1 <?xml version="1.0" encoding="UTF-8"?>
2 <chapter version="5.0" xml:lang="en-US" xmlns="http://docbook.org/ns/docbook" xmlns:xl="http://www.w3.org/1999/xlink" xml:id='settingupbonding'>
4 <title xml:id='settingupbonding.title'>Setting Up Network Interface Bonding</title>
7 <para>This chapter describes how to use multiple network interfaces in parallel to increase bandwidth and/or redundancy. Topics include:</para>
12 <xref linkend='dbdoclet.50438258_95824'/>
17 <xref linkend='dbdoclet.50438258_29107'/>
22 <xref linkend='dbdoclet.50438258_25353'/>
27 <xref linkend='dbdoclet.50438258_26899'/>
32 <xref linkend='dbdoclet.50438258_54769'/>
37 <xref linkend='dbdoclet.50438258_92244'/>
42 <note><para>Using network interface bonding is optional.</para></note>
45 <section xml:id="dbdoclet.50438258_95824">
46 <title>7.1 Network Interface Bonding Overview</title>
47 <para>Bonding, also known as link aggregation, trunking and port trunking, is a method of aggregating multiple physical network links into a single logical link for increased bandwidth.</para>
48 <para>Several different types of bonding are available in Linux. All these types are referred to as 'modes', and use the bonding kernel module.</para>
49 <para>Modes 0 to 3 allow load balancing and fault tolerance by using multiple interfaces. Mode 4 aggregates a group of interfaces into a single virtual interface where all members of the group share the same speed and duplex settings. This mode is described under IEEE spec 802.3ad, and it is referred to as either 'mode 4' or '802.3ad.'</para>
51 <section xml:id="dbdoclet.50438258_29107">
52 <title>7.2 Requirements</title>
53 <para>The most basic requirement for successful bonding is that both endpoints of the connection must be capable of bonding. In a normal case, the non-server endpoint is a switch. (Two systems connected via crossover cables can also use bonding.) Any switch used must explicitly handle 802.3ad Dynamic Link Aggregation.</para>
54 <para>The kernel must also be configured with bonding. All supported Lustre kernels have bonding functionality. The network driver for the interfaces to be bonded must have the ethtool functionality to determine slave speed and duplex settings. All recent network drivers implement it.</para>
55 <para>To verify that your interface works with ethtool, run:</para>
56 <screen># which ethtool
61 Supported ports: [ TP MII ]
62 Supported link modes: 10baseT/Half 10baseT/Full/
63 100baseT/Half 100baseT/Full
64 Supports auto-negotiation: Yes
65 Advertised link modes: 10baseT/Half 10baseT/Full
66 100baseT/Half 100baseT/Full
67 Advertised auto-negotiation: Yes
74 Supports Wake-on: pumbg
76 Current message level: 0x00000001 (1)
82 Supported ports: [ TP MII ]
83 Supported link modes: 10baseT/Half 10baseT/Full
84 100baseT/Half 100baseT/Full
85 Supports auto-negotiation: Yes
86 Advertised link modes: 10baseT/Half 10baseT/Full
87 100baseT/Half 100baseT/Full
88 Advertised auto-negotiation: Yes
95 Supports Wake-on: pumbg
97 Current message level: 0x00000007 (7)
99 To quickly check whether your kernel supports bonding, run:
100 # grep ifenslave /sbin/ifup
105 <section xml:id="dbdoclet.50438258_25353">
106 <title>7.3 Bonding Module Parameters</title>
107 <para>Bonding module parameters control various aspects of bonding.</para>
108 <para>Outgoing traffic is mapped across the slave interfaces according to the transmit hash policy. For Lustre, we recommend that you set the xmit_hash_policy option to the layer3+4 option for bonding. This policy uses upper layer protocol information if available to generate the hash. This allows traffic to a particular network peer to span multiple slaves, although a single connection does not span multiple slaves.</para>
109 <screen>$ xmit_hash_policy=layer3+4
111 <para>The miimon option enables users to monitor the link status. (The parameter is a time interval in milliseconds.) It makes an interface failure transparent to avoid serious network degradation during link failures. A reasonable default setting is 100 milliseconds; run:</para>
114 <para>For a busy network, increase the timeout.</para>
116 <section xml:id="dbdoclet.50438258_26899">
