3

I want have 2Gbps bandwidth (peer to peer) with 2 1Gbps NIC. It's on HP MicroServer Gen8, 2 on board NICs.

I have bind 2 NIC teaming with LACP, dynamic load balance, but one interface has large traffic, other has not traffic. (NIC0 has traffic NIC1 has no traffic, and I exchaged NIC0 and NIC1 in swich hub, NIC1 has traffic, and NIC0 no)

traffic

the switch hub configuration (h3c s5008pv2) below:


GigabitEthernet1/0/4 current state: UP
 IP Packet Frame Type: PKTFMT_ETHNT_2, Hardware Address: ----------------
 Description: GigabitEthernet1/0/4 Interface
 Loopback is not set
 Media type is twisted pair, Port hardware type is 1000_BASE_T
 1000Mbps-speed mode, full-duplex mode
 Link speed type is autonegotiation, link duplex type is autonegotiation
 Flow-control is not enabled
 The Maximum Frame Length is 9600
 Broadcast MAX-ratio: 100%
 Unicast MAX-ratio: 100%
 Multicast MAX-ratio: 100%
 PVID: 1
 Mdi type: auto
 Port link-type: trunk
  VLAN passing  : 1(default vlan), 10, 16, 192, 200
  VLAN permitted: 1(default vlan), 2-4094
  Trunk port encapsulation: IEEE 802.1q
 Port priority: 0
 Last clearing of counters:  Never
 Peak value of input: 251958 bytes/sec, at 2000-05-02 18:09:53
 Peak value of output: 1280345 bytes/sec, at 2000-05-02 17:38:18
 Last 300 seconds input:  5 packets/sec 5956 bytes/sec 0%
 Last 300 seconds output:  1 packets/sec 206 bytes/sec 0%
 Input (total):  14079379 packets, 1348337038 bytes
     14072290 unicasts, 3183 broadcasts, 3906 multicasts, 0 pauses
 Input (normal):  14079379 packets, 1348337038 bytes
     14072290 unicasts, 3183 broadcasts, 3906 multicasts, 0 pauses
 Input:  0 input errors, 0 runts, - giants, - throttles
     0 CRC, - frame, 0 overruns, 0 aborts
     - ignored, - parity errors
 Output (total): 2114263 packets, 522544778 bytes
     253952 unicasts, 1683663 broadcasts, 176648 multicasts, 0 pauses
 Output (normal): 2114263 packets, 522544778 bytes
     253952 unicasts, 1683663 broadcasts, 176648 multicasts, 0 pauses
 Output: 0 output errors, - underruns, - buffer failures
     0 aborts, 0 deferred, 0 collisions, - late collisions
     - lost carrier, - no carrier




 Bridge-Aggregation3 current state: UP
 IP Packet Frame Type: PKTFMT_ETHNT_2, Hardware Address: ----------------
 Description: Bridge-Aggregation3 Interface
 2Gbps-speed mode, full-duplex mode
 Link speed type is autonegotiation, link duplex type is autonegotiation
 PVID: 1
 Port link-type: trunk
  VLAN passing  : 1(default vlan), 10, 16, 192, 200
  VLAN permitted: 1(default vlan), 2-4094
  Trunk port encapsulation: IEEE 802.1q
 Last clearing of counters:  Never
 Last 300 seconds input:  9 packets/sec 9458 bytes/sec    0%
 Last 300 seconds output:  11 packets/sec 1188 bytes/sec    0%
 Input (total):  24112713 packets, 2337853081 bytes
     24091994 unicasts, 5166 broadcasts, 15553 multicasts
 Input (normal):  24112713 packets, 2337853081 bytes
     24091994 unicasts, 5166 broadcasts, 15553 multicasts
 Input:  0 input errors, 0 runts, - giants, - throttles
     0 CRC, - frame, 0 overruns, 0 aborts
     - ignored, - parity errors
 Output (total): 115638980 packets, 164823779464 bytes
     113243640 unicasts, 2068522 broadcasts, 326818 multicasts, 0 pauses
 Output (normal): 115638980 packets, 164823779464 bytes
         113243640 unicasts, 2068522 broadcasts, 326818 multicasts, 0 pauses
 Output: 0 output errors, - underruns, - buffer failures
     0 aborts, 0 deferred, 0 collisions, - late collisions
     - lost carrier, - no carrier

GigabitEthernet1/0/4 is the port which seems not work(but status is UP, and NIC status in win2012 is ACTIVE)

Bridge-Aggregation3 is the bond port(GigaEth1/0/3 & GigaEth1/0/4).

3 Answers 3

3

You need multiple TCP connections over your LACP link to get aggregated performance. With only one TCP session you’ll get best NIC in the team performance and that’s it! Good wrap up and a clarifying picture are both here:

https://community.arubanetworks.com/t5/Controller-Based-WLANs/Will-configuring-LACP-increase-throughput-on-the-link/ta-p/517925

2

You will never get 2Gbps bandwith peer to peer with LACP, you will only get 2Gbps combined bandwith (theoretically) with multiple clients accessing the server at the same time.

2
  • I have tried transfer large file from two different computer to this server at the same time, each one have 1gbps max, but only1gbps on this server
    – user219031
    Jan 1, 2017 at 15:47
  • 2
    It’s about TCP sessions, not any mystical “clients”.
    – RiGiD5
    Mar 21, 2020 at 22:01
1

If I'm understanding you correctly, there is always activity to the same switch port regardless of which NIC it's plugged in to? This rather suggests the problem is with the switch.

You mention that you've set the server up to use LACP - this needs to be configured on the switch as well as on the server.

I believe Windows also supports "Static Teaming" and "Switch Independent" settings for NIC teaming. If you cannot configure your switch to use LACP then consider using one of these modes as appropriate.

2
  • 1
    tks, switch (h3c s5008) has set bridge aggregration with lacp.
    – user219031
    Jan 1, 2017 at 15:50
  • if set to static aggregation, same problem.
    – user219031
    Jan 1, 2017 at 15:56

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .