HDD temps in dashboard 30 minutes out-of-date?

[CS01-HS]

I run the autofan plugin (which I've set to check HDD temps every 5 minutes) and the HDDTemp docker for Grafana.

 

Comparing HDD temps reported by those to temps in unRAID dashboard, dashboard seems about 30 minutes behind. I'm wondering why that is and if it's necessary. Neither the plugin nor docker wake my sleeping drives.

 

Below is the drive temp code from autofan (which uses hdparm and smartctl.)

Note: I've heavily customized autofan so I'm not sure what parts of this code are the old version, which my customization was based on, and what parts I tweaked. I notice the current version doesn't pass --nocheck standby to smartctl (and maybe the old one didn't either.)

function_get_highest_hd_temp() {
  HIGHEST_TEMP_HDD=0
  HIGHEST_TEMP_HDD_LABEL=''
  for DISK in "${HD[@]}"; do
    SLEEPING=`hdparm -C ${DISK} | grep -c standby`
    if [[ $SLEEPING -eq 0 ]]; then
      if [[ $DISK == /dev/nvme[0-9] ]]; then
        CURRENT_TEMP_HDD=$(smartctl -A $DISK | awk '$1=="Temperature:" {print $2;exit}')
      else
        CURRENT_TEMP_HDD=$(smartctl --nocheck standby -A $DISK | awk '$1==190||$1==194 {print $10;exit}')
      fi
      if [[ $HIGHEST_TEMP_HDD -le $CURRENT_TEMP_HDD ]]; then
        HIGHEST_TEMP_HDD=$CURRENT_TEMP_HDD
        HIGHEST_TEMP_HDD_LABEL=$DISK
      fi
    fi
  done
}

 

[JorgeB]

10 minutes ago, CS01-HS said:

dashboard seems about 30 minutes behind

You can change that on Settings -> Disk Settings -> Tunable (poll_attributes):

 

Default is 1800s (30 minutes)

[CS01-HS]

14 minutes ago, JorgeB said:

You can change that on Settings -> Disk Settings -> Tunable (poll_attributes):

 

Default is 1800s (30 minutes)

Great, thanks!

 

EDIT: I found this detailed explanation for the 30 minute default. "Pretty large disruption in I/O flow" makes me nervous but I've run the 5 minute check for months now and haven't noticed any performance issues on my relatively weak system. Hmm.

 

https://forums.unraid.net/bug-reports/prereleases/unraid-os-version-690-beta25-available-r990/page/2/?tab=comments#comment-9930

 

Edited September 8, 2020 by CS01-HS

[JorgeB]

18 minutes ago, CS01-HS said:

EDIT: I found this detailed explanation for the 30 minute default. "Pretty large disruption in I/O flow" makes me nervous but I've run the 5 minute check for months now and haven't noticed any performance issues on my relatively weak system. Hmm

Yes, 5 minutes should be fine, it's what I use also for a long time.