On Mon, 25 Apr 2011 at 09:46, Dave Chinner wrote:
> BTW, what are your mount options? If it is the problem I suspect it
> is, then using noatime with stop it from occurring....
When mounted with noatime, running du(1) accross the XFS volume still
triggered the OOM killer, but the backtrace looked somewhat different
this time:
date invoked oom-killer: gfp_mask=0x84d0, order=0, oom_adj=0, oom_score_adj=0
Call Trace:
[ecda5c70] [c0009ce4] show_stack+0x70/0x1bc (unreliable)
[ecda5cb0] [c008f508] T.528+0x74/0x1cc
[ecda5d00] [c008f734] T.526+0xd4/0x2a0
[ecda5d40] [c008fb7c] out_of_memory+0x27c/0x360
[ecda5d90] [c0093b3c] __alloc_pages_nodemask+0x6f8/0x708
[ecda5e30] [c0017e9c] pte_alloc_one+0x30/0x90
[ecda5e40] [c00a9464] __pte_alloc+0x2c/0x100
[ecda5e60] [c00a95a8] handle_mm_fault+0x70/0x110
[ecda5e90] [c0016d78] do_page_fault+0x358/0x504
[ecda5f40] [c0012938] handle_page_fault+0xc/0x80
--- Exception: 401 at 0xffdd5e0
LR = 0xffd8a04
Full dmesg & slabinfo & more on:
http://nerdbynature.de/bits/2.6.39-rc4/oom/
(slabinfo-5.txt.bz2, messages-5.txt)
Christian.
--
BOFH excuse #176:
vapors from evaporating sticky-note adhesives
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
