[nsp] "ignore" on PA-FE-TX, 12.2(18)S4

[Scott Madley]

Hello,

Ignores are incremented on a VIP-PA when there are insufficient particle
buffers in the global pool to replace the interface rx ring buffers.
Ignores are also incremented when a packet arrives on an interface and there
a no available MEMD buffers in an interface pool.

One solution is to configure all interfaces on your router for the same MTU,
as this creates one large MEMD pool for packet buffering vs the having the
MEMD buffer carving algorithm do it's thing slicing up the available
SRAM/SDRAM.  This however, is not always the most desirable solution. :)

Hit me off-list if you desire more elaboration.

Cheers!
-SM





-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net]On Behalf Of Sebastian Abt
Sent: Monday, June 14, 2004 5:05 PM
To: lee.e.rian at census.gov
Cc: Gert Doering; cisco-nsp at puck.nether.net
Subject: Re: [nsp] "ignore" on PA-FE-TX, 12.2(18)S4


* lee.e.rian at census.gov wrote:
> My impression is that ignores are caused by an output interface queue
> being full.

Afaik, this happens/can happen when forwarding between high-speed and
low-speed interfaces.  Your higher-speed input interface's buffer pool
can be filled and tied up waiting for transmission on your lower-speed
interface.

regards,
sebastian

--
whois: sabt-ripe - phone: +49 (0)160 90759764
email: sabt at sabt.net - pgp-pubkey: 0xD008DA9C
_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/