| blob | 61f516b135b415986217f130fcadf064a6bf5c83 |
2 Linux Ethernet Bonding Driver HOWTO
4 Latest update: 23 September 2009
6 Initial release : Thomas Davis <tadavis at lbl.gov>
7 Corrections, HA extensions : 2000/10/03-15 :
8 - Willy Tarreau <willy at meta-x.org>
9 - Constantine Gavrilov <const-g at xpert.com>
10 - Chad N. Tindel <ctindel at ieee dot org>
11 - Janice Girouard <girouard at us dot ibm dot com>
12 - Jay Vosburgh <fubar at us dot ibm dot com>
14 Reorganized and updated Feb 2005 by Jay Vosburgh
15 Added Sysfs information: 2006/04/24
16 - Mitch Williams <mitch.a.williams at intel.com>
18 Introduction
19 ============
21 The Linux bonding driver provides a method for aggregating
22 multiple network interfaces into a single logical "bonded" interface.
23 The behavior of the bonded interfaces depends upon the mode; generally
24 speaking, modes provide either hot standby or load balancing services.
25 Additionally, link integrity monitoring may be performed.
27 The bonding driver originally came from Donald Becker's
28 beowulf patches for kernel 2.0. It has changed quite a bit since, and
29 the original tools from extreme-linux and beowulf sites will not work
30 with this version of the driver.
32 For new versions of the driver, updated userspace tools, and
33 who to ask for help, please follow the links at the end of this file.
35 Table of Contents
36 =================
38 1. Bonding Driver Installation
40 2. Bonding Driver Options
42 3. Configuring Bonding Devices
43 3.1 Configuration with Sysconfig Support
44 3.1.1 Using DHCP with Sysconfig
45 3.1.2 Configuring Multiple Bonds with Sysconfig
46 3.2 Configuration with Initscripts Support
47 3.2.1 Using DHCP with Initscripts
48 3.2.2 Configuring Multiple Bonds with Initscripts
49 3.3 Configuring Bonding Manually with Ifenslave
50 3.3.1 Configuring Multiple Bonds Manually
51 3.4 Configuring Bonding Manually via Sysfs
53 4. Querying Bonding Configuration
54 4.1 Bonding Configuration
55 4.2 Network Configuration
57 5. Switch Configuration
59 6. 802.1q VLAN Support
61 7. Link Monitoring
62 7.1 ARP Monitor Operation
63 7.2 Configuring Multiple ARP Targets
64 7.3 MII Monitor Operation
66 8. Potential Trouble Sources
67 8.1 Adventures in Routing
68 8.2 Ethernet Device Renaming
69 8.3 Painfully Slow Or No Failed Link Detection By Miimon
71 9. SNMP agents
73 10. Promiscuous mode
75 11. Configuring Bonding for High Availability
76 11.1 High Availability in a Single Switch Topology
77 11.2 High Availability in a Multiple Switch Topology
78 11.2.1 HA Bonding Mode Selection for Multiple Switch Topology
79 11.2.2 HA Link Monitoring for Multiple Switch Topology
81 12. Configuring Bonding for Maximum Throughput
82 12.1 Maximum Throughput in a Single Switch Topology
83 12.1.1 MT Bonding Mode Selection for Single Switch Topology
84 12.1.2 MT Link Monitoring for Single Switch Topology
85 12.2 Maximum Throughput in a Multiple Switch Topology
86 12.2.1 MT Bonding Mode Selection for Multiple Switch Topology
87 12.2.2 MT Link Monitoring for Multiple Switch Topology
89 13. Switch Behavior Issues
90 13.1 Link Establishment and Failover Delays
91 13.2 Duplicated Incoming Packets
93 14. Hardware Specific Considerations
94 14.1 IBM BladeCenter
96 15. Frequently Asked Questions
98 16. Resources and Links
101 1. Bonding Driver Installation
102 ==============================
104 Most popular distro kernels ship with the bonding driver
105 already available as a module and the ifenslave user level control
106 program installed and ready for use. If your distro does not, or you
107 have need to compile bonding from source (e.g., configuring and
108 installing a mainline kernel from kernel.org), you'll need to perform
109 the following steps:
111 1.1 Configure and build the kernel with bonding
112 -----------------------------------------------
114 The current version of the bonding driver is available in the
115 drivers/net/bonding subdirectory of the most recent kernel source
116 (which is available on http://kernel.org). Most users "rolling their
117 own" will want to use the most recent kernel from kernel.org.
119 Configure kernel with "make menuconfig" (or "make xconfig" or
120 "make config"), then select "Bonding driver support" in the "Network
121 device support" section. It is recommended that you configure the
122 driver as module since it is currently the only way to pass parameters
123 to the driver or configure more than one bonding device.
125 Build and install the new kernel and modules, then continue
126 below to install ifenslave.
128 1.2 Install ifenslave Control Utility
129 -------------------------------------
131 The ifenslave user level control program is included in the
132 kernel source tree, in the file Documentation/networking/ifenslave.c.
133 It is generally recommended that you use the ifenslave that
134 corresponds to the kernel that you are using (either from the same
135 source tree or supplied with the distro), however, ifenslave
136 executables from older kernels should function (but features newer
137 than the ifenslave release are not supported). Running an ifenslave
138 that is newer than the kernel is not supported, and may or may not
139 work.
141 To install ifenslave, do the following:
143 # gcc -Wall -O -I/usr/src/linux/include ifenslave.c -o ifenslave
144 # cp ifenslave /sbin/ifenslave
146 If your kernel source is not in "/usr/src/linux," then replace
147 "/usr/src/linux/include" in the above with the location of your kernel
148 source include directory.
150 You may wish to back up any existing /sbin/ifenslave, or, for
151 testing or informal use, tag the ifenslave to the kernel version
152 (e.g., name the ifenslave executable /sbin/ifenslave-2.6.10).
154 IMPORTANT NOTE:
156 If you omit the "-I" or specify an incorrect directory, you
157 may end up with an ifenslave that is incompatible with the kernel
158 you're trying to build it for. Some distros (e.g., Red Hat from 7.1
159 onwards) do not have /usr/include/linux symbolically linked to the
160 default kernel source include directory.
162 SECOND IMPORTANT NOTE:
163 If you plan to configure bonding using sysfs, you do not need
164 to use ifenslave.
166 2. Bonding Driver Options
167 =========================
169 Options for the bonding driver are supplied as parameters to the
170 bonding module at load time, or are specified via sysfs.
172 Module options may be given as command line arguments to the
173 insmod or modprobe command, but are usually specified in either the
174 /etc/modules.conf or /etc/modprobe.conf configuration file, or in a
175 distro-specific configuration file (some of which are detailed in the next
176 section).
178 Details on bonding support for sysfs is provided in the
179 "Configuring Bonding Manually via Sysfs" section, below.
181 The available bonding driver parameters are listed below. If a
182 parameter is not specified the default value is used. When initially
183 configuring a bond, it is recommended "tail -f /var/log/messages" be
184 run in a separate window to watch for bonding driver error messages.
186 It is critical that either the miimon or arp_interval and
187 arp_ip_target parameters be specified, otherwise serious network
188 degradation will occur during link failures. Very few devices do not
189 support at least miimon, so there is really no reason not to use it.
191 Options with textual values will accept either the text name
192 or, for backwards compatibility, the option value. E.g.,
193 "mode=802.3ad" and "mode=4" set the same mode.
195 The parameters are as follows:
197 ad_select
199 Specifies the 802.3ad aggregation selection logic to use. The
200 possible values and their effects are:
202 stable or 0
204 The active aggregator is chosen by largest aggregate
205 bandwidth.
