Hello,

yes output drops in the 8 ports group is a clear sign of the oversubscription effect on WS-4548, but they don't count all of these drops.

It should be enough to demonstrate the problem but it doesn't provide a precise measure of how many frames are dropped.

For us it was enough to demonstrate to management that these linecards are not good in modern GE server farms

see also the following document:

http://www.cisco.com/en/US/products/hw/switches/ps700/products_tech_note09186a008015bfd6.shtml#oversub

>>If the frames received on any group exceeds the bandwidth of 1 Gbps,       the device starts to drop the frames. These drops are not obvious as they are       dropped at the internal ASIC rather than the actual interfaces. This can lead       to slow throughput of packets across the device.

The Rx-No-Pkt-Buff does not depend on the total traffic rate. It       depends on the amount of the packets that are stored in the Rx FIFO buffer of       the module ASIC. The size of this buffer is only 16 KB. It is counted with       short bursty traffic flows when some packets fill this buffer. Thus,       Rx-No-Pkt-Buff on each port can be counted when the total traffic rate of this       ASIC port group exceeds 1 Gbps, since WS-X4548-GB-RJ45 is 8:1 oversubscribed       module.

Hope to help

Giuseppe