So, after the yesterday experiences – Using more SDRAM to system and less in GPU – the kernel panic still happens.
Had seen a lot of people talking about heavy processes, and i think it must be connected, because system running idle hasn’t any problems.
So, as for today i’ve tryed a different approach: after reading some things about the interrupts in excess for such a machine – about 8000 getted with vmstat – and some saying it’s normal, and it’s related to the NIC driver from Broadcom which is using about 20% of resources, and while a new driver does’t appear, i’ve tryed a firmware update.
So, we’ve got a new kernel as i can see, and possible some more things. The upgrade firmware is quite easy to do, just follow the instructions where: https://github.com/Hexxeh/rpi-update
So at this point:
[email protected]:~$ vmstat
procs ———–memory———- —swap– —–io—- -system– —-cpu—-
r b swpd free buff cache si so bi bo in cs us sy id wa
0 0 0 90272 40652 44264 0 0 0 0 402 142 17 3 79 1
[email protected]:~$ uname -a
Linux raspberrypi 3.1.9+ #101 PREEMPT Mon Jun 4 17:19:44 BST 2012 armv6l GNU/Linux
Until now system looks stable, in this 3 hours… let’s see if it handles our stuff.
We’ve some heavy processes running using about 30% CPU, and since they’re threaded i think this could be the potential problem with the kernel panic interrupt fatality, but it’s kind of a wild guess…
top – 13:47:12 up 3:11, 1 user, load average: 0.28, 0.25, 0.27
Tasks: 63 total, 1 running, 62 sleeping, 0 stopped, 0 zombie
Cpu(s): 19.8%us, 3.3%sy, 0.0%ni, 74.6%id, 1.0%wa, 0.0%hi, 1.3%si, 0.0%st
Mem: 223048k total, 132824k used, 90224k free, 40692k buffers
Swap: 0k total, 0k used, 0k free, 44264k cached
Will update these random thoughts with more news when some stability is achieved