I have an rsync that backs up my main server to a backup server and this used to work absolutely fine but it seems that recently it's started making the 2nd server reboot as it seems to max out 1 of the CPU cores and makes it start overheating.

 

 

You can see how it oddly spreads the load across the cores. It appears that it's SSHD that causes the large amount of CPU usage.

 

Sometimes the transfer rates drops from about 110meg/sec to much much lower and the CPU usage drops right down and then subsequently the temps and everything is fine but obviously backup then takes a crazy amount of time. I have the turbo write plugin enabled and set to automatic but doesn't seem to make much difference.

 

Anyone got any ideas as to why this is happening? The 2nd server is a core i5 but it's literally only used as a copy of my main server.

Post diags but it the CPU overheats when maxing out a core there are cooling issues, clean the cooler and check the thermal paste.

The server reboots so can't get diags. Nothing has physically changed on the hardware and I used to run this script daily without fuss so I feel like something on unraid itself has changed.

Again if the CPU is overheating it's a cooling issue, it should not overheat even if all cores were at 100%, but it might not even be that.

I'll try applying some new thermal paste to it, I don't know if atmospheric conditions can affect it as my backup server is in my loft (attic).

You can also enable the syslog server and post that after it crashes, it should show if it's an overheating problem.

Cool thanks, I've set it up to log to my main server so I'll be able to collect the logs to see what it says.

It seems that reapplying new thermal paste has made zero difference. There's nothing wrong with the fan on the CPU either so it's a very odd problem. Will hopefully get something in the logs.