I had been running a version of v5 for ages without issues, and upgraded straight to v6.1.6 a couple weeks ago following the instructions to reformat the USB key. Since then the system has been 'freezing up' most nights: no GUI access, no share access, but it pings and I can SSH in and grab stuff via WinSCP (though some commands don't work). I upgraded to 6.1.8 yesterday, but it still froze up again this morning. I tried to capture diagnostics but the command did not fully complete; I've attached what it did generate. When it's frozen up like this, I can't run the powerdown command (just doesn't actually powerdown anything).
Specs:
Gigabyte GA-MA74GM-S2H
2x1GB 800Mhz DDR2
AMD Athlon 64 X2 Dual Core 6000+ 3.00GHz
4 data drives, 1 cache, 1 parity, all ReiserFS
SNMP plugin and the Powerdown plugin installed.
So far I have tried:
Disabling the cache drive on all disks
Disabling Docker
Checked SMART status of all disks (the quick test) and all look OK. One disk has some current pending sectors but this has not increased in a long time.
Last night I wrote a script that SSH's in and grabs the top 5 processes by CPU usage every 2 minutes to see if that would turn up anything useful (date; ps aux | sort -rk 3,3 | head -n 5). It shows this process grabbing a bit more CPU % over the course of the night:
/usr/local/sbin/shfs /mnt/user -disks 31 2048000000 -o noatime,big_writes,allow_other -o remember=0
Gonna be honest, don't know what that actually is indicating but it doesn't seem right. Attached the output as processes.txt.
Last night I believe the shares stopped working between about 10pm and 2:30am, but there's nothing in the syslog that shows anything funky happening.
I've attached some logs. The server is currently in it's half responsive state, so if anyone has any troubleshooting suggestions that I can run over SSH, or logs and files I can grab, I'll leave it like this for a few hours.
Thanks in advance!
towerzip.zip