117 <title>7.4 Setting Up Bonding</title>
118 <para>To set up bonding:</para>
122 <para>Create a virtual 'bond' interface by creating a configuration file in:</para>
123 <screen>/etc/sysconfig/network-scripts/ # vi /etc/sysconfig/ network-scripts/ifcfg-\
128 <para>Append the following lines to the file.</para>
130 IPADDR=192.168.10.79 # Use the free IP Address of your network
132 NETMASK=255.255.255.0
139 <para> Attach one or more slave interfaces to the bond interface. Modify the eth0 and eth1 configuration files (using a VI text editor).</para>
142 <para>Use the VI text editor to open the eth0 configuration file.</para>
143 <screen># vi /etc/sysconfig/network-scripts/ifcfg-eth0
147 <para>Modify/append the eth0 file as follows:</para>
157 <para>Use the VI text editor to open the eth1 configuration file.</para>
158 <screen># vi /etc/sysconfig/network-scripts/ifcfg-eth1
162 <para>Modify/append the eth1 file as follows:</para>
174 <para>Set up the bond interface and its options in /etc/modprobe.conf. Start the slave interfaces by your normal network method.</para>
175 <screen># vi /etc/modprobe.conf
179 <para>Append the following lines to the file.</para>
180 <screen>alias bond0 bonding
181 options bond0 mode=balance-alb miimon=100
185 <para>Load the bonding module.</para>
186 <screen># modprobe bonding
188 # ifenslave bond0 eth0 eth1
194 <para>Start/restart the slave interfaces (using your normal network method).</para>
197 <para>You must modprobe the bonding module for each bonded interface. If you wish to create bond0 and bond1, two entries in modprobe.conf are required.</para></note>
199 <para>The examples below are from RedHat systems. For setup use: /etc/sysconfig/networking-scripts/ifcfg-* The website referenced below includes detailed instructions for other configuration methods, instructions to use DHCP with bonding, and other setup details. We strongly recommend you use this website.</para>
200 <para><link xl:href="http://www.linuxfoundation.org/collaborate/workgroups/networking/bonding">http://www.linuxfoundation.org/collaborate/workgroups/networking/bonding</link></para>
203 <para>Check /proc/net/bonding to determine status on bonding. There should be a file there for each bond interface.</para>
204 <screen># cat /proc/net/bonding/bond0
205 Ethernet Channel Bonding Driver: v3.0.3 (March 23, 2006)
207 Bonding Mode: load balancing (round-robin)
209 MII Polling Interval (ms): 0
213 Slave Interface: eth0
215 Link Failure Count: 0
216 Permanent HW addr: 4c:00:10:ac:61:e0
218 Slave Interface: eth1
220 Link Failure Count: 0
221 Permanent HW addr: 00:14:2a:7c:40:1d
225 <para>Use ethtool or ifconfig to check the interface state. ifconfig lists the first bonded interface as 'bond0.'</para>
227 bond0 Link encap:Ethernet HWaddr 4C:00:10:AC:61:E0
228 inet addr:192.168.10.79 Bcast:192.168.10.255 \ Mask:255.255.255.0
229 inet6 addr: fe80::4e00:10ff:feac:61e0/64 Scope:Link
230 UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
231 RX packets:3091 errors:0 dropped:0 overruns:0 frame:0
232 TX packets:880 errors:0 dropped:0 overruns:0 carrier:0
233 collisions:0 txqueuelen:0
234 RX bytes:314203 (306.8 KiB) TX bytes:129834 (126.7 KiB)
236 eth0 Link encap:Ethernet HWaddr 4C:00:10:AC:61:E0
237 inet6 addr: fe80::4e00:10ff:feac:61e0/64 Scope:Link
238 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
239 RX packets:1581 errors:0 dropped:0 overruns:0 frame:0
240 TX packets:448 errors:0 dropped:0 overruns:0 carrier:0
241 collisions:0 txqueuelen:1000
242 RX bytes:162084 (158.2 KiB) TX bytes:67245 (65.6 KiB)
243 Interrupt:193 Base address:0x8c00
245 eth1 Link encap:Ethernet HWaddr 4C:00:10:AC:61:E0
246 inet6 addr: fe80::4e00:10ff:feac:61e0/64 Scope:Link
247 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
248 RX packets:1513 errors:0 dropped:0 overruns:0 frame:0
249 TX packets:444 errors:0 dropped:0 overruns:0 carrier:0
250 collisions:0 txqueuelen:1000
251 RX bytes:152299 (148.7 KiB) TX bytes:64517 (63.0 KiB)
252 Interrupt:185 Base address:0x6000
259 <title>7.4.1 Examples</title>
260 <para>This is an example showing modprobe.conf entries for bonding Ethernet interfaces eth1 and eth2 to bond0:</para>
