input queue drops on vlan cisco 6500

[Zeeshan Siddiqui]

Experts can you please help with the following I can  see drops on one of our busy L3 vlan in the input queue and are going up very frequently

System image file is "sup-bootflash:s72033-psv-mz.122-18.SXD3.bin

Hardware  =  6509

Input queue: 9/75/198/0 (size/max/drops/flushes); Total output drops: 0

  Queueing strategy: fifo

  Output queue: 0/40 (size/max)

  5 minute input rate 18217000 bits/sec, 9700 packets/sec

  5 minute output rate 138055000 bits/sec, 14592 packets/sec

  L2 Switched: ucast: 357314 pkt, 437562133 bytes - mcast: 1489 pkt, 152397 bytes

  L3 in Switched: ucast: 495853 pkt, 109667501 bytes - mcast: 0 pkt, 0 bytes mcast

  L3 out Switched: ucast: 728332 pkt, 883364068 bytes mcast: 0 pkt, 0 bytes

     541665 packets input, 119252541 bytes, 0 no buffer

     Received 1534 broadcasts (19542 IP multicast)

     0 runts, 0 giants, 3 throttles

     0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

     742857 packets output, 894351706 bytes, 0 underruns

     0 output errors, 0 interface resets

     0 output buffer failures, 0 output buffers swapped out

Any help is much appreciated is our L3 engine on the 6509 not able to handle that amount of traffic

[skarthic]

When your input queue is getting filled, this means that the packets are getting punted to the RP/SP CPU which is not the optimal way of packet forwarding on the Cat6ks.Ideally, the PFC should handle all the packets.

So we will have to identify why are the packets getting sent to the CPU.One useful tool to identify this is the Netdr capture. Please look at the following document the explains how this is done.

https://supportforums.cisco.com/docs/DOC-15608

From the Netdr captured packets, try to identify a pattern in the type of packets getting sent to the CPU. That should give us some clue on why we are seeing drops.

Meanwhile, also check for 'show process cpu' on both RP and SP to see if get anything interesting.

Thanks.