This is an incomplete issue, I’m asking for help to further troubleshoot the problem.

Recently my server has been having some intermittent issues where the dashboard reports 50% cpu utilisation, and all docker containers become unresponsive , even the shutdown commands do too.

The only temporary method to resolve is to hard shutdown it down. Sometimes this will reoccur within a day, but sometime longer, I had a steak recently of 2 months, it fooled me into thinking all my problems were over.

When sshd into the server, top reports normal utilisation, showing no major processes, with total utilisation no where near 50%. The kernel log also shows no messages related to this.

Where can I look next to debug?

Please post the diagnostics

@JorgeB I presume that will be most useful during an event, as I restarted it since last time, I shall respond next once it reoccurs.

An event finally reoccured, trying to run diagnostics during it hung for a while (maybe I'm not patient enough?), and no information seems present in the syslog

About 22:20 is when I noticed, and restarted the server
```

Sep 5 19:47:01 Tower crond[2013]: failed parsing crontab for user root: Invalid frequency setting of /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php >/dev/null 2>&1

Sep 5 20:07:01 Tower crond[2013]: failed parsing crontab for user root: Invalid frequency setting of /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php >/dev/null 2>&1

Sep 5 20:20:32 Tower bluetoothd[7554]: src/device.c:set_wake_allowed_complete() Set device flags return status: Invalid Parameters

Sep 5 20:20:32 Tower kernel: Bluetooth: hci2: No such LE device 70:91:f3:9e:38:ea (0x0)

Sep 5 20:28:01 Tower crond[2013]: failed parsing crontab for user root: Invalid frequency setting of /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php >/dev/null 2>&1

Sep 5 22:19:30 Tower rc.rsyslogd: Syslog server daemon... Started.

Sep 5 22:19:30 Tower root: Delaying execution of fix common problems scan for 10 minutes

Sep 5 22:19:30 Tower unassigned.devices: Mounting 'Auto Mount' Devices...

Sep 5 22:19:30 Tower emhttpd: Starting services...

Sep 5 22:19:30 Tower emhttpd: shcmd (70): /etc/rc.d/rc.samba reload

```

Edited September 6, 2025Sep 6 by hcookie
Selfdeprecation

I'm afraid that without anything relevant logged, there's not much to go on, one thing you can try is to boot the server in safe mode with all docker containers/VMs disabled, let it run as a basic NAS for a few days and see if it still happens, if it doesn't, start turning on the other services one by one, including the individual docker containers.

So further debugging during an event, revealed the true cause of this behaviour is a lock of /mnt/user. As the containers mounted that for appdata, they froze.

I tried moving them directly to /mnt/cache to prevent over use of the shfs, but that merely kept some containers alive when it happened again. What are some troubleshooting steps I could take? I have already updated to the latest release of unraid

Without anything logged, it won't be easy to debug, but you can still try what I mentioned above.

As I run a number of important containers on this system, I can’t quite do that. I’m looking at buying a second system atm. Here is what little is logged, https://pastebin.com/ThAe4HF3

Edited September 20, 2025Sep 20 by hcookie
Add log

Unfortunately, still nothing relevant that I can see.

I do believe it is the same issue as https://forums.unraid.net/topic/129333-unraid-fuse-dies-during-docker-container-operation-whole-os-gets-stuck/