207 Reselection of the active aggregator occurs only when all
208 slaves of the active aggregator are down or the active
209 aggregator has no slaves.
211 This is the default value.
213 bandwidth or 1
215 The active aggregator is chosen by largest aggregate
216 bandwidth. Reselection occurs if:
218 - A slave is added to or removed from the bond
220 - Any slave's link state changes
222 - Any slave's 802.3ad association state changes
224 - The bond's administrative state changes to up
226 count or 2
228 The active aggregator is chosen by the largest number of
229 ports (slaves). Reselection occurs as described under the
230 "bandwidth" setting, above.
232 The bandwidth and count selection policies permit failover of
233 802.3ad aggregations when partial failure of the active aggregator
234 occurs. This keeps the aggregator with the highest availability
235 (either in bandwidth or in number of ports) active at all times.
237 This option was added in bonding version 3.4.0.
239 arp_interval
241 Specifies the ARP link monitoring frequency in milliseconds.
243 The ARP monitor works by periodically checking the slave
244 devices to determine whether they have sent or received
245 traffic recently (the precise criteria depends upon the
246 bonding mode, and the state of the slave). Regular traffic is
247 generated via ARP probes issued for the addresses specified by
248 the arp_ip_target option.
250 This behavior can be modified by the arp_validate option,
251 below.
253 If ARP monitoring is used in an etherchannel compatible mode
254 (modes 0 and 2), the switch should be configured in a mode
255 that evenly distributes packets across all links. If the
256 switch is configured to distribute the packets in an XOR
257 fashion, all replies from the ARP targets will be received on
258 the same link which could cause the other team members to
259 fail. ARP monitoring should not be used in conjunction with
260 miimon. A value of 0 disables ARP monitoring. The default
261 value is 0.
263 arp_ip_target
265 Specifies the IP addresses to use as ARP monitoring peers when
266 arp_interval is > 0. These are the targets of the ARP request
267 sent to determine the health of the link to the targets.
268 Specify these values in ddd.ddd.ddd.ddd format. Multiple IP
269 addresses must be separated by a comma. At least one IP
270 address must be given for ARP monitoring to function. The
271 maximum number of targets that can be specified is 16. The
272 default value is no IP addresses.
274 arp_validate
276 Specifies whether or not ARP probes and replies should be
277 validated in the active-backup mode. This causes the ARP
278 monitor to examine the incoming ARP requests and replies, and
279 only consider a slave to be up if it is receiving the
280 appropriate ARP traffic.
282 Possible values are:
284 none or 0
286 No validation is performed. This is the default.
288 active or 1
290 Validation is performed only for the active slave.
292 backup or 2
294 Validation is performed only for backup slaves.
296 all or 3
298 Validation is performed for all slaves.
300 For the active slave, the validation checks ARP replies to
301 confirm that they were generated by an arp_ip_target. Since
302 backup slaves do not typically receive these replies, the
303 validation performed for backup slaves is on the ARP request
304 sent out via the active slave. It is possible that some
305 switch or network configurations may result in situations
306 wherein the backup slaves do not receive the ARP requests; in
307 such a situation, validation of backup slaves must be
308 disabled.
310 This option is useful in network configurations in which
311 multiple bonding hosts are concurrently issuing ARPs to one or
312 more targets beyond a common switch. Should the link between
313 the switch and target fail (but not the switch itself), the
314 probe traffic generated by the multiple bonding instances will
315 fool the standard ARP monitor into considering the links as
316 still up. Use of the arp_validate option can resolve this, as
317 the ARP monitor will only consider ARP requests and replies
318 associated with its own instance of bonding.
320 This option was added in bonding version 3.1.0.
322 downdelay
324 Specifies the time, in milliseconds, to wait before disabling
325 a slave after a link failure has been detected. This option
326 is only valid for the miimon link monitor. The downdelay
327 value should be a multiple of the miimon value; if not, it
328 will be rounded down to the nearest multiple. The default
329 value is 0.
331 fail_over_mac
333 Specifies whether active-backup mode should set all slaves to
334 the same MAC address at enslavement (the traditional
335 behavior), or, when enabled, perform special handling of the
336 bond's MAC address in accordance with the selected policy.
338 Possible values are:
340 none or 0
342 This setting disables fail_over_mac, and causes
343 bonding to set all slaves of an active-backup bond to
344 the same MAC address at enslavement time. This is the
345 default.
347 active or 1
349 The "active" fail_over_mac policy indicates that the
350 MAC address of the bond should always be the MAC
351 address of the currently active slave. The MAC
352 address of the slaves is not changed; instead, the MAC
353 address of the bond changes during a failover.
355 This policy is useful for devices that cannot ever
356 alter their MAC address, or for devices that refuse
357 incoming broadcasts with their own source MAC (which
358 interferes with the ARP monitor).
360 The down side of this policy is that every device on
361 the network must be updated via gratuitous ARP,
362 vs. just updating a switch or set of switches (which
363 often takes place for any traffic, not just ARP
364 traffic, if the switch snoops incoming traffic to
365 update its tables) for the traditional method. If the
366 gratuitous ARP is lost, communication may be
367 disrupted.
369 When this policy is used in conjuction with the mii
370 monitor, devices which assert link up prior to being
371 able to actually transmit and receive are particularly
372 susceptible to loss of the gratuitous ARP, and an
373 appropriate updelay setting may be required.
375 follow or 2
377 The "follow" fail_over_mac policy causes the MAC
378 address of the bond to be selected normally (normally
379 the MAC address of the first slave added to the bond).
380 However, the second and subsequent slaves are not set
381 to this MAC address while they are in a backup role; a
382 slave is programmed with the bond's MAC address at
383 failover time (and the formerly active slave receives
384 the newly active slave's MAC address).
386 This policy is useful for multiport devices that
387 either become confused or incur a performance penalty
388 when multiple ports are programmed with the same MAC
389 address.
392 The default policy is none, unless the first slave cannot
393 change its MAC address, in which case the active policy is
394 selected by default.
396 This option may be modified via sysfs only when no slaves are
397 present in the bond.
399 This option was added in bonding version 3.2.0. The "follow"
400 policy was added in bonding version 3.3.0.
402 lacp_rate
404 Option specifying the rate in which we'll ask our link partner
405 to transmit LACPDU packets in 802.3ad mode. Possible values
406 are:
408 slow or 0
409 Request partner to transmit LACPDUs every 30 seconds
411 fast or 1
412 Request partner to transmit LACPDUs every 1 second
414 The default is slow.
416 max_bonds
418 Specifies the number of bonding devices to create for this
419 instance of the bonding driver. E.g., if max_bonds is 3, and
420 the bonding driver is not already loaded, then bond0, bond1
421 and bond2 will be created. The default value is 1. Specifying
422 a value of 0 will load bonding, but will not create any devices.
424 miimon
426 Specifies the MII link monitoring frequency in milliseconds.
427 This determines how often the link state of each slave is
428 inspected for link failures. A value of zero disables MII
429 link monitoring. A value of 100 is a good starting point.
430 The use_carrier option, below, affects how the link state is
431 determined. See the High Availability section for additional
432 information. The default value is 0.
434 mode
436 Specifies one of the bonding policies. The default is
437 balance-rr (round robin). Possible values are:
439 balance-rr or 0
441 Round-robin policy: Transmit packets in sequential
442 order from the first available slave through the
443 last. This mode provides load balancing and fault
444 tolerance.
