雙網(wǎng)卡綁定問題

echoaix

IBM 刀片小機(jī) 運(yùn)行l(wèi)inux 5片刀。
#uname -a
2.6.32-343.el6.ppc64 #1 SMP Mon Nov 19 16:49:16 EST 2012 ppc64 ppc64 ppc64 GNU/Linux
每片刀上eth0和eth1雙網(wǎng)卡綁定。
ifcfg-bong0內(nèi)容：
DEVICE=bond0                     
IPADDR=192.168.3.51               
NETMASK=255.255.255.0            
NETWORK=192.168.3.0               
BROADCAST=192.168.3.255           
ONBOOT=yes                        
GATEWAY=192.168.3.254            
BOOTPROTO=static                  
USERCTL=no                        
BONDING_OPTS="mode=1 miimon=500"            
DEFROUTE=yes

ifcfg-eth0內(nèi)容
DEVICE=eth0     
TYPE=Ethernet   
ONBOOT=yes      
BOOTPROTO=none  
USERCTL=no      
MASTER=bond0   
SLAVE=yes      

ifcfg-eth1內(nèi)容
DEVICE=eth1     
TYPE=Ethernet   
ONBOOT=yes      
BOOTPROTO=none  
USERCTL=no      
MASTER=bond0   
SLAVE=yes

/etc/rc.d/rc.local的內(nèi)容
ifenslave bond0 eth0 eth1         
route add default gw 192.168.3.254

現(xiàn)在的問題 只有一臺(tái)正常路由正常
正常的路由：
# route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.168.3.0     *               255.255.255.0   U     0      0        0 bond0
link-local      *               255.255.0.0     U     1004   0        0 bond0
default         192.168.3.254   0.0.0.0         UG    0      0        0 bond0

其它的不對(duì)的：
# route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.168.3.0     *               255.255.255.0   U     0      0        0 bond0
192.168.3.0     *               255.255.255.0   U     1      0        0 eth1
192.168.3.0     *               255.255.255.0   U     1      0        0 eth0
link-local      *               255.255.0.0     U     1004   0        0 bond0
default         192.168.3.254   0.0.0.0         UG    0      0        0 bond0
default         192.168.3.254   0.0.0.0         UG    0      0        0 eth0  

配置完全都一樣的。網(wǎng)卡文件也一樣的（只是IP設(shè)置不同）。咨詢一下什么問題，謝謝各位了。  



                     

chenyx

去掉rc.local里面的內(nèi)容呢?

milujite

1. 1. For a channel bonding interface to be valid, the bonding module must be loaded. To ensure that the module is loaded when the channel bonding interface is brought up, create a new file as root named +bonding.conf+** in the +/etc/modprobe.d/+ directory. Note that you can name this file anything you like as long as it ends with a .conf extension. Insert the following line in this new file:
   
2. /etc/modprobe.d/bonding.conf :
   
3. alias bond<N> bonding
   
4. Replace <N> with the  interface number, such as 0. For each configured channel bonding  interface, there must be a corresponding entry in your new  /etc/modprobe.d/<bonding>.conf file.
   
5. 2. To create a channel bonding interface, create a file in the /etc/sysconfig/network-scripts/ directory called ifcfg-bond<N> , replacing <N> with the number for the interface, such as 0. The contents of the file can be identical to whatever type of interface is getting bonded, such as an Ethernet interface. The only difference is that the DEVICE= directive must be bond<N> , replacing <N> with the number for the interface.
   
6. The following is a sample channel bonding configuration file:
   
7. DEVICE=bond0
   
8. IPADDR=192.168.1.1
   
9. NETMASK=255.255.255.0
   
10. ONBOOT=yes
    
11. BOOTPROTO=none
    
12. USERCTL=no
    
13. BONDING_OPTS="<bonding parameters separated by spaces>"
    
14. 3. After the channel bonding interface is created, the network interfaces to be bound together must be configured by adding the MASTER= and SLAVE= directives to their configuration files. The configuration files for each of the channel-bonded interfaces can be nearly identical.  For example, if two Ethernet interfaces are being channel bonded, both eth0 and eth1 may look like the following example:
    
15. DEVICE=eth<N>
    
16. BOOTPROTO=none
    
17. ONBOOT=yes
    
18. MASTER=bond0
    
19. SLAVE=yes
    
20. USERCTL=no
    
21. In this example, replace <N> with the numerical value for the interface.
    
22. 4. Restart the service network and make the bond0 up.
    
