[c-nsp] 6704-10GE huge input drops (flushes)

[Antonio Soares]

Somewhere in the past the queue was increased due to the same type of symptoms it seems. But the impact this time is worse (bgp/isis sessions going down).


Regards,

Antonio Soares, CCIE #18473 (R&S/SP)
amsoares at netcabo.pt
http://www.ccie18473.net


-----Original Message-----
From: Peter Rathlev [mailto:peter at rathlev.dk] 
Sent: terça-feira, 7 de Maio de 2013 06:22
To: Antonio Soares
Cc: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] 6704-10GE huge input drops (flushes)

On Mon, 2013-05-06 at 23:59 +0100, Antonio Soares wrote:
> Before:
> ++++++++++++++++++++++++++++++++
> cat6k#sh int te1/1 | inc drops
>   Input queue: 1/2000/10310609/10310609 (size/max/drops/flushes); 
> Total output drops: 0 cat6k#
> ++++++++++++++++++++++++++++++++
> 
> After:
> ++++++++++++++++++++++++++++++++
> cat6k#sh int te1/1 | inc drops
>   Input queue: 0/2000/15863293/15863293 (size/max/drops/flushes); 
> Total output drops: 0 cat6k#
> ++++++++++++++++++++++++++++++++
> 
> This is a L3 interface is connected directly to a 12K (sip-601+spa-10ge).

Hm... the queue size of 2000 is default for a switchport where 75 is default for a L3 port. Did you just happen to increase it with "hold-queue 2000 in"?

The drops might not just be the port. Traffic directed at the CPU might get dropped and will (AFAIK) be counted towards drops on the input port.

--
Peter