446 active-backup or 1
448 Active-backup policy: Only one slave in the bond is
449 active. A different slave becomes active if, and only
450 if, the active slave fails. The bond's MAC address is
451 externally visible on only one port (network adapter)
452 to avoid confusing the switch.
454 In bonding version 2.6.2 or later, when a failover
455 occurs in active-backup mode, bonding will issue one
456 or more gratuitous ARPs on the newly active slave.
457 One gratuitous ARP is issued for the bonding master
458 interface and each VLAN interfaces configured above
459 it, provided that the interface has at least one IP
460 address configured. Gratuitous ARPs issued for VLAN
461 interfaces are tagged with the appropriate VLAN id.
463 This mode provides fault tolerance. The primary
464 option, documented below, affects the behavior of this
465 mode.
467 balance-xor or 2
469 XOR policy: Transmit based on the selected transmit
470 hash policy. The default policy is a simple [(source
471 MAC address XOR'd with destination MAC address) modulo
472 slave count]. Alternate transmit policies may be
473 selected via the xmit_hash_policy option, described
474 below.
476 This mode provides load balancing and fault tolerance.
478 broadcast or 3
480 Broadcast policy: transmits everything on all slave
481 interfaces. This mode provides fault tolerance.
483 802.3ad or 4
485 IEEE 802.3ad Dynamic link aggregation. Creates
486 aggregation groups that share the same speed and
487 duplex settings. Utilizes all slaves in the active
488 aggregator according to the 802.3ad specification.
490 Slave selection for outgoing traffic is done according
491 to the transmit hash policy, which may be changed from
492 the default simple XOR policy via the xmit_hash_policy
493 option, documented below. Note that not all transmit
494 policies may be 802.3ad compliant, particularly in
495 regards to the packet mis-ordering requirements of
496 section 43.2.4 of the 802.3ad standard. Differing
497 peer implementations will have varying tolerances for
498 noncompliance.
500 Prerequisites:
502 1. Ethtool support in the base drivers for retrieving
503 the speed and duplex of each slave.
505 2. A switch that supports IEEE 802.3ad Dynamic link
506 aggregation.
508 Most switches will require some type of configuration
509 to enable 802.3ad mode.
511 balance-tlb or 5
513 Adaptive transmit load balancing: channel bonding that
514 does not require any special switch support. The
515 outgoing traffic is distributed according to the
516 current load (computed relative to the speed) on each
517 slave. Incoming traffic is received by the current
518 slave. If the receiving slave fails, another slave
519 takes over the MAC address of the failed receiving
520 slave.
522 Prerequisite:
524 Ethtool support in the base drivers for retrieving the
525 speed of each slave.
527 balance-alb or 6
529 Adaptive load balancing: includes balance-tlb plus
530 receive load balancing (rlb) for IPV4 traffic, and
531 does not require any special switch support. The
532 receive load balancing is achieved by ARP negotiation.
533 The bonding driver intercepts the ARP Replies sent by
534 the local system on their way out and overwrites the
535 source hardware address with the unique hardware
536 address of one of the slaves in the bond such that
537 different peers use different hardware addresses for
538 the server.
540 Receive traffic from connections created by the server
541 is also balanced. When the local system sends an ARP
542 Request the bonding driver copies and saves the peer's
543 IP information from the ARP packet. When the ARP
544 Reply arrives from the peer, its hardware address is
545 retrieved and the bonding driver initiates an ARP
546 reply to this peer assigning it to one of the slaves
547 in the bond. A problematic outcome of using ARP
548 negotiation for balancing is that each time that an
549 ARP request is broadcast it uses the hardware address
550 of the bond. Hence, peers learn the hardware address
551 of the bond and the balancing of receive traffic
552 collapses to the current slave. This is handled by
553 sending updates (ARP Replies) to all the peers with
554 their individually assigned hardware address such that
555 the traffic is redistributed. Receive traffic is also
556 redistributed when a new slave is added to the bond
557 and when an inactive slave is re-activated. The
558 receive load is distributed sequentially (round robin)
559 among the group of highest speed slaves in the bond.
561 When a link is reconnected or a new slave joins the
562 bond the receive traffic is redistributed among all
563 active slaves in the bond by initiating ARP Replies
564 with the selected MAC address to each of the
565 clients. The updelay parameter (detailed below) must
566 be set to a value equal or greater than the switch's
567 forwarding delay so that the ARP Replies sent to the
568 peers will not be blocked by the switch.
570 Prerequisites:
572 1. Ethtool support in the base drivers for retrieving
573 the speed of each slave.
575 2. Base driver support for setting the hardware
576 address of a device while it is open. This is
577 required so that there will always be one slave in the
578 team using the bond hardware address (the
579 curr_active_slave) while having a unique hardware
580 address for each slave in the bond. If the
581 curr_active_slave fails its hardware address is
582 swapped with the new curr_active_slave that was
583 chosen.
585 num_grat_arp
587 Specifies the number of gratuitous ARPs to be issued after a
588 failover event. One gratuitous ARP is issued immediately after
589 the failover, subsequent ARPs are sent at a rate of one per link
590 monitor interval (arp_interval or miimon, whichever is active).
592 The valid range is 0 - 255; the default value is 1. This option
593 affects only the active-backup mode. This option was added for
594 bonding version 3.3.0.
596 num_unsol_na
598 Specifies the number of unsolicited IPv6 Neighbor Advertisements
599 to be issued after a failover event. One unsolicited NA is issued
600 immediately after the failover.
602 The valid range is 0 - 255; the default value is 1. This option
603 affects only the active-backup mode. This option was added for
604 bonding version 3.4.0.
606 primary
608 A string (eth0, eth2, etc) specifying which slave is the
609 primary device. The specified device will always be the
610 active slave while it is available. Only when the primary is
611 off-line will alternate devices be used. This is useful when
612 one slave is preferred over another, e.g., when one slave has
613 higher throughput than another.
615 The primary option is only valid for active-backup mode.
617 primary_reselect
619 Specifies the reselection policy for the primary slave. This
620 affects how the primary slave is chosen to become the active slave
621 when failure of the active slave or recovery of the primary slave
622 occurs. This option is designed to prevent flip-flopping between
623 the primary slave and other slaves. Possible values are:
625 always or 0 (default)
627 The primary slave becomes the active slave whenever it
628 comes back up.
630 better or 1
632 The primary slave becomes the active slave when it comes
633 back up, if the speed and duplex of the primary slave is
634 better than the speed and duplex of the current active
635 slave.
637 failure or 2
639 The primary slave becomes the active slave only if the
640 current active slave fails and the primary slave is up.
642 The primary_reselect setting is ignored in two cases:
644 If no slaves are active, the first slave to recover is
645 made the active slave.
647 When initially enslaved, the primary slave is always made
648 the active slave.
650 Changing the primary_reselect policy via sysfs will cause an
651 immediate selection of the best active slave according to the new
652 policy. This may or may not result in a change of the active
653 slave, depending upon the circumstances.
655 This option was added for bonding version 3.6.0.
657 updelay
659 Specifies the time, in milliseconds, to wait before enabling a
660 slave after a link recovery has been detected. This option is
661 only valid for the miimon link monitor. The updelay value
662 should be a multiple of the miimon value; if not, it will be
663 rounded down to the nearest multiple. The default value is 0.
665 use_carrier
667 Specifies whether or not miimon should use MII or ETHTOOL
668 ioctls vs. netif_carrier_ok() to determine the link
669 status. The MII or ETHTOOL ioctls are less efficient and
670 utilize a deprecated calling sequence within the kernel. The
671 netif_carrier_ok() relies on the device driver to maintain its
672 state with netif_carrier_on/off; at this writing, most, but
673 not all, device drivers support this facility.
