[smartmontools-support] smartd complains about failing to read temperature, but smartctl/smartd manually can

[Boether, Maximilian]

Dear Support-Team,

every 30 minutes, smartd complains that it failed to read the temperature of one HDD which is part of a RAID. This is what the logs show:
Mar  8 17:05:32 fb14srv2 smartd[1222]: Device: /dev/bus/0 [megaraid_disk_04], failed to read Temperature
Mar  8 17:35:31 fb14srv2 smartd[1222]: Device: /dev/bus/0 [megaraid_disk_04], failed to read Temperature

But when I check using smartd manually, I get a valid result:
smartd -q onecheck:
Device: /dev/bus/0 [megaraid_disk_04], opened SCSI device
Device: /dev/bus/0 [megaraid_disk_04], SMART health: passed
Device: /dev/bus/0 [megaraid_disk_04], initial Temperature is 35 Celsius (Min/Max 28/54)
    [trip Temperature is 68 Celsius]

Also using smartctl everything seems to be fine:
smartctl -a -d megaraid,4  /dev/sda2:
=== START OF READ SMART DATA SECTION ===
SMART Health Status: OK

Current Drive Temperature:     36 C
Drive Trip Temperature:        68 C

What can cause such an issue? This is really strange to me.

Thank you very much for your help.

Kind regards
Maximilian Böther