Re: oom-killer why ?

[Marco Nietz <m.nietz-mm@iplabs.de>]

Christoph Lameter schrieb:

> Hmmm... That should be fairly stable. I wonder how prostgres handles the
> buffers? If the pages are mlocked and are required to be in lowmem then what
> you saw could be related to the postgres configuration.

Don't know it exactly, but will try to find it out. And yes, the Machine
was fairly stable until i raised up shared buffers.

> The problem is that the boot messages are cut off we cannot see the basic
> operating system configuration and the hardware that was detected.

Haven't got more Information than the ones if've posted, sorry-

Maybe this short Overview helps:

It's a Dell Poweredge 1950, Dual Quad Core with 2.66GHz and 16G Ram. The
Machine has two Raid-Controllers. One used for the OS and the other one
for a Direct Attached Storage (MD-3000). This Storage is controlled with
multipath-tools and used for Database Storage.


Best Regards
Marco



--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>