675 If bonding insists that the link is up when it should not be,
676 it may be that your network device driver does not support
677 netif_carrier_on/off. The default state for netif_carrier is
678 "carrier on," so if a driver does not support netif_carrier,
679 it will appear as if the link is always up. In this case,
680 setting use_carrier to 0 will cause bonding to revert to the
681 MII / ETHTOOL ioctl method to determine the link state.
683 A value of 1 enables the use of netif_carrier_ok(), a value of
684 0 will use the deprecated MII / ETHTOOL ioctls. The default
685 value is 1.
687 xmit_hash_policy
689 Selects the transmit hash policy to use for slave selection in
690 balance-xor and 802.3ad modes. Possible values are:
692 layer2
694 Uses XOR of hardware MAC addresses to generate the
695 hash. The formula is
697 (source MAC XOR destination MAC) modulo slave count
699 This algorithm will place all traffic to a particular
700 network peer on the same slave.
702 This algorithm is 802.3ad compliant.
704 layer2+3
706 This policy uses a combination of layer2 and layer3
707 protocol information to generate the hash.
709 Uses XOR of hardware MAC addresses and IP addresses to
710 generate the hash. The formula is
712 (((source IP XOR dest IP) AND 0xffff) XOR
713 ( source MAC XOR destination MAC ))
714 modulo slave count
716 This algorithm will place all traffic to a particular
717 network peer on the same slave. For non-IP traffic,
718 the formula is the same as for the layer2 transmit
719 hash policy.
721 This policy is intended to provide a more balanced
722 distribution of traffic than layer2 alone, especially
723 in environments where a layer3 gateway device is
724 required to reach most destinations.
726 This algorithm is 802.3ad compliant.
728 layer3+4
730 This policy uses upper layer protocol information,
731 when available, to generate the hash. This allows for
732 traffic to a particular network peer to span multiple
733 slaves, although a single connection will not span
734 multiple slaves.
736 The formula for unfragmented TCP and UDP packets is
738 ((source port XOR dest port) XOR
739 ((source IP XOR dest IP) AND 0xffff)
740 modulo slave count
742 For fragmented TCP or UDP packets and all other IP
743 protocol traffic, the source and destination port
744 information is omitted. For non-IP traffic, the
745 formula is the same as for the layer2 transmit hash
746 policy.
748 This policy is intended to mimic the behavior of
749 certain switches, notably Cisco switches with PFC2 as
750 well as some Foundry and IBM products.
752 This algorithm is not fully 802.3ad compliant. A
753 single TCP or UDP conversation containing both
754 fragmented and unfragmented packets will see packets
755 striped across two interfaces. This may result in out
756 of order delivery. Most traffic types will not meet
757 this criteria, as TCP rarely fragments traffic, and
758 most UDP traffic is not involved in extended
759 conversations. Other implementations of 802.3ad may
760 or may not tolerate this noncompliance.
762 The default value is layer2. This option was added in bonding
763 version 2.6.3. In earlier versions of bonding, this parameter
764 does not exist, and the layer2 policy is the only policy. The
765 layer2+3 value was added for bonding version 3.2.2.
768 3. Configuring Bonding Devices
769 ==============================
771 You can configure bonding using either your distro's network
772 initialization scripts, or manually using either ifenslave or the
773 sysfs interface. Distros generally use one of two packages for the
774 network initialization scripts: initscripts or sysconfig. Recent
775 versions of these packages have support for bonding, while older
776 versions do not.
778 We will first describe the options for configuring bonding for
779 distros using versions of initscripts and sysconfig with full or
780 partial support for bonding, then provide information on enabling
781 bonding without support from the network initialization scripts (i.e.,
782 older versions of initscripts or sysconfig).
784 If you're unsure whether your distro uses sysconfig or
785 initscripts, or don't know if it's new enough, have no fear.
786 Determining this is fairly straightforward.
788 First, issue the command:
790 $ rpm -qf /sbin/ifup
792 It will respond with a line of text starting with either
793 "initscripts" or "sysconfig," followed by some numbers. This is the
794 package that provides your network initialization scripts.
796 Next, to determine if your installation supports bonding,
797 issue the command:
799 $ grep ifenslave /sbin/ifup
801 If this returns any matches, then your initscripts or
802 sysconfig has support for bonding.
804 3.1 Configuration with Sysconfig Support
805 ----------------------------------------
807 This section applies to distros using a version of sysconfig
808 with bonding support, for example, SuSE Linux Enterprise Server 9.
810 SuSE SLES 9's networking configuration system does support
811 bonding, however, at this writing, the YaST system configuration
812 front end does not provide any means to work with bonding devices.
813 Bonding devices can be managed by hand, however, as follows.
815 First, if they have not already been configured, configure the
816 slave devices. On SLES 9, this is most easily done by running the
817 yast2 sysconfig configuration utility. The goal is for to create an
818 ifcfg-id file for each slave device. The simplest way to accomplish
819 this is to configure the devices for DHCP (this is only to get the
820 file ifcfg-id file created; see below for some issues with DHCP). The
821 name of the configuration file for each device will be of the form:
823 ifcfg-id-xx:xx:xx:xx:xx:xx
825 Where the "xx" portion will be replaced with the digits from
826 the device's permanent MAC address.
828 Once the set of ifcfg-id-xx:xx:xx:xx:xx:xx files has been
829 created, it is necessary to edit the configuration files for the slave
830 devices (the MAC addresses correspond to those of the slave devices).
831 Before editing, the file will contain multiple lines, and will look
832 something like this:
834 BOOTPROTO='dhcp'
835 STARTMODE='on'
836 USERCTL='no'
837 UNIQUE='XNzu.WeZGOGF+4wE'
838 _nm_name='bus-pci-0001:61:01.0'
840 Change the BOOTPROTO and STARTMODE lines to the following:
842 BOOTPROTO='none'
843 STARTMODE='off'
845 Do not alter the UNIQUE or _nm_name lines. Remove any other
846 lines (USERCTL, etc).
848 Once the ifcfg-id-xx:xx:xx:xx:xx:xx files have been modified,
849 it's time to create the configuration file for the bonding device
850 itself. This file is named ifcfg-bondX, where X is the number of the
851 bonding device to create, starting at 0. The first such file is
852 ifcfg-bond0, the second is ifcfg-bond1, and so on. The sysconfig
853 network configuration system will correctly start multiple instances
854 of bonding.
856 The contents of the ifcfg-bondX file is as follows:
858 BOOTPROTO="static"
859 BROADCAST="10.0.2.255"
860 IPADDR="10.0.2.10"
861 NETMASK="255.255.0.0"
862 NETWORK="10.0.2.0"
863 REMOTE_IPADDR=""
864 STARTMODE="onboot"
865 BONDING_MASTER="yes"
866 BONDING_MODULE_OPTS="mode=active-backup miimon=100"
867 BONDING_SLAVE0="eth0"
868 BONDING_SLAVE1="bus-pci-0000:06:08.1"
870 Replace the sample BROADCAST, IPADDR, NETMASK and NETWORK
871 values with the appropriate values for your network.
873 The STARTMODE specifies when the device is brought online.
874 The possible values are:
876 onboot: The device is started at boot time. If you're not
877 sure, this is probably what you want.
879 manual: The device is started only when ifup is called
880 manually. Bonding devices may be configured this
881 way if you do not wish them to start automatically
882 at boot for some reason.
