If you scroll upper, there are written times, it varies from ours to days.
System will reboot after NMI and continues working, but
while rebooting data are not accessible
start/stop count is incrementing on disks - power lost for disks during reboot
auto VM run is working, but if all machines run in the same time and they are running “big tasks“, system gets some overload (I am running in one VM ubuntu with docker and some containers and there are some docker jobs consuming CPU) … when I run it manually, I don’t have this problem
containers may have problem too if all running parallel, I run them manually one by one (I have script with short sleep)
I think 3) and 4) maybe caused by too old HW (small performance). But reboots are very unpleasant.
The main problem is I was using TrueNAS Core on this HW "many” years without any issues and after migration to TrueNAS CE (Scale), it is not stable (“enough“).
I can install clean debian or clean freebsd and configure samba share, virtual machiness. But TrueNAS has nice dashboard, reports, audit logs, etc …
I think that it is not CPU problem. I suspect more SAS controller operations, but not sure. I have feeling if I run docker containers and don’t write “bigger” files to FS and don’t write “bigger“ files to storage, system is stable. I am not sure about exactly testing of this hypothesis. I also think that we have no solution. Solution is wait for 26.04 if new kernel fix it or don’t use TrueNAS CE.
I was thinking about to turn off hyperthreading, but not realised yet.
There won’t be a “26.04” version, it’ll be “26” (then “26.1”, “26.2”, etc.)
Joke aside, I’m note sure a new kernel would solve the issue since I can’t find any bugfix close to my issue (the bugreports I’ve found are still open).
I might give the 26 Beta 2 and/or 26 RC a try though.
Please, share info about 26RC. What is your backup plan? XigmaNAS? OMV? Pure FreeBSD and own config? Pure Linux install with own config? Proxmox with virtualised TrueNAS? I don’t want to change HW, I like consumption of this CPU and I want 9 HDD connected so I don’t want to change controller. Thank you
I’m running a very similar setup (Gen8, G2020T, 16 GB ECC RAM - tested, no errors) and I’m seeing the exact same random restarts.
The system is basically idle most of the time (just file storage). The behavior is identical on my side.
Same thing happens on 25.04 and 25.10 - I rolled back to 24.10 and it’s stable there, no restarts.
At first I thought it might be aging hardware, but after reading this thread and seeing similar reports, it really starts to look more like a software/kernel issue than a hardware problem.
