FreeNAS server randomly restarts (usually when watching something on plex) - Log says nothing

[DrKK]

Random reboots that cannot be chalked up to PSU, thermal issues, or hardware incompatibilities/driver problems, are almost always memory related, in my experience.

I was going to say the 3V was more than 10% out of whack, but the 3V rail is actually a 3.3V rail, I believe, in which case that seems spot on.

So...yeah, you got me. Test the memory.

 

[Ericloewe]

the 3V rail is actually a 3.3V rail, I believe, in which case that seems spot on.

Yup, that IPMI team was lazy.

 

[Supa]

Random reboots that cannot be chalked up to PSU, thermal issues, or hardware incompatibilities/driver problems, are almost always memory related, in my experience.

I was going to say the 3V was more than 10% out of whack, but the 3V rail is actually a 3.3V rail, I believe, in which case that seems spot on.

So...yeah, you got me. Test the memory.

6 passes with memtest so far. No errors :3

 

[DrKK]

This has to be annoying as hell Supa for you. I feel your pain. I hope we can figure this out. Have you reseated the CPU? Have you reseated/replaced all SATA power cables and data cables? We're starting to get into the "pretty unlikely" zone here.

 

[DrKK]

Also, have you tried a new power supply, as we discussed at the beginning? Power supplies are by far the most commonly breaking pieces of equipment---and fortunately, they are also inexpensive.

Just because the voltages seem good in IPMI at particular times does not mean that under load, or under other conditions, the rails stay up at the nominal voltage, and/or the quality of power delivered is sufficient. You cannot tell if a power supply is good or bad based upon IPMI readings. You can really only tell by swapping it out, and seeing if your problem goes away.

 

[Supa]

Haven't messed with any of that.. but I'll try.

What makes it hard to debug is it seems like it reboots randomly.. so it's hard to make a change and then test it.

I might just submit an RMA though with Seasonic and see if a replacement helps.


I reseated the CPU, reseated all the PSU/Sata cables, used some new SATA cables.... Did some cable management... so we'll see how it goes....

 

[invent0r]

hi, do you have any update on this post?
Im currently having exactly the same issue and my hardware is not even 1 year old...

 

[Supa]

Honestly I kid you not.. I took out everything, did some cable management, re-inserted all my cables... and it's been good ever since.. So I'm still clueless.

 

[invent0r]

Damn gonna try the same thing when i find the time

 

[invent0r]

@Supa I fixed it, seems like my hard drives were running hot and resulting in a shutdown, i disabled my cloud storage and i haven't had issues since, now on to finding what resulted in the heat increase.

 

[Supa]

@Supa I fixed it, seems like my hard drives were running hot and resulting in a shutdown, i disabled my cloud storage and i haven't had issues since, now on to finding what resulted in the heat increase.

Yeah I don't even use cloud storage. How do you know they're hot?

Sent from my SM-G935T using Tapatalk

 

[Mirfster]

Could always query SMART Data...
Example for disk "da10": smartctl -a /dev/da10 | grep "Temperature"

Also, if you properly configured the "S.M.A.R.T Service"; you should have set an "Informational" and/or "Critical" value that would send you an e-mail notifying you that your drives are getting hot.