HDD doesn't hibernate / TCRP 0.9.3.0 / DSM 7.1.1-42962 update 2 / HP microserver gen8

[mandgeo]

I sure think i've posted my configuration on title.

 

The problem ...
scemd generates alot of errors, that are written on log file. File located on DSM partition. As far as I understand, DSM is replicated on every drive on system, so my hard drives doesn't get the chance to sleep.

 

Quote

Dec 01 09:54:13 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:54:23 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:54:33 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:54:43 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:54:53 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:55:03 SyNAS scemd[14480]: oob/oob_reg_read.c:46 Read OOB reg 316 failed, ret=-1
Dec 01 09:55:03 SyNAS scemd[14480]: polling_hdd_backplane.c:39 ioctl device failed

 

There is also docker who writes something at regular interval, but it's a baby compared to scemd.

 

What have I done so far? I've managed to stop generating the scemd log file, by editing /etc.defaults/syslog-ng/patterndb.d/scemd.conf. The problem is that the messges are also written in /var/log/messages file. 

 

I've tried to create a filter, so that error messages could not be generated, but with no effect. 

What's the approach should i take in this case? Where are this messages coming from? I saw DSM 6 users complaining about this problem.

I've also tried to play with TCRP  user_config.json. I've changed maxdrives and internal sata parameters (maybe not the exact names), with no effect. 

 

Thank you!

[I am Don Hoolio]

Hello,

 

Did you, or anyone else, find a solution?

[mandgeo]

No solution. I've managed to stop loging oob errors in message, but drives doesn't sleep.

 

Test made today. Hibernation set to 10 minutes. iostat from 60 to 60 seconds and believe me that i have mode than 10 readings showing zero activity on sda, sdc &sdd. The result is active status, zero hibernation. If I remember well, sdc&sdd are part of raid 1 array.
External hard drives are showing stanby.

So, after this test, I can only conclude that the problem is not related with logs.

[scoobdriver]

On 12/7/2022 at 1:55 PM, mandgeo said:

No solution. I've managed to stop loging oob errors in message, but drives doesn't sleep.

Would you mind sharing how, also have a GEN8 and getting this error message in the log . 

 

Regards

[Keli]

Same here.. I can't seem to get the drives to hibernate and by extension the whole system via the 'auto poweroff' feature.

Things I've tried so far:

• Editing the file '/etc.defaults/syslog-ng/patterndb.d/scemd.conf' so that '/var/log/scemd.log points to /dev/null'
• Using  hdparm -Y /dev/sdx to put the drives on standby mode - they quickly resume, even drives not used by DSM (Uninitialized)
• Tried this supposed solution - it didn't work 
• Wiped the drive clean and reinstalled DSM from scratch then tested with only the default apps installed

Specs of my system:

Motherboard: Gigabyte GA-H61M-DS2

Processor: i5-3470

Drive: Hitachi HTS541680J9SA00

Loader: tinycore-redpill-uefi.v0.9.4.0

Platform: DS3622xs+

DSM Version: DSM 7.1.1-42962 Update 3 (Manually updated from Update 1 via ./rploader.sh postupdate)

 

Something else I tried.. I ran 'cat /sys/power/state' which returned 'disk' meaning that system hibernation is supported. Naturally I ran 'echo disk > /sys/power/state' as root which put the system into hibernation, so it works! I was able to wake it up with a magic packet with a notification popping up that the NAS was shutdown improperly. 

My scemd.log is full of these errors which get logged every few seconds as from the screenshot below.