FreeBSD 10.1 crashing under load

[Christian Kujau]

Hi,

I have a Lenovo Ideapad S10 netbook running FreeBSD 10.1-p5 and 
ever since I switched from Linux Fedora to FreeBSD the system crashes 
regularly when the system is under a bit of load :-\

This netbook is operated in a headless mode, so I'm not using Wifi, 
Bluetooth or other fancy hardware features this netbook might have, only 
the system's disk and ethernet.

I've put a few more details below, but what I wanted to ask is: are there 
any flags to set in FreeBSD so that it's more chatty when it crashes? 
Right now the system just reboots but leaves nothing in the system log as 
to _why_ it crashed. I've configured a crash dump device as per [0] but no 
crash dump is ever taken. I'm running with encrypted swap (via GELI), 
maybe crashdumping doesn't work with encrypted swap?

I tried to set ddb_enable=YES in rc.conf but the standard kernel doesn't 
appear to have debugging enabled:

# /etc/rc.d/ddb start
/etc/rc.d/ddb: WARNING: failed precmd routine for ddb

# sysctl debug.ddb.scripting.scripts
sysctl: unknown oid 'debug.ddb.scripting.scripts': No such file or directory

I don't have a serial console for this machine.

Does anybody have any hints to to debug this thing? I haven't touched any 
boot parameters yet, anybody got a hint which knob to turn there?

Thanks,
Christian.

[0] https://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug.html

$ kldstat 
Id Refs Address    Size     Name
 1   28 0xc0400000 13a367c  kernel
 2    1 0xc606c000 12000    geom_eli.ko
 3    1 0xc607f000 22000    crypto.ko
 4    1 0xc5f7b000 4000     fdescfs.ko
 5    1 0xc628a000 8000     linprocfs.ko
 6    2 0xc629e000 4a000    linux.ko
 7    1 0xc6298000 3000     linsysfs.ko
 8    1 0xc61b9000 2000     blank_saver.ko
 9    1 0xc68c4000 5000     nullfs.ko

The only "exotic" part here is the enabled Linux compatibility: I'm 
running a (headless) sysutils/linux-crashplan installation on this system 
but for some reason it's not running with a native FreeBSD JRE but needs 
java/linux-sun-jre17 to work.

When running, this spawns some ~40 java processes that can be quite
active on both CPU and network & disk I/O. It uses a lot of memory too
and the machine does only have 1GB of physical memory (+1GB swap) but even
if the application would use too much memory I'd expect the application to
crash, not the whole system. The following is from "top -b" right before
the machine crashed:

===================================================================
97 processes:  1 running, 96 sleeping
Mem: 57M Active, 630M Inact, 281M Wired, 3104K Cache, 176M Buf, 6808K Free
Swap: 1024M Total, 92M Used, 932M Free, 8% Inuse

  PID USERNAME    THR PRI NICE   SIZE    RES STATE   C   TIME    WCPU 
COMMAND
91068 crashplan     1  52   19  1179M 69228K biord   0   0:21  36.18% java
24505 crashplan     1  40   19  1179M 69228K futex   1  11:13   0.78% java
24509 crashplan     1  40   19  1179M 69228K futex   0   0:10   0.49% java
40623 crashplan     1  39   19  1179M 69228K futex   1 179:40   0.00% java
 5750 chrisk        1  20    0 15144K  2744K select  0  39:18   0.00% screen
 5764 root          1  52    0 11048K  1276K wait    0  35:42   0.00% bash
===================================================================



-- 
BOFH excuse #236:

Fanout dropping voltage too much, try cutting some of those little traces