Unscheduled reboot, /data/crash/info points to one of the disks and to Plex Media Server?

[GLaDER]

It might be the case that I have already found the culprit (a Jail running Plex Media Server), in which case I would greatly appreciate your support understanding why the jail miss behaved. Regardless, maybe this will also help someone else in their troubleshooting!

SYSTEM INFO
UPS: APC Back-UPS Pro 650 VA
Motherboard: SuperMicro A2SDI-4C-HLN4F-O
RAM: 16 GB (2x Samsung 8GB DDR4 ECC REG 2400Mhz x4 SR)
Boot Drive: Samsung 840 SSD
Pool: 4x Toshiba HDWG180 (8 TB)
OS: TrueNAS-12.0-U4

During the night I had an unscheduled reboot. At first I thought it was related to a power outage and a problem with my new UPS. Upon closer inspection it seems like the reboot might be related to an issue with one of my drives.

I could not find anything useful in messages (apart from the time the system came back online); but /data/crash/info.last contains information with the same timestamp as the reboot occurred.
Dump header from device: /dev/ada4p1
Architecture: amd64
Architecture Version: 4
Dump Length: 657408
Blocksize: 512
Compression: none
Dumptime: Sun Jul 18 03:27:17 2021
Hostname: freenas.local
Magic: FreeBSD Text Dump
Version String: FreeBSD 12.2-RELEASE-p6 df578562304(HEAD) TRUENAS
Panic String: page fault
Dump Parity: 33138809
Bounds: 1
Dump Status: good


I unpacked textdumt.tar.last.gz and looked at ddb.txt. The file is large... but there were plenty of references to Plex Script Host in the file.

Show : Truncated output from ddb.txt


[...]
db:0:kdb.enter.default> show allpcpu
Current CPU: 2

cpuid = 0
dynamic pcpu = 0x137b180
curthread = 0xfffff803533c3740: pid 44174 tid 100521 "Plex Script Host"
curpcb = 0xfffff803533c3ce0
fpcurthread = 0xfffff803533c3740: pid 44174 "Plex Script Host"
[...]

cpuid = 1
dynamic pcpu = 0xfffffe008d6fa180
curthread = 0xfffff801ea878740: pid 45395 tid 101603 "Plex Script Host"
curpcb = 0xfffff801ea878ce0
fpcurthread = 0xfffff801ea878740: pid 45395 "Plex Script Host"
[...]

cpuid = 2
dynamic pcpu = 0xfffffe008d707180
curthread = 0xfffff8003fc42000: pid 4283 tid 101146 "Plex Media Server"
curpcb = 0xfffff8003fc425a0
fpcurthread = 0xfffff8003fc42000: pid 4283 "Plex Media Server"
[...]


msgbuf.txt seems to confirm my suspicion:

Fatal trap 12: page fault while in kernel mode
cpuid = 2; apic id = 10
fault virtual address = 0x3454f8f9
fault code = supervisor read data, page not present
instruction pointer = 0x20:0xffffffff827e52ce
stack pointer = 0x28:0xfffffe013173a658
frame pointer = 0x28:0xfffffe013173a690
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 4283 (Plex Media Server)
trap number = 12
panic: page fault
cpuid = 2
time = 1626571637


I very seldom have unexpected reboots, but I want to migrate the PMS Jail to another system anyway, so I'll disable and remove the jail.

QUESTION:

1. Does anyone recognize this? Do you think PMS is the reason for the reboot?
   I have been running PMS on this server for a very long time and I have rarely have any unexpected reboots.
   
   
2. Are there more places where I should look for logs, to assist me in my detective work?
   
   
3. Since the crash/info.last file mentioned a drive, should I be concerned about the health of that disk? Or how does Dump header from device: play into the crash reports?

 

[Jager42]

I am dealing with this issue as well, have not really found any details for a cause yet but have had at least 5 crashes due to this. I am seeing very similar messages in my logs compared to the ones you posted. I am pretty sure this crash is caused when Plex is doing its maintenance tasks as the crash is always durring the timeframe I have set for the tasks.

I also want to link other posts talking about this:

Plex Media Scanner kernel panics at least once a day

Hi there, I'm currently running into the problem that my TrueNAS 12.0-U2.1 (see Signature) install is seemingly randomly rebooting. I already had a look at the debug logs and it seems to experience an page fault kernel panic almost every day. I already ran multiple burn in tests in Windows for...

www.truenas.com

Plex Server - Fatal trap 12: page fault while in kernel mode

This is the only logs from Plex right before the crash. This is from the file called “Plex Media Server.log”. All other log files did not have entries around the time of the crash. The crash happened at 06:34. I have my doubts that it’s a corrupted media file as it crashed two days in a row and...

forums.plex.tv

 

[GLaDER]

I am dealing with this issue as well, have not really found any details for a cause yet but have had at least 5 crashes due to this. I am seeing very similar messages in my logs compared to the ones you posted. I am pretty sure this crash is caused when Plex is doing its maintenance tasks as the crash is always durring the timeframe I have set for the tasks.

I also want to link other posts talking about this:

Plex Media Scanner kernel panics at least once a day

Hi there, I'm currently running into the problem that my TrueNAS 12.0-U2.1 (see Signature) install is seemingly randomly rebooting. I already had a look at the debug logs and it seems to experience an page fault kernel panic almost every day. I already ran multiple burn in tests in Windows for...

www.truenas.com

Plex Server - Fatal trap 12: page fault while in kernel mode

This is the only logs from Plex right before the crash. This is from the file called “Plex Media Server.log”. All other log files did not have entries around the time of the crash. The crash happened at 06:34. I have my doubts that it’s a corrupted media file as it crashed two days in a row and...

forums.plex.tv

How do you run PMS? I have a jail where I have installed PMS using pkg.

 

[Jager42]

How do you run PMS? I have a jail where I have installed PMS using pkg.

I have mine set up the same way, installed in a jail via PKG using the normal (non-plexpass) version.