Kurt Roeckx wrote:
> It's running linux-image-2.6.32-5-amd64 2.6.32-45
>
> Note that this has nothing to do with vserver, just that people
> seeing it and reported it so far happened to run that version.
Hm. "page allocation failure" usually means that (some particular
kind of) memory has been exhausted, so a lot of different bugs could
produce that same symptom.
So I am worried about lumping all these cases together.
> Attached are 2 dmesg logs from homer.
Thanks. If I remember correctly, you set
vm.min_free_kbytes = 65536
and the problems went away.
The page http://www.acc.umu.se/~maswan/linux-netperf.txt you
mentioned says:
It might also be a good idea to increase vm/min_free_kbytes,
especially if you have e1000 with NAPI or similar. A sensible
value is 16M or 64M: vm/min_free_kbytes = 65536
However, you are using r8169 here, not e1000. Any leads towards
finding the resource hog, source of fragmentation, or other cause
would be welcome. We need to be setting the sysctl appropriately
automatically.
Hope that helps,
Jonathan
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120710194425.GG8439@burratino
