On Sun, May 14, 2000 at 12:49:22PM -0700, Ben Greear wrote:
> Andi Kleen wrote:
>
> > You can increase the kernel buffers with the SO_RCVBUF/SO_SNDBUF setsockops
> > and the net/core/[rw]mem_{max,default} sysctls. For large values you
> > may also need to increase the vm/freepages settings (see the man pages
> > for more details)
>
> Which man pages would these be? Can this be done across the board through
> the /proc FS?
e.g. socket(7)
>
> > There are unfortunately not counters for all oom packet drop cases,
> > so some get silently dropped. It may also be dropped somewhere else
> > (like the switch).
>
> Yep, the af_packet.c file definately does not have counters for OOM
> on receive. Should be trivial to add them, but might be a pain to display
> them... Supposing I did want to add the counters & display them. Is
> that normally done through the proc FS, or just through IOCTLs?
2.3 af_packet has drop counters. There are just some other cases which are
not, or not consistently intrumented (generic net core, network drivers)
Ideally both: /proc for the usual cases and a ioctl or cmsg for high speed
packet users.
-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
