我在centos 5.5中创build了一个bonding界面，我select了mode-balance，使用slave eth6和eth7，并且eth6和eth7的连接正常，这里是configuration文件：
[root@50:B3:42:00:00:A3 network-scripts]# cat ifcfg-bond1 DEVICE=bond1 IPADDR=172.16.252.225 NETMASK=255.255.0.0 GATEWAY=172.16.0.1 MTU=9000 ONBOOT==yes BOOTPROTO=none USERCTL=on BONDING_OPTS="mode=balance-rr miimon=100" [root@50:B3:42:00:00:A3 network-scripts]# cat ifcfg-eth6 DEVICE=eth6 USERCTL=no ONBOOT=yes MASTER=bond1 SLAVE=yes BOOTPROTO=none [root@50:B3:42:00:00:A3 network-scripts]# cat ifcfg-eth7 DEVICE=eth7 USERCTL=no ONBOOT=yes MASTER=bond1 SLAVE=yes BOOTPROTO=none 并且绑定界面已经成功build立，sysfs：
[root@50:B3:42:00:00:A3 network-scripts]# cat /sys/class/net/bond1/bonding/slaves eth6 eth7 [root@50:B3:42:00:00:A3 network-scripts]# cat /sys/class/net/bond1/bonding/mode balance-rr 0 [root@50:B3:42:00:00:A3 network-scripts]# cat /sys/class/net/bond1/bonding/mii miimon mii_status [root@50:B3:42:00:00:A3 network-scripts]# cat /sys/class/net/bond1/bonding/miimon 100 [root@50:B3:42:00:00:A3 network-scripts]# cat /proc/net/bonding/bond1 Ethernet Channel Bonding Driver: v3.4.0 (October 7, 2008) Bonding Mode: load balancing (round-robin) MII Status: up MII Polling Interval (ms): 100 Up Delay (ms): 0 Down Delay (ms): 0 Slave Interface: eth6 MII Status: up Link Failure Count: 0 Permanent HW addr: 50:b3:42:00:00:74 Slave Interface: eth7 MII Status: up Link Failure Count: 0 Permanent HW addr: 50:b3:42:00:00:75
我可以ping通债券接口，链接状态正常：
# ping 172.16.252.225 PING 172.16.252.225 (172.16.252.225) 56(84) bytes of data. 64 bytes from 172.16.252.225: icmp_seq=1 ttl=64 time=1.88 ms 64 bytes from 172.16.252.225: icmp_seq=2 ttl=64 time=0.122 ms 64 bytes from 172.16.252.225: icmp_seq=3 ttl=64 time=0.112 ms 64 bytes from 172.16.252.225: icmp_seq=4 ttl=64 time=0.110 ms 64 bytes from 172.16.252.225: icmp_seq=5 ttl=64 time=0.117 ms
然后用IOmeter通过bonding接口对磁盘进行stream读写操作，查看数据stream的分布情况，绑定接口的模式为balance-rr，使用命令分析数据stream：
sar -n DEV 2 100
但结果与预期不一样：
# sar -n DEV 2 100 Linux 2.6.18-194.17.1.el5 (50:B3:42:00:00:A3) 04/28/2012 06:32:32 PM IFACE rxpck/s txpck/s rxbyt/s txbyt/s rxcmp/s txcmp/s rxmcst/s 06:32:34 PM lo 0.00 0.00 0.00 0.00 0.00 0.00 0.00 06:32:34 PM eth0 318.78 4316.24 19691.88 439105.58 0.00 0.00 0.00 06:32:34 PM inner0 0.00 0.51 0.00 47.72 0.00 0.00 0.00 06:32:34 PM eth5 1.02 0.00 93.40 0.00 0.00 0.00 0.00 06:32:34 PM eth6 4499.49 0.00 2631274.62 0.00 0.00 0.00 0.00 06:32:34 PM eth7 236.55 0.00 14350.76 0.00 0.00 0.00 0.00 06:32:34 PM eth8 0.00 0.00 0.00 0.00 0.00 0.00 0.00 06:32:34 PM sit0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 06:32:34 PM bond0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 06:32:34 PM bond1 4736.04 0.00 2645625.38 0.00 0.00 0.00 0.00
我们可以看到数据stream不平衡，大部分数据stream在eth6上，mabye有些参数设置不正确，还有一些其他的错误，你能帮我吗？非常感谢！
机器和系统信息：
Linux 2.6.18-194.17.1.el5 x86_64 GNU/Linux
邦定驱动版本：
v3.4.0 (October 7, 2008)
网卡：
intel 82574L 1000Mb/s
开关：
H3C S5800 # interface GigabitEthernet1/0/7 port link-aggregation group 1 # interface GigabitEthernet1/0/9 port link-aggregation group 1 # interface GigabitEthernet1/0/12 port link-aggregation group 1 # interface GigabitEthernet1/0/22 port link-aggregation group
我的机器连接到交换机上的这4个端口。 同时我configuration了相应接口的链路端口types为access 。
您不显示您的交换机configuration。 通常，balance-rr模式需要交换机上的Etherchannel（Cisco）或中继configuration才能正常工作。 您是否在H3C S5800交换机上build立了一个分组？
在这里看到笔记 。
The balance-rr, balance-xor and broadcast modes generally require that the switch have the appropriate ports grouped together. The nomenclature for such a group differs between switches, it may be called an "etherchannel" (as in the Cisco example, above), a "trunk group" or some other similar variation.