See the attached diagnostics and syslog.

 

Every sunday in the early hours I have the parity check scheduled and until recently had the CA backup later on.

 

Every sunday morning I wake up to uptime robot notifications of my services being down. I visit the webui and it initially loads but buttons do nothing and the server is effectively frozen (I end up having to manually restart the server).

The server runs perfectly happy for a few days at a time but around sunday/monday something always goes wrong. I suspect that my drives (which are 7 years old) are causing the issue. If I can get confirmation that is the only issue that could be causing this I will invest in new drives, I just need to know beforehand.

 

I have checked my Netdata for metric correlations and find disk I/O to massively rise before the freeze, suggesting its the parity check that is doing this.

 

Please let me know if there's more info needed.

 

Thanks.

syslog-127.0.0.1(1).log thin-diagnostics-20241104-1026.zip

@JorgeB Your name was mentioned as someone who is quite knowledgeable when it comes to diagnosing logs. Any input is appreciated !

There's nothing relevant logged, though there are some disk errors, but a server crashing during a parity check is usually a hardware problem.

47 minutes ago, JorgeB said:

There's nothing relevant logged, though there are some disk errors, but a server crashing during a parity check is usually a hardware problem.

Ok, is there anything I can do to isolate it to ensure its a disk issue rather than anything else?

Thanks.

Disk errors are for disk2, you can run an extended SMART test, but unlikely that the server crashing is related to that, usual suspects are RAM, board, CPU or PSU, in no particular order.