884 hotplug: The device is started by a hotplug event. This is not
885 a valid choice for a bonding device.
887 off or ignore: The device configuration is ignored.
889 The line BONDING_MASTER='yes' indicates that the device is a
890 bonding master device. The only useful value is "yes."
892 The contents of BONDING_MODULE_OPTS are supplied to the
893 instance of the bonding module for this device. Specify the options
894 for the bonding mode, link monitoring, and so on here. Do not include
895 the max_bonds bonding parameter; this will confuse the configuration
896 system if you have multiple bonding devices.
898 Finally, supply one BONDING_SLAVEn="slave device" for each
899 slave. where "n" is an increasing value, one for each slave. The
900 "slave device" is either an interface name, e.g., "eth0", or a device
901 specifier for the network device. The interface name is easier to
902 find, but the ethN names are subject to change at boot time if, e.g.,
903 a device early in the sequence has failed. The device specifiers
904 (bus-pci-0000:06:08.1 in the example above) specify the physical
905 network device, and will not change unless the device's bus location
906 changes (for example, it is moved from one PCI slot to another). The
907 example above uses one of each type for demonstration purposes; most
908 configurations will choose one or the other for all slave devices.
910 When all configuration files have been modified or created,
911 networking must be restarted for the configuration changes to take
912 effect. This can be accomplished via the following:
914 # /etc/init.d/network restart
916 Note that the network control script (/sbin/ifdown) will
917 remove the bonding module as part of the network shutdown processing,
918 so it is not necessary to remove the module by hand if, e.g., the
919 module parameters have changed.
921 Also, at this writing, YaST/YaST2 will not manage bonding
922 devices (they do not show bonding interfaces on its list of network
923 devices). It is necessary to edit the configuration file by hand to
924 change the bonding configuration.
926 Additional general options and details of the ifcfg file
927 format can be found in an example ifcfg template file:
929 /etc/sysconfig/network/ifcfg.template
931 Note that the template does not document the various BONDING_
932 settings described above, but does describe many of the other options.
934 3.1.1 Using DHCP with Sysconfig
935 -------------------------------
937 Under sysconfig, configuring a device with BOOTPROTO='dhcp'
938 will cause it to query DHCP for its IP address information. At this
939 writing, this does not function for bonding devices; the scripts
940 attempt to obtain the device address from DHCP prior to adding any of
941 the slave devices. Without active slaves, the DHCP requests are not
942 sent to the network.
944 3.1.2 Configuring Multiple Bonds with Sysconfig
945 -----------------------------------------------
947 The sysconfig network initialization system is capable of
948 handling multiple bonding devices. All that is necessary is for each
949 bonding instance to have an appropriately configured ifcfg-bondX file
950 (as described above). Do not specify the "max_bonds" parameter to any
951 instance of bonding, as this will confuse sysconfig. If you require
952 multiple bonding devices with identical parameters, create multiple
953 ifcfg-bondX files.
955 Because the sysconfig scripts supply the bonding module
956 options in the ifcfg-bondX file, it is not necessary to add them to
957 the system /etc/modules.conf or /etc/modprobe.conf configuration file.
959 3.2 Configuration with Initscripts Support
960 ------------------------------------------
962 This section applies to distros using a recent version of
963 initscripts with bonding support, for example, Red Hat Enterprise Linux
964 version 3 or later, Fedora, etc. On these systems, the network
965 initialization scripts have knowledge of bonding, and can be configured to
966 control bonding devices. Note that older versions of the initscripts
967 package have lower levels of support for bonding; this will be noted where
968 applicable.
970 These distros will not automatically load the network adapter
971 driver unless the ethX device is configured with an IP address.
972 Because of this constraint, users must manually configure a
973 network-script file for all physical adapters that will be members of
974 a bondX link. Network script files are located in the directory:
976 /etc/sysconfig/network-scripts
978 The file name must be prefixed with "ifcfg-eth" and suffixed
979 with the adapter's physical adapter number. For example, the script
980 for eth0 would be named /etc/sysconfig/network-scripts/ifcfg-eth0.
981 Place the following text in the file:
983 DEVICE=eth0
984 USERCTL=no
985 ONBOOT=yes
986 MASTER=bond0
987 SLAVE=yes
988 BOOTPROTO=none
990 The DEVICE= line will be different for every ethX device and
991 must correspond with the name of the file, i.e., ifcfg-eth1 must have
992 a device line of DEVICE=eth1. The setting of the MASTER= line will
993 also depend on the final bonding interface name chosen for your bond.
994 As with other network devices, these typically start at 0, and go up
995 one for each device, i.e., the first bonding instance is bond0, the
996 second is bond1, and so on.
998 Next, create a bond network script. The file name for this
999 script will be /etc/sysconfig/network-scripts/ifcfg-bondX where X is
1000 the number of the bond. For bond0 the file is named "ifcfg-bond0",
1001 for bond1 it is named "ifcfg-bond1", and so on. Within that file,
1002 place the following text:
1004 DEVICE=bond0
1005 IPADDR=192.168.1.1
1006 NETMASK=255.255.255.0
1007 NETWORK=192.168.1.0
1008 BROADCAST=192.168.1.255
1009 ONBOOT=yes
1010 BOOTPROTO=none
1011 USERCTL=no
1013 Be sure to change the networking specific lines (IPADDR,
1014 NETMASK, NETWORK and BROADCAST) to match your network configuration.
1016 For later versions of initscripts, such as that found with Fedora
1017 7 (or later) and Red Hat Enterprise Linux version 5 (or later), it is possible,
1018 and, indeed, preferable, to specify the bonding options in the ifcfg-bond0
1019 file, e.g. a line of the format:
1021 BONDING_OPTS="mode=active-backup arp_interval=60 arp_ip_target=192.168.1.254"
1023 will configure the bond with the specified options. The options
1024 specified in BONDING_OPTS are identical to the bonding module parameters
1025 except for the arp_ip_target field when using versions of initscripts older
1026 than and 8.57 (Fedora 8) and 8.45.19 (Red Hat Enterprise Linux 5.2). When
1027 using older versions each target should be included as a separate option and
1028 should be preceded by a '+' to indicate it should be added to the list of
1029 queried targets, e.g.,
1031 arp_ip_target=+192.168.1.1 arp_ip_target=+192.168.1.2
1033 is the proper syntax to specify multiple targets. When specifying
1034 options via BONDING_OPTS, it is not necessary to edit /etc/modules.conf or
1035 /etc/modprobe.conf.
1037 For even older versions of initscripts that do not support
1038 BONDING_OPTS, it is necessary to edit /etc/modules.conf (or
1039 /etc/modprobe.conf, depending upon your distro) to load the bonding module
1040 with your desired options when the bond0 interface is brought up. The
1041 following lines in /etc/modules.conf (or modprobe.conf) will load the
1042 bonding module, and select its options:
1044 alias bond0 bonding
1045 options bond0 mode=balance-alb miimon=100
1047 Replace the sample parameters with the appropriate set of
1048 options for your configuration.
1050 Finally run "/etc/rc.d/init.d/network restart" as root. This
1051 will restart the networking subsystem and your bond link should be now
1052 up and running.
1054 3.2.1 Using DHCP with Initscripts
1055 ---------------------------------
1057 Recent versions of initscripts (the versions supplied with Fedora
1058 Core 3 and Red Hat Enterprise Linux 4, or later versions, are reported to
1059 work) have support for assigning IP information to bonding devices via
1060 DHCP.
1062 To configure bonding for DHCP, configure it as described
1063 above, except replace the line "BOOTPROTO=none" with "BOOTPROTO=dhcp"
1064 and add a line consisting of "TYPE=Bonding". Note that the TYPE value
1065 is case sensitive.