23. service network restart
    
24. *Note :*
    
25. Parameters for the bonding kernel module must be specified as a space-separated list in the BONDING_OPTS="<bonding parameters>"  directive in the ifcfg-bond<N>  interface file. Do not specify options for the bonding device in /etc/modprobe.d/<bonding>.conf, or in the deprecated /etc/modprobe.conf file.
    

復(fù)制代碼

按RH的方式來吧，試試。路由盡量別寫RC.LOCAL，這文件只在開機(jī)時(shí)候執(zhí)行一次。其他時(shí)候網(wǎng)卡一重啟就丟路由了。建議路由寫到配置文件里。

echoaix

chenyx 發(fā)表于 2013-09-09 15:42
去掉rc.local里面的內(nèi)容呢?

謝謝
去掉也不行的，只是少了bond0的這條路由，夸網(wǎng)段直接不通了。

echoaix

補(bǔ)充一下：
正常的機(jī)器上 eth0和eth1 都是沒有IP地址的。
# ifconfig
bond0     Link encap:Ethernet  HWaddr 34:40:B5:CF:F0:5A  
          inet addr:192.168.3.51  Bcast:192.168.3.255  Mask:255.255.255.0
          inet6 addr: fe80::3640:b5ff:fecf:f05a/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST  MTU:1500  Metric:1
          RX packets:285 errors:0 dropped:0 overruns:0 frame:0
          TX packets:215 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:22316 (21.7 KiB)  TX bytes:22055 (21.5 KiB)

eth0      Link encap:Ethernet  HWaddr 34:40:B5:CF:F0:5A  
          UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
          RX packets:274 errors:0 dropped:0 overruns:0 frame:0
          TX packets:215 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:21810 (21.2 KiB)  TX bytes:22055 (21.5 KiB)

eth1      Link encap:Ethernet  HWaddr 34:40:B5:CF:F0:5A  
          UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
          RX packets:11 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:506 (506.0 b)  TX bytes:0 (0.0 b)

不正常的機(jī)器： eth0和eth1 都是有IP地址的。
# ifconfig
bond0     Link encap:Ethernet  HWaddr 34:40:B5:CF:F3:BA  
          inet addr:192.168.3.55  Bcast:192.168.3.255  Mask:255.255.255.0
          inet6 addr: fe80::3640:b5ff:fecf:f3ba/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST  MTU:1500  Metric:1
          RX packets:53321 errors:0 dropped:0 overruns:0 frame:0
          TX packets:31784 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:3082003 (2.9 MiB)  TX bytes:2318806 (2.2 MiB)

eth0      Link encap:Ethernet  HWaddr 34:40:B5:CF:F3:BA  
          inet addr:192.168.3.55  Bcast:192.168.3.255  Mask:255.255.255.0
          UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
          RX packets:42207 errors:0 dropped:0 overruns:0 frame:0
          TX packets:31784 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:2480752 (2.3 MiB)  TX bytes:2318806 (2.2 MiB)

eth1      Link encap:Ethernet  HWaddr 34:40:B5:CF:F3:BA  
          inet addr:192.168.3.55  Bcast:192.168.3.255  Mask:255.255.255.0
          UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
          RX packets:11114 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:601251 (587.1 KiB)  TX bytes:0 (0.0 b)

chenyx

這個(gè)就讓人費(fèi)解了.
看你的ifconfig的輸出,所有的mac都是一樣的,按道理bond已經(jīng)成功了.為什么ethX還會(huì)被設(shè)置上ip呢,奇怪