Re: Call Trace: page allocation failure - is it normal behaviour?
From: Andrew Morton (akpm_at_osdl.org)
Date: 02/04/04
- Previous message: David Weinehall: "Re: 2.6.2-rc2 nfsd+xfs spins in i_size_read()"
- In reply to: Alexander Y. Fomichev: "Call Trace: page allocation failure - is it normal behaviour?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 3 Feb 2004 16:10:55 -0800 To: "Alexander Y. Fomichev" <gluk@php4.ru>
"Alexander Y. Fomichev" <gluk@php4.ru> wrote:
>
> Hello,
>
> I noticed some call trace when testing box under heavy load.
> To create a load following jobs have been running simultaneously.
>
> ab2 -c 200 -n 10000000 http://192.168.114.239/
> fsx-linux -l 900000000 fsx-data3
> dbench 100
>
> adt root # w
> 19:24:32 up 14:58, 6 users, load average: 90.92, 83.97, 84.28
>
> Some times after dmesg has shown multiple call traces of two types:
>
> swapper: page allocation failure. order:2, mode:0x20
> Call Trace:
> [<c014059c>] __alloc_pages+0x30c/0x350
> [<c0140605>] __get_free_pages+0x25/0x40
> [<c01435a7>] cache_grow+0xc7/0x310
> [<c01438fe>] cache_alloc_refill+0x10e/0x2c0
> [<c0143e01>] __kmalloc+0x71/0x80
> [<c0266697>] alloc_skb+0x47/0xe0
> [<c0294e7e>] tcp_fragment+0x5e/0x340
> [<c02975b8>] tcp_write_wakeup+0xe8/0x280
> [<c0298870>] tcp_write_timer+0x0/0x130
> [<c029776d>] tcp_send_probe0+0x1d/0x110
> [<c0298933>] tcp_write_timer+0xc3/0x130
> [<c01298f7>] run_timer_softirq+0xe7/0x1d0
> [<c0124e2a>] do_softirq+0xca/0xd0
>
> ...
> bwt I've noticed no visible harm to system and question ruther is
> whether this behaviour is normal under such circumstances?
Yes, it is expected and the networking stack will recover. We'll remove
that debug code at some point.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
- Previous message: David Weinehall: "Re: 2.6.2-rc2 nfsd+xfs spins in i_size_read()"
- In reply to: Alexander Y. Fomichev: "Call Trace: page allocation failure - is it normal behaviour?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