1067 3.2.2 Configuring Multiple Bonds with Initscripts
1068 -------------------------------------------------
1070 Initscripts packages that are included with Fedora 7 and Red Hat
1071 Enterprise Linux 5 support multiple bonding interfaces by simply
1072 specifying the appropriate BONDING_OPTS= in ifcfg-bondX where X is the
1073 number of the bond. This support requires sysfs support in the kernel,
1074 and a bonding driver of version 3.0.0 or later. Other configurations may
1075 not support this method for specifying multiple bonding interfaces; for
1076 those instances, see the "Configuring Multiple Bonds Manually" section,
1077 below.
1079 3.3 Configuring Bonding Manually with Ifenslave
1080 -----------------------------------------------
1082 This section applies to distros whose network initialization
1083 scripts (the sysconfig or initscripts package) do not have specific
1084 knowledge of bonding. One such distro is SuSE Linux Enterprise Server
1085 version 8.
1087 The general method for these systems is to place the bonding
1088 module parameters into /etc/modules.conf or /etc/modprobe.conf (as
1089 appropriate for the installed distro), then add modprobe and/or
1090 ifenslave commands to the system's global init script. The name of
1091 the global init script differs; for sysconfig, it is
1092 /etc/init.d/boot.local and for initscripts it is /etc/rc.d/rc.local.
1094 For example, if you wanted to make a simple bond of two e100
1095 devices (presumed to be eth0 and eth1), and have it persist across
1096 reboots, edit the appropriate file (/etc/init.d/boot.local or
1097 /etc/rc.d/rc.local), and add the following:
1099 modprobe bonding mode=balance-alb miimon=100
1100 modprobe e100
1101 ifconfig bond0 192.168.1.1 netmask 255.255.255.0 up
1102 ifenslave bond0 eth0
1103 ifenslave bond0 eth1
1105 Replace the example bonding module parameters and bond0
1106 network configuration (IP address, netmask, etc) with the appropriate
1107 values for your configuration.
1109 Unfortunately, this method will not provide support for the
1110 ifup and ifdown scripts on the bond devices. To reload the bonding
1111 configuration, it is necessary to run the initialization script, e.g.,
1113 # /etc/init.d/boot.local
1115 or
1117 # /etc/rc.d/rc.local
1119 It may be desirable in such a case to create a separate script
1120 which only initializes the bonding configuration, then call that
1121 separate script from within boot.local. This allows for bonding to be
1122 enabled without re-running the entire global init script.
1124 To shut down the bonding devices, it is necessary to first
1125 mark the bonding device itself as being down, then remove the
1126 appropriate device driver modules. For our example above, you can do
1127 the following:
1129 # ifconfig bond0 down
1130 # rmmod bonding
1131 # rmmod e100
1133 Again, for convenience, it may be desirable to create a script
1134 with these commands.
1137 3.3.1 Configuring Multiple Bonds Manually
1138 -----------------------------------------
1140 This section contains information on configuring multiple
1141 bonding devices with differing options for those systems whose network
1142 initialization scripts lack support for configuring multiple bonds.
1144 If you require multiple bonding devices, but all with the same
1145 options, you may wish to use the "max_bonds" module parameter,
1146 documented above.
1148 To create multiple bonding devices with differing options, it is
1149 preferrable to use bonding parameters exported by sysfs, documented in the
1150 section below.
1152 For versions of bonding without sysfs support, the only means to
1153 provide multiple instances of bonding with differing options is to load
1154 the bonding driver multiple times. Note that current versions of the
1155 sysconfig network initialization scripts handle this automatically; if
1156 your distro uses these scripts, no special action is needed. See the
1157 section Configuring Bonding Devices, above, if you're not sure about your
1158 network initialization scripts.
1160 To load multiple instances of the module, it is necessary to
1161 specify a different name for each instance (the module loading system
1162 requires that every loaded module, even multiple instances of the same
1163 module, have a unique name). This is accomplished by supplying multiple
1164 sets of bonding options in /etc/modprobe.conf, for example:
1166 alias bond0 bonding
1167 options bond0 -o bond0 mode=balance-rr miimon=100
1169 alias bond1 bonding
1170 options bond1 -o bond1 mode=balance-alb miimon=50
1172 will load the bonding module two times. The first instance is
1173 named "bond0" and creates the bond0 device in balance-rr mode with an
1174 miimon of 100. The second instance is named "bond1" and creates the
1175 bond1 device in balance-alb mode with an miimon of 50.
1177 In some circumstances (typically with older distributions),
1178 the above does not work, and the second bonding instance never sees
1179 its options. In that case, the second options line can be substituted
1180 as follows:
1182 install bond1 /sbin/modprobe --ignore-install bonding -o bond1 \
1183 mode=balance-alb miimon=50
1185 This may be repeated any number of times, specifying a new and
1186 unique name in place of bond1 for each subsequent instance.
1188 It has been observed that some Red Hat supplied kernels are unable
1189 to rename modules at load time (the "-o bond1" part). Attempts to pass
1190 that option to modprobe will produce an "Operation not permitted" error.
1191 This has been reported on some Fedora Core kernels, and has been seen on
1192 RHEL 4 as well. On kernels exhibiting this problem, it will be impossible
1193 to configure multiple bonds with differing parameters (as they are older
1194 kernels, and also lack sysfs support).
1196 3.4 Configuring Bonding Manually via Sysfs
1197 ------------------------------------------
1199 Starting with version 3.0.0, Channel Bonding may be configured
1200 via the sysfs interface. This interface allows dynamic configuration
1201 of all bonds in the system without unloading the module. It also
1202 allows for adding and removing bonds at runtime. Ifenslave is no
1203 longer required, though it is still supported.
1205 Use of the sysfs interface allows you to use multiple bonds
1206 with different configurations without having to reload the module.
1207 It also allows you to use multiple, differently configured bonds when
1208 bonding is compiled into the kernel.
1210 You must have the sysfs filesystem mounted to configure
1211 bonding this way. The examples in this document assume that you
1212 are using the standard mount point for sysfs, e.g. /sys. If your
1213 sysfs filesystem is mounted elsewhere, you will need to adjust the
1214 example paths accordingly.
1216 Creating and Destroying Bonds
1217 -----------------------------
1218 To add a new bond foo:
1219 # echo +foo > /sys/class/net/bonding_masters
1221 To remove an existing bond bar:
1222 # echo -bar > /sys/class/net/bonding_masters
1224 To show all existing bonds:
1225 # cat /sys/class/net/bonding_masters
1227 NOTE: due to 4K size limitation of sysfs files, this list may be
1228 truncated if you have more than a few hundred bonds. This is unlikely
1229 to occur under normal operating conditions.
1231 Adding and Removing Slaves
1232 --------------------------
1233 Interfaces may be enslaved to a bond using the file
1234 /sys/class/net/<bond>/bonding/slaves. The semantics for this file
1235 are the same as for the bonding_masters file.
1237 To enslave interface eth0 to bond bond0:
1238 # ifconfig bond0 up
1239 # echo +eth0 > /sys/class/net/bond0/bonding/slaves
1241 To free slave eth0 from bond bond0:
1242 # echo -eth0 > /sys/class/net/bond0/bonding/slaves
1244 When an interface is enslaved to a bond, symlinks between the
1245 two are created in the sysfs filesystem. In this case, you would get
1246 /sys/class/net/bond0/slave_eth0 pointing to /sys/class/net/eth0, and
1247 /sys/class/net/eth0/master pointing to /sys/class/net/bond0.
