Re: [PATCH] Add source address to sunrpc svc errors

From: Valdis . Kletnieks
Date: Wed Aug 29 2007 - 14:28:42 EST


On Tue, 28 Aug 2007 15:19:07 EDT, "J. Bruce Fields" said:
> On Tue, Aug 28, 2007 at 03:12:26PM -0400, Valdis.Kletnieks@xxxxxx wrote:
> > On Mon, 27 Aug 2007 17:43:33 EDT, "J. Bruce Fields" said:
> > > I also wonder whether these shouldn't all be dprintk's instead of
> > > printk's. One misbehaving client could create a lot of noise in the
> > > logs.
> >
> > I shouldn't have to rebuild my kernel with debugging enabled just to see
> > who is throwing trash at my machine.
>
> Fair enough.
>
> The dprintk's throughout the nfs and sunrpc server and client code can
> be selectively enabled at runtime using a set of sysctls in
> sys/sunrpc/*_debug; see
>
> include/linux/lockd/debug.h
> include/linux/nfsd/debug.h

I looked at nfsd/debug.h, and saw the NFSD_DEBUG usage, and thought "Wow, if I
built it without RPC_DEBUG, I'm screwed". I didn't see where RPC_DEBUG was
enabled if CONFIG_SYSCTL was defined - having "compile in debugging code"
equated to "system has sysctl support" was as unexpected as the guy today who
didn't realize you needed SCSI support for an IDE CD/ROM. ;)

"You are trapped in a twisty maze of little #ifdef's, all different"... ;)

Attachment: pgp00000.pgp
Description: PGP signature