Hard drives temperatures show * but SMART is enabled and reports it


vizi0n

Recommended Posts

Hi,

 

I am kind of wondering what is happening. My hard drives do report SMART data, including temperature (194). None of the drive show the temperature in the GUI (see screenshots) even though they are spun up, except for the one that is mounted with the "Unassigned devices" plugin. Before adding that plugin, the drives would not report temperature at all, any of them.

 

Is there any tweak I can do?

screenshot_unraid.png.a765eeaf1584cfc6fc3ff78ddbc33740.png

Screen-Shot-2016-12-14-at-10_53.27-PM.png.bc993330a10fc5f7cb822b70490703bc.png

Link to comment

This was already done at the very beginning of my build, and eveytime I add a new drive because they are not recognized at all under unRAID if I dont initialize+JBOD them all. The drives are currently fully working. Only the GUI doesn't show the temperature is the devices list.

 

I kind of looked at the code of the unassigned devices plugin and found that it grabs the info using this command:

smartctl -A -d sat,12 $dev 2>/dev/null| /bin/grep -m 1 -i Temperature_Celsius | /bin/awk '{print $10}'"

 

Now for the Dynamix part, it seems like it grabs the info about the drives from /var/local/emhttp/disks.ini

 

How is that file generated? This file has the * in it at the temperature line

Link to comment
  • 3 months later...
3 minutes ago, vizi0n said:

Bump? I still haven't found how the disks.ini file is generated. I would love to see the actual temperature of my drives. And since smartctl can report it, why doesn't it show ?

disks.ini is generated by a background process in unRaid.  Temperatures are pushed into it at the frequency set in settings - disk settings - Poll Attributes (defaults to every 30 minutes)

 

Don't have an adaptec controller, so can't help you out beyond that info  :/

Link to comment

But the strange thing is that the smart reports of all drives on that controller are showing the temperature. As you can see the screenshot.

 

So unRAID should be able to read it, I guess.

 

Is it possible to manually run the script/process to update that data? I would love playing around with it

Link to comment
  • 2 years later...

Have you encountered for the potential 30 minute delay (with default poll settings) before the GUI updates these settings?   You can always reduced the poll interval (I have reduced mine to 5 minutes)  if you want but on some systems the polling results in disk being spun up.

Link to comment
  • 1 year later...
5 minutes ago, johnnie.black said:

If you're using a custom controller it's a known issue.

Yeah, i see. I investigated further, but couldn't find where the disks.ini is generated.

Maybe a user script updatiung the temps in that ini file would temp fix this problem, until it gets picked up and worked on...

Or modify the rc.diskinfo php script for writing that info into the ini file.

Edited by DarkMan83
Link to comment
  • 3 months later...

I just picked up an ML350P Gen8 and am going through the same issue currently.  I put the P420i controller into HBA mode after getting all the firmware up to date and just have been doing some testing before I try to migrate everything over.   I have a hodgepodge of SATA drives installed, different brands sizes, etc. and haven't noticed any issues with the fan speeds since running Unraid so I am taking that as a good sign.

 

Unraid lists no temps or SMART data for any of the drives, but I am able to retrieve it via the smartctl command as you mention.  One of my ideas was to try to use iLO and set an alarm for this but of course there is no easy way to install the agent that reports this data into Unraid, although I am considering trying to convert the rpm into a txz if there is no easy way to obtain the data in the GUI.  If you are still working on this, I would be glad to exchange notes here and see if we can't figure out a way to solve this.  It's so strange to me that Unraid can be so polished in many aspects and fall flat in others, especially when it's paid software and this mature.

 

Regardless, I am going to try some things tonight and will share what I come up with if noteworthy.

Link to comment

I just spent the whole evening trying to figure out what is going on here.  After playing around int he Global Disk Settings, trying various SMART controller types and testing, I think I have some answers.  I set the SMART controller to HP cciss globally and it wouldn't work.  I tried a few others, eventually landing on "SAT," whatever the hell that may mean.  To my surprise it returned all of the data I was looking for.  Every one of the fields are populated and temperature data works, although like in the case of OP, it doesn't transfer to the Main page.

