[Linux-cluster] GFS Crash

[Velu Erwan]

Chmouel Boudjnah a écrit :

>Hi,
>  
>
Hey chmouel,
Nice to see you ;)

>Using GFS on RHEL4 (kernel 2.6.9-22.0.2.ELsmp) i got a crash with
>it. 
>
>Before in the dmesg, there is some OOM-Killer activity with httpd.
>It seems to get something to do with the HIGHMEM and the 4GB of RAM.
>  
>
I don't know if its related directly with your troubles but I saw this 
in the 2.6.11 changelog:

[PATCH] vmscan: count writeback pages in nr_scanned
	
	OOM kills have been observed with 70% of the pages in lowmem being in the
	writeback state.  If we count those pages in sc->nr_scanned, the VM should
	throttle and wait for IO completion, instead of OOM killing.
	
	(akpm: this is how the code was designed to work - we broke it six months
	ago).

<andrea at suse.de>
	[PATCH] mm: adjust dirty threshold for lowmem-only mappings
	
	With Rik van Riel <riel at redhat.com>
	
	Simply running "dd if=/dev/zero of=/dev/hd<one you can miss>" can cause
	excessive amounts of dirty lowmem pagecache, due to the blockdev's
	non-highmem page allocation strategy.
	
	This patch effectively lowers the dirty limit for mappings which cannot be
	cached in highmem, counting the dirty limit as a percentage of lowmem
	instead.  This should prevent heavy block device writers from pushing the
	VM over the edge and triggering OOM kills.
	
	Signed-off-by: Rik van Riel <riel at redhat.com>
	Acked-by: Andrea Arcangeli <andrea at suse.de>
	Signed-off-by: Andrew Morton <akpm at osdl.org>
	Signed-off-by: Linus Torvalds <torvalds at osdl.org>


Erwan,