Mine isn't being fixed by changing that line. Any Suggestions?
How many minutes do you want between spin checks? 15 Friday, Aug 03 pass0 pass1 pass2 pass3 Tmax Tmean 08:15:51 Spin 9) ./spincheck.sh: line 85: let: Tsum=0+9): syntax error in expression (error token is ")") Spin 1) ./spincheck.sh: line 85: let: Tsum=9+1): syntax error in expression (error token is ")") UNKNOWN UNKNOWN ./spincheck.sh: line 42: printf: 9): invalid number ^9 5.00 ERRc= -28.57
@twsps, I wrote that over 2 years ago, and I guess you're using an old version. Please use the up-to-date version of the spincheck script from here: https://forums.freenas.org/index.php?resources/fan-scripts-for-supermicro-boards-using-pid-logic.24/
It's very likely you have something running the accesses the hard drives periodically. It's really not hard to force the drives to spin down but it takes a lot on learning how to command these things to spin down and then paying close attention as to what applications are being run which access the hard drives. For example, do you have any VMs or jails or plugins running as these are key things that cause the drives to be accessed and thus not sleep. You say you have nothing configured so you should be able to send the command manually to sleep the drives immediately. Listen for them to stop spinning. Then wait to see if they spin back up. If they do then you have some application which is using the drives/pool some way. Unfortunately I'm not familiar enough with FreeNAS 11.2 to provide further advice.
and after reboot/update.
First of all, good job in doing your research and running the tests. To make this persistent you could easily write a script (just like you would from the command line) into a text file and then execute that file via CRONTAB (CRON job). You can select if you would like to run it once or maybe even once a day. So once you have this all up and running, pay attention to your head loading count and the spinup count, you could inadvertently cause this to increase significantly which means the drives are cycling up and down too frequently which will likely cause a drive failure. In my opinion cycling more than once a hour is way too much but it's up to you to figure out what you desire.
import subprocess
import time
drives_to_check = ['ada3', 'ada4', 'ada5']
no_sleep_hours = ['00', '01']
seconds = 600
if time.strftime("%H") in no_sleep_hours:
exit()
o = subprocess.check_output(f'/usr/sbin/iostat -x -z -d {seconds} 2', shell=True)
for drive in drives_to_check:
if drive not in o.decode().split('extended')[-1]:
p = subprocess.check_output(f'/sbin/camcontrol cmd {drive} -a "E5 00 00 00 00 00 00 00 00 00 00 00" -r -', shell=True)
if p.decode()[27:29] != '00':
q = subprocess.check_output(f'/usr/local/sbin/ataidle -s /dev/{drive}', shell=True)
*/15 * * * * /usr/local/bin/python3 /path/to/the/folder/check_disk_usage_and_spindown.py
It's been a while since I've looked at this, but when I did (OP) getting drives to spin down was no problem. As time went on, FreeNAS would spin up the disks at various times with no apparent reason. Running your script every 15 minutes may put them back to sleep, but if they're spun up and down multiple times a day, at some point it's better just to leave them spinning.
import subprocess
from datetime import datetime
datetime.now().strftime('%Y-%m-%d %H:%M:%S')
def main():
check()
def check():
drives = {'ada0':'', 'ada1':'', 'ada2':'', 'ada3':'', 'ada4':'', 'ada5':''}
state = []
for drive in drives:
try:
o = subprocess.check_output(f'/sbin/camcontrol cmd {drive} -a "E5 00 00 00 00 00 00 00 00 00 00 00" -r -', shell=True)
drives[drive] = o.decode()[27:29]
except:
drives[drive] = '??'
with open("/path/to/folder/drive_stats.txt", "a") as myfile:
myfile.write(f"{datetime.now().strftime('%Y-%m-%d %H:%M:%S')} {drives}\n")
print(datetime.now().strftime('%Y-%m-%d %H:%M:%S'), f"{drives}")
if __name__ == '__main__':
main()