261 <screen># cat /etc/modprobe.conf
263 alias scsi_hostadapter sata_via
264 alias scsi_hostadapter1 usb-storage
265 alias snd-card-0 snd-via82xx
266 options snd-card-0 index=0
267 options snd-via82xx index=0
269 options bond0 mode=balance-alb miimon=100
270 options lnet networks=tcp
273 # cat /etc/sysconfig/network-scripts/ifcfg-bond0
276 NETMASK=255.255.255.0
277 IPADDR=192.168.10.79 # (Assign here the IP of the bonded interface.)
282 # cat /etc/sysconfig/network-scripts/ifcfg-eth0
285 HWADDR=4c:00:10:ac:61:e0
294 <para>In the following example, the bond0 interface is the master (MASTER) while eth0 and eth1 are slaves (SLAVE).</para>
295 <informaltable frame="none">
297 <colspec colname="c1" colwidth="100*"/>
300 <entry><para><emphasis role="bold">Note -</emphasis>All slaves of bond0 have the same MAC address (Hwaddr) - bond0. All modes, except TLB and ALB, have this MAC address. TLB and ALB require a unique MAC address for each slave.</para></entry>
305 <screen>$ /sbin/ifconfig
307 bond0Link encap:EthernetHwaddr 00:C0:F0:1F:37:B4
308 inet addr:XXX.XXX.XXX.YYY Bcast:XXX.XXX.XXX.255 Mask:255.255.252.0
309 UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
310 RX packets:7224794 errors:0 dropped:0 overruns:0 frame:0
311 TX packets:3286647 errors:1 dropped:0 overruns:1 carrier:0
312 collisions:0 txqueuelen:0
314 eth0Link encap:EthernetHwaddr 00:C0:F0:1F:37:B4
315 inet addr:XXX.XXX.XXX.YYY Bcast:XXX.XXX.XXX.255 Mask:255.255.252.0
316 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
317 RX packets:3573025 errors:0 dropped:0 overruns:0 frame:0
318 TX packets:1643167 errors:1 dropped:0 overruns:1 carrier:0
319 collisions:0 txqueuelen:100
320 Interrupt:10 Base address:0x1080
322 eth1Link encap:EthernetHwaddr 00:C0:F0:1F:37:B4
323 inet addr:XXX.XXX.XXX.YYY Bcast:XXX.XXX.XXX.255 Mask:255.255.252.0
324 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
325 RX packets:3651769 errors:0 dropped:0 overruns:0 frame:0
326 TX packets:1643480 errors:0 dropped:0 overruns:0 carrier:0
327 collisions:0 txqueuelen:100
328 Interrupt:9 Base address:0x1400
332 <section xml:id="dbdoclet.50438258_54769">
333 <title>7.5 Configuring Lustre with Bonding</title>
334 <para>Lustre uses the IP address of the bonded interfaces and requires no special configuration. It treats the bonded interface as a regular TCP/IP interface. If needed, specify bond0 using the Lustre networks parameter in /etc/modprobe.</para>
335 <screen>options lnet networks=tcp(bond0)
338 <section xml:id="dbdoclet.50438258_92244">
339 <title>7.6 Bonding References</title>
340 <para>We recommend the following bonding references:</para>
341 <itemizedlist><listitem>
342 <para> In the Linux kernel source tree, see documentation/networking/bonding.txt</para>
345 <para><link xl:href="http://linux-ip.net/html/ether-bonding.html">http://linux-ip.net/html/ether-bonding.html</link></para>
348 <para><link xl:href="http://www.sourceforge.net/projects/bonding">http://www.sourceforge.net/projects/bonding</link></para>
351 <para> Linux Foundation bonding website: <link xl:href="http://www.linuxfoundation.org/collaborate/workgroups/networking/bonding">http://www.linuxfoundation.org/collaborate/workgroups/networking/bonding</link></para>
354 <para>This is the most extensive reference and we highly recommend it. This website includes explanations of more complicated setups, including the use of DHCP with bonding.</para>