1249 This means that you can tell quickly whether or not an
1250 interface is enslaved by looking for the master symlink. Thus:
1251 # echo -eth0 > /sys/class/net/eth0/master/bonding/slaves
1252 will free eth0 from whatever bond it is enslaved to, regardless of
1253 the name of the bond interface.
1255 Changing a Bond's Configuration
1256 -------------------------------
1257 Each bond may be configured individually by manipulating the
1258 files located in /sys/class/net/<bond name>/bonding
1260 The names of these files correspond directly with the command-
1261 line parameters described elsewhere in this file, and, with the
1262 exception of arp_ip_target, they accept the same values. To see the
1263 current setting, simply cat the appropriate file.
1265 A few examples will be given here; for specific usage
1266 guidelines for each parameter, see the appropriate section in this
1267 document.
1269 To configure bond0 for balance-alb mode:
1270 # ifconfig bond0 down
1271 # echo 6 > /sys/class/net/bond0/bonding/mode
1272 - or -
1273 # echo balance-alb > /sys/class/net/bond0/bonding/mode
1274 NOTE: The bond interface must be down before the mode can be
1275 changed.
1277 To enable MII monitoring on bond0 with a 1 second interval:
1278 # echo 1000 > /sys/class/net/bond0/bonding/miimon
1279 NOTE: If ARP monitoring is enabled, it will disabled when MII
1280 monitoring is enabled, and vice-versa.
1282 To add ARP targets:
1283 # echo +192.168.0.100 > /sys/class/net/bond0/bonding/arp_ip_target
1284 # echo +192.168.0.101 > /sys/class/net/bond0/bonding/arp_ip_target
1285 NOTE: up to 16 target addresses may be specified.
1287 To remove an ARP target:
1288 # echo -192.168.0.100 > /sys/class/net/bond0/bonding/arp_ip_target
1290 Example Configuration
1291 ---------------------
1292 We begin with the same example that is shown in section 3.3,
1293 executed with sysfs, and without using ifenslave.
1295 To make a simple bond of two e100 devices (presumed to be eth0
1296 and eth1), and have it persist across reboots, edit the appropriate
1297 file (/etc/init.d/boot.local or /etc/rc.d/rc.local), and add the
1298 following:
1300 modprobe bonding
1301 modprobe e100
1302 echo balance-alb > /sys/class/net/bond0/bonding/mode
1303 ifconfig bond0 192.168.1.1 netmask 255.255.255.0 up
1304 echo 100 > /sys/class/net/bond0/bonding/miimon
1305 echo +eth0 > /sys/class/net/bond0/bonding/slaves
1306 echo +eth1 > /sys/class/net/bond0/bonding/slaves
1308 To add a second bond, with two e1000 interfaces in
1309 active-backup mode, using ARP monitoring, add the following lines to
1310 your init script:
1312 modprobe e1000
1313 echo +bond1 > /sys/class/net/bonding_masters
1314 echo active-backup > /sys/class/net/bond1/bonding/mode
1315 ifconfig bond1 192.168.2.1 netmask 255.255.255.0 up
1316 echo +192.168.2.100 /sys/class/net/bond1/bonding/arp_ip_target
1317 echo 2000 > /sys/class/net/bond1/bonding/arp_interval
1318 echo +eth2 > /sys/class/net/bond1/bonding/slaves
1319 echo +eth3 > /sys/class/net/bond1/bonding/slaves
1322 4. Querying Bonding Configuration
1323 =================================
1325 4.1 Bonding Configuration
1326 -------------------------
1328 Each bonding device has a read-only file residing in the
1329 /proc/net/bonding directory. The file contents include information
1330 about the bonding configuration, options and state of each slave.
1332 For example, the contents of /proc/net/bonding/bond0 after the
1333 driver is loaded with parameters of mode=0 and miimon=1000 is
1334 generally as follows:
1336 Ethernet Channel Bonding Driver: 2.6.1 (October 29, 2004)
1337 Bonding Mode: load balancing (round-robin)
1338 Currently Active Slave: eth0
1339 MII Status: up
1340 MII Polling Interval (ms): 1000
1341 Up Delay (ms): 0
1342 Down Delay (ms): 0
1344 Slave Interface: eth1
1345 MII Status: up
1346 Link Failure Count: 1
1348 Slave Interface: eth0
1349 MII Status: up
1350 Link Failure Count: 1
1352 The precise format and contents will change depending upon the
1353 bonding configuration, state, and version of the bonding driver.
1355 4.2 Network configuration
1356 -------------------------
1358 The network configuration can be inspected using the ifconfig
1359 command. Bonding devices will have the MASTER flag set; Bonding slave
1360 devices will have the SLAVE flag set. The ifconfig output does not
1361 contain information on which slaves are associated with which masters.
1363 In the example below, the bond0 interface is the master
1364 (MASTER) while eth0 and eth1 are slaves (SLAVE). Notice all slaves of
1365 bond0 have the same MAC address (HWaddr) as bond0 for all modes except
1366 TLB and ALB that require a unique MAC address for each slave.
1368 # /sbin/ifconfig
1369 bond0 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1370 inet addr:XXX.XXX.XXX.YYY Bcast:XXX.XXX.XXX.255 Mask:255.255.252.0
1371 UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
1372 RX packets:7224794 errors:0 dropped:0 overruns:0 frame:0
1373 TX packets:3286647 errors:1 dropped:0 overruns:1 carrier:0
1374 collisions:0 txqueuelen:0
1376 eth0 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1377 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
1378 RX packets:3573025 errors:0 dropped:0 overruns:0 frame:0
1379 TX packets:1643167 errors:1 dropped:0 overruns:1 carrier:0
1380 collisions:0 txqueuelen:100
1381 Interrupt:10 Base address:0x1080
1383 eth1 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1384 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
1385 RX packets:3651769 errors:0 dropped:0 overruns:0 frame:0
1386 TX packets:1643480 errors:0 dropped:0 overruns:0 carrier:0
1387 collisions:0 txqueuelen:100
1388 Interrupt:9 Base address:0x1400
1390 5. Switch Configuration
1391 =======================
1393 For this section, "switch" refers to whatever system the
1394 bonded devices are directly connected to (i.e., where the other end of
1395 the cable plugs into). This may be an actual dedicated switch device,
1396 or it may be another regular system (e.g., another computer running
1397 Linux),
1399 The active-backup, balance-tlb and balance-alb modes do not
1400 require any specific configuration of the switch.
1402 The 802.3ad mode requires that the switch have the appropriate
1403 ports configured as an 802.3ad aggregation. The precise method used
1404 to configure this varies from switch to switch, but, for example, a
1405 Cisco 3550 series switch requires that the appropriate ports first be
1406 grouped together in a single etherchannel instance, then that
1407 etherchannel is set to mode "lacp" to enable 802.3ad (instead of
1408 standard EtherChannel).
1410 The balance-rr, balance-xor and broadcast modes generally
1411 require that the switch have the appropriate ports grouped together.
1412 The nomenclature for such a group differs between switches, it may be
1413 called an "etherchannel" (as in the Cisco example, above), a "trunk
1414 group" or some other similar variation. For these modes, each switch
1415 will also have its own configuration options for the switch's transmit
1416 policy to the bond. Typical choices include XOR of either the MAC or
1417 IP addresses. The transmit policy of the two peers does not need to
1418 match. For these three modes, the bonding mode really selects a
1419 transmit policy for an EtherChannel group; all three will interoperate
1420 with another EtherChannel group.
