 times and that the heads remain loaded the entire time the power was applied. This means that the drives did not spin down nor unload the heads, both of which are desirable aspects for a frequently used NAS.

These appear to be new drives so I must ask, did you perform any burn-in testing on them? If not, I'd recommend you do this to weed out any surface damage and give you some peace of mind. If you do not, that is fine, it is just a recommendation.

A concern... You have had your hard drives running for basically 30 days power-on time and have cycled power to them 272 times. I have had my four hard drives for 3.5 years and power cycled them half of what you have done. That is a lot. These values could be the same because the drive electronics mirrors the head load/unload cycles in which case sleeping the drive (motor spins down not power turned off), or you really have cycled power that many times. A little investigation on the make/model of the drive would result in knowing if the two values mirror each other at all times.

I would welcome a way to solve the problem. For example, to recommend me tools, commands with which I can find out which file or what data the read head reads / writes, or which application accesses the files. In my opinion, this will help to find out where the problem might be. If that's your interest at all.

If I knew the inner working of the TrueNAS application then I'd offer some advice. In the old days we would have the source code, modify it, and compile it. Anything good that came out of it would likely be incorporated into the FreeNAS. I have not done that in a great many years.

Good luck, hope you find the solution you are looking for.

 

[truefriend-cz]

@truefriend-cz
You cut off part of the output, the part that states the hard drive make/model and is important to me, I do not need the serial number.


If that is the problem you are reporting then I would agree that it is a problem. I was thinking you wanted the drives to sleep for long periods of time. You need to ensure that the drives are setup as follows: HDD Standby: ALWAYS ON, Advanced Power Management: DISABLED, Acoustic Level: DISABLED, S.M.A.R.T.: ENABLED/TRUE and this should eliminate the drives from sleeping, UNLESS the drive is a laptop type drive where it will put itself to sleep constantly. I have my test NAS setup with old laptop hard drives. They work for testing but they park after a few seconds on non-use. It makes some of my testing slow but they do what I need and they were scrap parts. And you can play with these settings but look at the SMART report ID's 192 and 193 to see if the heads are unloading still.

As for your drives, the data provided indicated that your drives were powered off 272 times and that the heads remain loaded the entire time the power was applied. This means that the drives did not spin down nor unload the heads, both of which are desirable aspects for a frequently used NAS.

These appear to be new drives so I must ask, did you perform any burn-in testing on them? If not, I'd recommend you do this to weed out any surface damage and give you some peace of mind. If you do not, that is fine, it is just a recommendation.

A concern... You have had your hard drives running for basically 30 days power-on time and have cycled power to them 272 times. I have had my four hard drives for 3.5 years and power cycled them half of what you have done. That is a lot. These values could be the same because the drive electronics mirrors the head load/unload cycles in which case sleeping the drive (motor spins down not power turned off), or you really have cycled power that many times. A little investigation on the make/model of the drive would result in knowing if the two values mirror each other at all times.


If I knew the inner working of the TrueNAS application then I'd offer some advice. In the old days we would have the source code, modify it, and compile it. Anything good that came out of it would likely be incorporated into the FreeNAS. I have not done that in a great many years.

Good luck, hope you find the solution you are looking for.

SDA:

Code:

smartctl 7.2 2020-12-30 r5155 [x86_64-linux-5.10.105+truenas] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org


=== START OF INFORMATION SECTION ===
Model Family:     Western Digital Red Pro
Device Model:     WDC WD6003FFBX-68MU3N0
Serial Number:    ---
LU WWN Device Id: 5 000cca 0bdee0dfe
Firmware Version: 83.00A83
User Capacity:    6,001,175,126,016 bytes [6.00 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    7200 rpm
Form Factor:      3.5 inches
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ACS-2, ATA8-ACS T13/1699-D revision 4
SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Sat Mar 19 23:52:34 2022 CET
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:  (0x82) Offline data collection activity
                                        was completed without error.
                                        Auto Offline Data Collection: Enabled.
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:                (   87) seconds.
Offline data collection
capabilities:                    (0x5b) SMART execute Offline immediate.
                                        Auto Offline data collection on/off support.
                                        Suspend Offline collection upon new
                                        command.
                                        Offline surface scan supported.
                                        Self-test supported.
                                        No 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:        ( 693) minutes.
SCT capabilities:              (0x003d) 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     0x000b   100   100   016    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0004   132   132   054    Old_age   Offline      -       96
  3 Spin_Up_Time            0x0007   149   149   024    Pre-fail  Always       -       404 (Average 407)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       99
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   100   100   000    Old_age   Always       -       758
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       92
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       273
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       273
194 Temperature_Celsius     0x0002   153   153   000    Old_age   Always       -       39 (Min/Max 18/52)
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x000a   200   200   000    Old_age   Always       -       0


SMART Error Log Version: 1
No Errors Logged


SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]


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.

SDB:

Code:

smartctl 7.2 2020-12-30 r5155 [x86_64-linux-5.10.105+truenas] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org


=== START OF INFORMATION SECTION ===
Model Family:     Western Digital Red Pro
Device Model:     WDC WD6003FFBX-68MU3N0
Serial Number:    ---
LU WWN Device Id: 5 000cca 0bdf05ff0
Firmware Version: 83.00A83
User Capacity:    6,001,175,126,016 bytes [6.00 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    7200 rpm
Form Factor:      3.5 inches
Device is:        In smartctl database [for details use: -P show]
ATA Version is:   ACS-2, ATA8-ACS T13/1699-D revision 4
SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Sat Mar 19 23:53:57 2022 CET
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:  (0x82) Offline data collection activity
                                        was completed without error.
                                        Auto Offline Data Collection: Enabled.
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:                (   87) seconds.
Offline data collection
capabilities:                    (0x5b) SMART execute Offline immediate.
                                        Auto Offline data collection on/off support.
                                        Suspend Offline collection upon new
                                        command.
                                        Offline surface scan supported.
                                        Self-test supported.
                                        No 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:        ( 639) minutes.
SCT capabilities:              (0x003d) 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     0x000b   100   100   016    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0004   132   132   054    Old_age   Offline      -       96
  3 Spin_Up_Time            0x0007   148   148   024    Pre-fail  Always       -       406 (Average 407)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       99
  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   067    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   128   128   020    Old_age   Offline      -       18
  9 Power_On_Hours          0x0012   100   100   000    Old_age   Always       -       758
 10 Spin_Retry_Count        0x0012   100   100   060    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       92
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       273
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       273
194 Temperature_Celsius     0x0002   150   150   000    Old_age   Always       -       40 (Min/Max 17/52)
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x000a   200   200   000    Old_age   Always       -       0


SMART Error Log Version: 1
No Errors Logged


SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]


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.

My settings of the disks from first time:

 

[joeschmuck]

So you have some good quality hard drives with a 5 year warranty. Again, I'd recommend you change your settings to have the HDD Standby to ALWAYS ON instead of 30 minutes. I'm sure that if you really want the drives to sleep, you will find a solution.

 

[Jessep]

Where is your System Dataset located?