Login or Register to Ask a Question and Join Our Community


UNIX for Advanced & Expert Users

Bonding IEEE 802.3ad Dynamic link aggregation : Bond showing less than desired throughput

Login to Discuss or Reply to this Discussion in Our Community

 
Thread Tools Search this Thread
Top Forums UNIX for Advanced & Expert Users Bonding IEEE 802.3ad Dynamic link aggregation : Bond showing less than desired throughput
# 1  
Old 04-26-2017
Bonding IEEE 802.3ad Dynamic link aggregation : Bond showing less than desired throughput
Hi All,
I have done IEEE 802.3ad Dynamic link aggregation bond configuration with name bond0 which has 4 slaves (each 25GB/s) in it on cent os 6.8. Issue i am facing is bonding throughput is only 50GB/s not 100GB/s. below are the configuration files :
Code:
DEVICE=bond0
IPADDR=xx.xx.xx.xx
NETMASK=yyy.yyy.yyy.yyy
GATEWAY=zz.zz.zz.zz
USERCTL=no
BOOTPROTO=none
ONBOOT=yes
BONDING_OPTS="mode=4 miimon=100 xmit_hash_policy=1"
MTU=9000
IPV6INIT=yes
IPV6ADDR=xxxx:xxxx:xxxx:xxxx:xxxx
#IPV6_DEFAULTGW=yyyy:yyyy:yyyy:yyyy
NM_CONTROLLED=no

TYPE=Ethernet
BOOTPROTO=static
DEVICE=eth0
ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED=no
HWADDR=9C:DC:71:47:3C:90
UUID=6e157026-ea38-4299-a183-68159ed86236

TYPE=Ethernet
BOOTPROTO=none
DEVICE=eth1
ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED=no
HWADDR=9C:DC:71:47:3C:91
UUID=11d180a3-e481-4372-83a3-8fec06fb1425

TYPE=Ethernet
BOOTPROTO=none
DEVICE=eth2
ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED=no
HWADDR=9C:DC:71:47:9B:60
UUID=7b9812ac-9926-417a-9825-6c1f3f285117

TYPE=Ethernet
BOOTPROTO=none
DEVICE=eth3
ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED=no
HWADDR=9C:DC:71:47:9B:61
UUID=5b0d4d31-0c7e-4fce-b1ab-87111b4f7942

also below is the output of /proc/net/bonding/bond0 which shows number of ports only 2 not 4 :
Code:
Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011)

Bonding Mode: IEEE 802.3ad Dynamic link aggregation
Transmit Hash Policy: layer3+4 (1)
MII Status: up
MII Polling Interval (ms): 100
Up Delay (ms): 0
Down Delay (ms): 0

802.3ad info
LACP rate: slow
Min links: 0
Aggregator selection policy (ad_select): stable
Active Aggregator Info:
        Aggregator ID: 9
        Number of ports: 2
        Actor Key: 1
        Partner Key: 65
        Partner Mac Address: 00:d7:8f:7f:5d:8b

Slave Interface: eth0
MII Status: up
Speed: 25000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 9c:dc:71:47:3c:90
Aggregator ID: 9
Slave queue ID: 0

Slave Interface: eth1
MII Status: up
Speed: 25000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 9c:dc:71:47:3c:91
Aggregator ID: 10
Slave queue ID: 0

Slave Interface: eth2
MII Status: up
Speed: 25000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 9c:dc:71:47:9b:60
Aggregator ID: 9
Slave queue ID: 0

Slave Interface: eth3
MII Status: up
Speed: 25000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 9c:dc:71:47:9b:61
Aggregator ID: 10
Slave queue ID: 0

also while making bond up i am getting below error :

Code:
# ifup bond0
Device eth1 has different MAC address than expected, ignoring.
Unable to start slave device ifcfg-eth1 for master bond0.
Device eth2 has different MAC address than expected, ignoring.
Unable to start slave device ifcfg-eth2 for master bond0.
Device eth3 has different MAC address than expected, ignoring.
Unable to start slave device ifcfg-eth3 for master bond0.

Please suggest to find out the root cause of it.