 

As mentioned before, the Dynamix webGUI is pulling the data from the /var/local/emhttp/disks.ini file.  You would think this information might be related to the data in the files contained in the /var/local/emhttp/smart/ directory which seem to query the SMART data from the disk, but it appears that the files in the smart folder are not connected in any way.  In my case, the directory contains basic text file for each disk that contains what would be the smartctl output for that device, as well as a file of the same name with a .ssa extension.  No matter what SMART controller I select in the disk settings, the information in these files does not seem to change and just reports the following:

 

Quote

 

smartctl 7.1 2019-12-30 r5022 [x86_64-linux-4.19.107-Unraid] (local build)
Copyright (C) 2002-19, Bruce Allen, Christian Franke, www.smartmontools.org

/dev/sdc: requires option '-d cciss,N'
Please specify device type with the -d option.

Use smartctl -h to get a usage summary

 


After doing some more searching, none of these files have anything to do with the Dashboard or Main pages.  It seems that others have had issues with the Areca controllers in which it has been stated that the smart reporting on Dashboard and Main pages are hard-coded in emhttp and the parameters are not able to be defined by the user.  I checked out the webgui code and it and I want to say they are currently working on a fix for this, as there was a commit last month removing the smartctl command from the monitor script.  The extent of my PHP knowledge is reading a couple chapters of PHP in 24 hours 20 years ago when I was in middle school, so I could be full of bs here.  I just spent way too much time into this last night when I could have been implementing a script to alert me to issues via another method.  Hopefully some devs can chime in on what is going on or if anyone here is familiar with the codebase and wants to check it out.  It definitely seems like something too simple to just leave not working properly, especially when thats kind of an important feature for this OS.

 

 

Link to comment
4 hours ago, Benson said:

Are you sure disk would report temperature in standby / idle ? Otherwise, you are wakeup those disk.

Yes, I was getting all smart data when running manually.  Looking at files in /var/local/emhttp/smart, it is clear that the underlying command hard-coded into the dynamix webgui is not running properly nor is it affected by what settings are entered for either each disk or globally.  See my last post though, I explored the code a little bit last night and believe that this is something currently be resolved and would expect to see in next beta release.  Lots of lines were removed in and only one added, and I am not even sure how to interpret it correctly:

if (file_exists("$file") && exec("grep -Pom1 '^SMART.*: \K[A-Z]+' ".escapeshellarg($file)." |tr -d '\n' 2>/dev/null", $ssa) && in_array("$ssa",$failed)) {

as referenced in this commit on July 12: https://github.com/limetech/webgui/commit/6f8507e5474e9b77fef836ee7379a1bee25a7a5b

 

  • Like 1
Link to comment
  • 1 month later...
On 8/25/2020 at 1:46 PM, greg_gorrell said:

I tried a few others, eventually landing on "SAT," whatever the hell that may mean.

SAT means ATA disk connected via a SCSI port, i.e. SATA. The Smartctl code defaults all sdX devices to scsi then tries to do an open and scsi inquiry. If the drive vendor is ATA then device type in smartctl is changed from SCSI to SAT.

 

root@Tower:/usr/local/sbin# ./smartctl -r ioctl /dev/sdc
smartctl 7.2 2020-09-19 r5083 [x86_64-linux-5.8.12-Unraid] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

>>>> do_scsi_cmnd_io: sg_io_ver=3
 [inquiry: 12 00 00 00 24 00 ]
  scsi_status=0x0, sg_transport_status=0x0, sg_driver_status=0x0
  sg_info=0x0  sg_duration=0 milliseconds  resid=0
>>>> do_scsi_cmnd_io: sg_io_ver=3
 [ata pass-through(16): 85 08 0e 00 00 00 01 00 00 00 00 00 00 00 ec 00 ]
  scsi_status=0x0, sg_transport_status=0x0, sg_driver_status=0x0
  sg_info=0x0  sg_duration=2 milliseconds  resid=0
/dev/sdc [SAT]: Device open changed type from 'scsi' to 'sat'
ATA device successfully opened

Use 'smartctl -a' (or '-x') to print SMART (and more) information
 

You can use -d test to see this.

 

root@Tower:/usr/local/sbin# ./smartctl -d test /dev/sdc
smartctl 7.2 2020-09-19 r5083 [x86_64-linux-5.8.12-Unraid] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

/dev/sdc: Device of type 'scsi' [SCSI] detected
/dev/sdc [SAT]: Device open changed type from 'scsi' to 'sat'
/dev/sdc [SAT]: Device of type 'sat' [ATA] opened

 

 

 

Link to comment
  • 1 year later...

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.