What exactly do TX pause counters mean?

[clayhawks]

I have a pair of 5548UP Nexus switches connected to a 2232PP FEX. Some of the FEX ports are connected to HP Flex Fabric modules, and I'm seeing lots of "TX pause" frames. Can anyone explain exactly what this counter means? Is my FEX telling the connected device to pause their transmission, or does this mean that the connected device is sending pause frames to my FEX telling the FEX to pause?

wme-r7nex5k-1# sh int e102/1/1
Ethernet102/1/1 is up
  Hardware: 1000/10000 Ethernet, address: 5897.1ea9.34c2 (bia 5897.1ea9.34c2)
  Description: *** Connected to XXXXXXXXXXX ***
  MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec
  reliability 255/255, txload 1/255, rxload 1/255
  Encapsulation ARPA
  Port mode is trunk
  full-duplex, 10 Gb/s, media type is 10G
  Beacon is turned off
  Input flow-control is off, output flow-control is on
  Rate mode is dedicated
  Switchport monitor is off
  EtherType is 0x8100
  Last link flapped 4week(s) 2day(s)
  Last clearing of "show interface" counters 12w4d
  30 seconds input rate 22629552 bits/sec, 4910 packets/sec
  30 seconds output rate 30351576 bits/sec, 5941 packets/sec
  Load-Interval #2: 5 minute (300 seconds)
    input rate 19.95 Mbps, 4.39 Kpps; output rate 24.73 Mbps, 5.39 Kpps
  RX
    22539910578 unicast packets  2069107 multicast packets  1747946 broadcast packets
    22543727631 input packets  27020659779780 bytes
    16312444684 jumbo packets  0 storm suppression bytes
    0 runts  0 giants  0 CRC  0 no buffer
    0 input error  0 short frame  0 overrun   0 underrun  0 ignored
    0 watchdog  0 bad etype drop  0 bad proto drop  0 if down drop
    0 input with dribble  0 input discard
    0 Rx pause
  TX
    153686449300 unicast packets  1159976745 multicast packets  359279597 broadcast packets
    155205705642 output packets  218143523955218 bytes
    138086374308 jumbo packets
    0 output errors  0 collision  0 deferred  0 late collision
    0 lost carrier  0 no carrier  0 babble 0 output discard
    734044274 Tx pause
  1 interface resets

[Keny Perez · Accepted solution]

You might not have enough uplinks between your FEX and N5K so the FEX sends pause frames down the link to the server.

See below:

http://www.cisco.com/c/en/us/td/docs/switches/datacenter/nexus5000/sw/troubleshooting/guide/N5K_Troubleshooting_Guide/n5K_ts_qos.html#pgfId-1039331

If it helps, please rate it.

-Kenny

[clayhawks]

Thanks! So the TX pause frames are sourced by the 5548/FEX and passed on the wire to the destination which in this case is the HP FlexFabric module. Is that accurate? 

[Keny Perez]

without looking at any more logs, I would actually guess that your FEX cannot effectively deliver the frames to the N5K due to the lack of enough uplinks and sends a Tx Pause frame down the server while it waits to have a chance to send the frames upstream to the N5K.

-Kenny

[clayhawks]

Thanks. That is the information I was looking for. I suspected that this was the cause of the TX pause frames, but I wanted to get some corroboration.

[Keny Perez]

My pleasure :)

-Kenny