asr 1k QFP drops

[serge123]

Hi all,

Anyone know what could be causing these drops in the QFP:

They are incrementing

sh platform hardware qfp active statistics drop
-------------------------------------------------------------------------
Global Drop Stats                         Packets                  Octets
-------------------------------------------------------------------------
IpTtlExceeded                                 352                  110011
Ipv4Null0                                   26616                 2715082
Ipv4mcRpfFailed                              2332                  441011
TailDrop                                   705820               604363927   

Hardware is SPA-1X10GE-L-V2:

5 minute input rate 11538000 bits/sec, 7899 packets/sec
5 minute output rate 140081000 bits/sec, 24471 packets/sec

[Leo Laohoo]

What exactly is your RP and ESP card?  

[serge123]

ASR1000-RP1

ASR1000-ESP20

[Leo Laohoo]

Do you have QoS enabled?  

How often or regular are the counters incrementing?

[serge123]

Yes I do  have QOS but I do not see any drops in there

QFP Counters were clear yesterday

here is another snapshot

-------------------------------------------------------------------------

Global Drop Stats                         Packets                  Octets

-------------------------------------------------------------------------

Icmp                                           15                     990

IpTtlExceeded                                4662                 1322852

Ipv4Null0                                  896464                91686673

Ipv4mcRpfFailed                             21512                 4056681

TailDrop                                  7767449              4472261037

[Vishesh Verma]

Hi Serge,

Here is a good document to begin troubleshooting -

http://www.cisco.com/c/en/us/support/docs/routers/asr-1000-series-aggregation-services-routers/110531-asr-packet-drop.html

IpTtlExceeded increments when TTL exceeds

Ipv4Null0 packets that do not have a route and dropped to null0

Ipv4mcRpfFailed are mcast RPF failures

TailDrop highly likely to be dropped by ESP, if packet burst is exceeding wire rate and queue is full.

Please share the show interface and show policy-map interface outputs as well.

-Vishesh

[serge123]

Hi All,

I have attach Show int for LAN and WAN side before and after cleared counters.

Show policy map I have over 20 vrf on it but none of them are exceeding or dropping packets

[Leo Laohoo]

  Last clearing of "show interface" counters 00:24:01
  Input queue: 0/375/0/0 (size/max/drops/flushes); Total output drops: 73230

In 24 minutes, output drops are incrementing?  This is on your WAN link?  

This is the problem.  Whatever is the connected to this link is showing signs of not being able to cope with the traffic being pushed by this interface.

[serge123]

Thanks Leo. I wanted to make sure it was not the ESP.  I will have the wan device investigate and let you guy,s know the outcome. Thank you all

[Leo Laohoo]

Please let us know of the outcome of the investigation, Serge.  

Another thing, can you verify that you have a 10 Gbps WAN link.  Because if you have, for example, a 5 Gbps WAN link then you'll need to configure traffic shaping policy on your WAN interface.

[Leo Laohoo]

Can you post the output to the command "sh interface <BLAH>"?