System crashed, "page fault while in kernel mode", current process=arc_reclaim_thread

[Arik Yavilevich]

Hi,

I have been benchmarking the system and it crashed. Am I right to assume the issue is with "arc_reclaim_thread"?

Code:

Fatal trap 12: page fault while in kernel mode
cpuid = 3; apic id = 03
fault virtual address   = 0x50
fault code     = supervisor read data, page not present
instruction pointer   = 0x20:0xffffffff80366ee4
stack pointer    = 0x28:0xfffffe023a4137a0
frame pointer    = 0x28:0xfffffe023a4137c0
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     = 7 (arc_reclaim_thread)

How would I go about understanding more about what happened?
Is there a way to configure the system to log the stack trace when a fault like this happens?

The machine is for testing, not production. It is a VM in ESXi. 4 vcpus and 8GB vram.

Thanks.

Code:

+--------------------------------------------------------------------------------+
+ FreeNAS-9.3-STABLE-201509160044 +
+--------------------------------------------------------------------------------+
Operating system type: FreeBSD
Operating system release: 9.3-RELEASE-p25
Operating system revision: 199506
Kernel version: FreeBSD 9.3-RELEASE-p25 #0 r281084+d3a5bf7: Tue Sep 15 17:52:04 PDT 2015
root@build3.ixsystems.com:/tank/home/jkh/build/FN/objs/os-base/amd64/tank/home/jkh/build/FN/FreeBSD/src/sys/FREENAS.amd64
Hostname: freenas.local
Name of kernel file booted: /boot/kernel/kernel

 

[dlavigne]

It is a VM in ESXi. 4 vcpus

Does it crash if you use 1 vcpu? From http://doc.freenas.org/9.3/freenas_install.html#vmware-esxi:

Under “CPUs”, make sure that only 1 virtual processor is listed, otherwise you will be unable to start any FreeNAS® services.

 

[Arik Yavilevich]

I would like to update that this crash was probably a result of a hardware issue.
After repeated cases of instability I was able to identify rare RAM issues. Only detectable with the new memtest86 v6.2. The CPU/board was not able to run memory at full speed in dual channel mode. Ended up downgrading to a slower profile in BIOS to resolve issues.
fyi.