Re: page allocation failed in 2.4.24

[Marcelo Tosatti <marcelo.tosatti _at_ xxxxxxxxxxxx>]

On Sun, 1 Feb 2004, Mike Gabriel wrote:

> hi there,
>
> can anyone tell me what these messages mean? running linux-2.4.24 on asus
> p4p800 deluxe.
>
> Jan 31 02:38:02 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0xf0/0)
> Jan 31 02:38:02 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0x1d2/0)
> Jan 31 02:38:02 galileo kernel: VM: killing process slapd
> Jan 31 02:38:50 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0xf0/0)
> Jan 31 02:38:50 galileo kernel: ENOMEM in journal_get_undo_access_R9add2900,
> retrying.
> Jan 31 02:38:53 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0xf0/0)
> Jan 31 02:39:19 galileo last message repeated 3 times
> Jan 31 02:39:25 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0x1d2/0)
> Jan 31 02:39:25 galileo kernel: VM: killing process sshd
> Jan 31 02:39:25 galileo kernel: __alloc_pages: 0-order allocation failed
> (gfp=0x1d2/0)
> Jan 31 02:39:25 galileo kernel: VM: killing process sshd
>
> I get plenty of them during copying (cp -av <source> <dest>) a huge amount of
> data, which is densely filled with hard links from one raid to another... the
> cp command runs in a screen session. on huge directories, it starts blasting
> other processes from the system (sshd, nmbd, spong-client, etc.)

Hi Mike,

How much swap do you have on this system and how much swap space is free
when you start seeing the allocation failures?

"vmstat 2" output (start it before running the tests) and "cat
/proc/slabinfo" output (before and during the tests) are useful to find
out what is happening.

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