me too-> RE: 2.2.0 proc stuck in __down (D state)

David Mansfield (david@cobite.com)
Tue, 26 Jan 1999 10:48:42 -0500 (EST)


I got this too running a similar load test. My system is Intel PPro
200(UP) w/96MB ram. Kernel 2.2.0. Gcc 2.7.2.3. My scenario:

I had 10 processes chewing on memory (10mb each) and a

make "MAKE=make -j" zImage

This caused a bit of a slowdown :-) When the smoke cleared they had all
exited with signal 7 or 9, dmesg showed a bunch of 'Out of memory for
process xxx' and top was stuck in __down_interruptible according to ps
-alx.

Note, during the heavy swapping period a few oddities arose: for about
ten seconds in the middle, ALL disk I/O froze. Then the kernel climbed
back on the horse and went on with it's business. Secondly, as soon as
heavy swapping/context switching starts all process information becomes
unattainable through ps, top, or vmstat. Additionally, at a certain point
when OOM was reached, ALL the memory eaters were killed. This may have
been a side effect of the 'fork bomb' make -j though, which was entirely
unsuccessful at building my kernel.

David

-- 
/==============================\
| David Mansfield              |
| david@cobite.com             |
\==============================/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/