Best practice for production servers: To reboot or not to reboot?

[Jerry Feldman]

On 09/15/2010 09:57 AM, Richard Pieri wrote:
> On Sep 15, 2010, at 1:44 AM, Derek Martin wrote:
>  =20
>> ...where he was clearly talking about virtual memory, and then you
>>    =20
> No, actually, it was not clear, given the context of memory fragmentati=
on and applications that suffer for it.  If an application is paging to d=
isk then you have far, far worse problems to deal with than memory fragme=
ntation.
>  =20
This is true.Memory fragmentation in modern Linux and Unix systems is
generally not an issue except in a small number of specific cases, as
the LWN article points out. In my travels, I have rarely seen an
application that wired memory (eg. for those that may not know, wired
means that the affected memory is locked in RAM and not pageable). Some
of the kernel is wired. Additionally, buffers that are used for I/O need
to locked in RAM until the I/O is complete, especially DMA.

--=20
Jerry Feldman <gaf-mNDKBlG2WHs at public.gmane.org>
Boston Linux and Unix
PGP key id: 537C5846
PGP Key fingerprint: 3D1B 8377 A3C0 A5F2 ECBB  CA3B 4607 4319 537C 5846