---------- Post updated at 08:42 AM ---------- Previous update was at 02:10 AM ----------

Any solution ... kindly suggest
# 2  
Old 04-27-2017
I am not an expert with link aggregation, maybe nobody here is.
It seems there is aggregated eth0 with eth2, and eth1 with eth3.
The only unusual setting is the BOOTPROTO=static for eth0.
You have set xmit_hash_policy=1.
Maybe you should change to xmit_hash_policy=layer3+4, then two hosts can use more interfaces if more services (ports) are involved. But the other way is controlled by the LAN switch.
And - maybe everything works as designed, according to this article.
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. IP Networking

Link Aggregation

Hi ihave three link of internet and iwant to put one linux front of a firewall that this three linux speard firewall such az load balance and fialover but dont close any port and protocol and only firewall have been internet what way can i use for it ? are there any script and services do that... (0 Replies)
Discussion started by: mnnn
0 Replies

2. Red Hat

Bonding a Bond with LACP

Does anyone know if it's possible to bond two bonds together? My situation is I have two older Cisco switches that cannot carry a LACP (bond level 4) aggregated between them, but separate aggregates can be setup on the switches themselves. In order to have redundancy of two switches I would... (0 Replies)
Discussion started by: christr
0 Replies

3. Solaris

solaris link aggregation problem , once i reboot it is not showing, not able to ping the default gat

Hi All, I am trying to aggregate the NIC's,(igb2 and igb3) (igb0 is used by the physical system and igb1 is used by primary-vsw0) to create the domains on that for faster data transfer, I followed the process for creating the aggregation, dladm create-aggr -d igb2 -d igb3 1 after creating the... (2 Replies)
Discussion started by: buildscm
2 Replies

4. Solaris

Throughput problems with solaris aggregation

Hello gurus, I have the following configuration in the server side: # dladm show-aggr key: 33 (0x0021) policy: L4 address: 0:14:4f:6c:11:8 (auto) device address speed duplex link state nxge0 0:14:4f:6c:11:8 1000 Mbps ... (3 Replies)
Discussion started by: FERCA
3 Replies

5. IP Networking

Interface bonding / Link aggregation (Multiple)

Hello, I've been using mode 4 with four slaves, however looking at ifconfig showed that the traffic was not balanced correctly between the interfaces, the outgoing traffic has been alot higher on the last slave. Example: eth0 RX 123.2 GiB TX 22.5 GiB eth1 RX 84.8 GiB TX 8.3 GiB eth2... (3 Replies)
Discussion started by: TehOne
3 Replies

6. Solaris

Link aggregation

Me again :) I'm trying to find a page describing the L2, L3 und L4 modes of dladm. It's nice to read "hashed by ip header", but how should I use that? On the file-server it's ok to have the six interfaces serving six clients each on it's own. But an rsync connection via switch between two... (8 Replies)
Discussion started by: PatrickBaer
8 Replies

7. IP Networking

bonding without switch link aggregation

I have some linux machines that I am trying to increase the throughput to on a single connection. They connect to a switch with two 1GbE lines and the switch does not have Link Aggregation support for these machines. I have tried bonding with balance-rr, balance-alb, but the machines can only... (4 Replies)
Discussion started by: Eruditass
4 Replies

8. IP Networking

Gigabit Link throughput

As a rule of thumb in doing calculations, what figure would you use in Mbytes/sec? I know the answer varies grealty on the topolgy of the network but I wonde what newteok engineers use a rough rule of thumb? Many thanks. (1 Reply)
Discussion started by: debd
1 Replies

9. IP Networking

bonding lacp and link aggregation

Hello, I am trying to get clarity on a few things and am looking for some info. In every article I have read about link aggregation and lacp, it can be used combine physical links to create 1 logical link for increased bandwidth. But what it doesn't say is if this is limited by source/dst. ... (1 Reply)
Discussion started by: jaredo
1 Replies

10. Solaris

Link Aggregation

Hi there I have a requirement to provide failover to our customer boxes in case of interface / switch failure, I have been looking at Solaris Link Aggregation with LACP and I wanted to ask a question Ive seen multiple websites that say the following Does this also mean that if the... (2 Replies)
Discussion started by: hcclnoodles
2 Replies
Login or Register to Ask a Question