1423 6. 802.1q VLAN Support
1424 ======================
1426 It is possible to configure VLAN devices over a bond interface
1427 using the 8021q driver. However, only packets coming from the 8021q
1428 driver and passing through bonding will be tagged by default. Self
1429 generated packets, for example, bonding's learning packets or ARP
1430 packets generated by either ALB mode or the ARP monitor mechanism, are
1431 tagged internally by bonding itself. As a result, bonding must
1432 "learn" the VLAN IDs configured above it, and use those IDs to tag
1433 self generated packets.
1435 For reasons of simplicity, and to support the use of adapters
1436 that can do VLAN hardware acceleration offloading, the bonding
1437 interface declares itself as fully hardware offloading capable, it gets
1438 the add_vid/kill_vid notifications to gather the necessary
1439 information, and it propagates those actions to the slaves. In case
1440 of mixed adapter types, hardware accelerated tagged packets that
1441 should go through an adapter that is not offloading capable are
1442 "un-accelerated" by the bonding driver so the VLAN tag sits in the
1443 regular location.
1445 VLAN interfaces *must* be added on top of a bonding interface
1446 only after enslaving at least one slave. The bonding interface has a
1447 hardware address of 00:00:00:00:00:00 until the first slave is added.
1448 If the VLAN interface is created prior to the first enslavement, it
1449 would pick up the all-zeroes hardware address. Once the first slave
1450 is attached to the bond, the bond device itself will pick up the
1451 slave's hardware address, which is then available for the VLAN device.
1453 Also, be aware that a similar problem can occur if all slaves
1454 are released from a bond that still has one or more VLAN interfaces on
1455 top of it. When a new slave is added, the bonding interface will
1456 obtain its hardware address from the first slave, which might not
1457 match the hardware address of the VLAN interfaces (which was
1458 ultimately copied from an earlier slave).
1460 There are two methods to insure that the VLAN device operates
1461 with the correct hardware address if all slaves are removed from a
1462 bond interface:
1464 1. Remove all VLAN interfaces then recreate them
1466 2. Set the bonding interface's hardware address so that it
1467 matches the hardware address of the VLAN interfaces.
1469 Note that changing a VLAN interface's HW address would set the
1470 underlying device -- i.e. the bonding interface -- to promiscuous
1471 mode, which might not be what you want.
1474 7. Link Monitoring
1475 ==================
1477 The bonding driver at present supports two schemes for
1478 monitoring a slave device's link state: the ARP monitor and the MII
1479 monitor.
1481 At the present time, due to implementation restrictions in the
1482 bonding driver itself, it is not possible to enable both ARP and MII
1483 monitoring simultaneously.
1485 7.1 ARP Monitor Operation
1486 -------------------------
1488 The ARP monitor operates as its name suggests: it sends ARP
1489 queries to one or more designated peer systems on the network, and
1490 uses the response as an indication that the link is operating. This
1491 gives some assurance that traffic is actually flowing to and from one
1492 or more peers on the local network.
1494 The ARP monitor relies on the device driver itself to verify
1495 that traffic is flowing. In particular, the driver must keep up to
1496 date the last receive time, dev->last_rx, and transmit start time,
1497 dev->trans_start. If these are not updated by the driver, then the
1498 ARP monitor will immediately fail any slaves using that driver, and
1499 those slaves will stay down. If networking monitoring (tcpdump, etc)
1500 shows the ARP requests and replies on the network, then it may be that
1501 your device driver is not updating last_rx and trans_start.
1503 7.2 Configuring Multiple ARP Targets
1504 ------------------------------------
1506 While ARP monitoring can be done with just one target, it can
1507 be useful in a High Availability setup to have several targets to
1508 monitor. In the case of just one target, the target itself may go
1509 down or have a problem making it unresponsive to ARP requests. Having
1510 an additional target (or several) increases the reliability of the ARP
1511 monitoring.
1513 Multiple ARP targets must be separated by commas as follows:
1515 # example options for ARP monitoring with three targets
1516 alias bond0 bonding
1517 options bond0 arp_interval=60 arp_ip_target=192.168.0.1,192.168.0.3,192.168.0.9
1519 For just a single target the options would resemble:
1521 # example options for ARP monitoring with one target
1522 alias bond0 bonding
1523 options bond0 arp_interval=60 arp_ip_target=192.168.0.100
1526 7.3 MII Monitor Operation
1527 -------------------------
1529 The MII monitor monitors only the carrier state of the local
1530 network interface. It accomplishes this in one of three ways: by
1531 depending upon the device driver to maintain its carrier state, by
1532 querying the device's MII registers, or by making an ethtool query to
1533 the device.
1535 If the use_carrier module parameter is 1 (the default value),
1536 then the MII monitor will rely on the driver for carrier state
1537 information (via the netif_carrier subsystem). As explained in the
1538 use_carrier parameter information, above, if the MII monitor fails to
1539 detect carrier loss on the device (e.g., when the cable is physically
1540 disconnected), it may be that the driver does not support
1541 netif_carrier.
1543 If use_carrier is 0, then the MII monitor will first query the
1544 device's (via ioctl) MII registers and check the link state. If that
1545 request fails (not just that it returns carrier down), then the MII
1546 monitor will make an ethtool ETHOOL_GLINK request to attempt to obtain
1547 the same information. If both methods fail (i.e., the driver either
1548 does not support or had some error in processing both the MII register
1549 and ethtool requests), then the MII monitor will assume the link is
1550 up.
1552 8. Potential Sources of Trouble
1553 ===============================
1555 8.1 Adventures in Routing
1556 -------------------------
1558 When bonding is configured, it is important that the slave
1559 devices not have routes that supersede routes of the master (or,
1560 generally, not have routes at all). For example, suppose the bonding
1561 device bond0 has two slaves, eth0 and eth1, and the routing table is
1562 as follows:
1564 Kernel IP routing table
1565 Destination Gateway Genmask Flags MSS Window irtt Iface
1566 10.0.0.0 0.0.0.0 255.255.0.0 U 40 0 0 eth0
1567 10.0.0.0 0.0.0.0 255.255.0.0 U 40 0 0 eth1
1568 10.0.0.0 0.0.0.0 255.255.0.0 U 40 0 0 bond0
1569 127.0.0.0 0.0.0.0 255.0.0.0 U 40 0 0 lo
1571 This routing configuration will likely still update the
1572 receive/transmit times in the driver (needed by the ARP monitor), but
1573 may bypass the bonding driver (because outgoing traffic to, in this
1574 case, another host on network 10 would use eth0 or eth1 before bond0).
1576 The ARP monitor (and ARP itself) may become confused by this
1577 configuration, because ARP requests (generated by the ARP monitor)
1578 will be sent on one interface (bond0), but the corresponding reply
1579 will arrive on a different interface (eth0). This reply looks to ARP
1580 as an unsolicited ARP reply (because ARP matches replies on an
1581 interface basis), and is discarded. The MII monitor is not affected
1582 by the state of the routing table.
1584 The solution here is simply to insure that slaves do not have
1585 routes of their own, and if for some reason they must, those routes do
1586 not supersede routes of their master. This should generally be the
1587 case, but unusual configurations or errant manual or automatic static
1588 route additions may cause trouble.
1590 8.2 Ethernet Device Renaming
1591 ----------------------------
1593 On systems with network configuration scripts that do not
1594 associate physical devices directly with network interface names (so
1595 that the same physical device always has the same "ethX" name), it may
1596 be necessary to add some special logic to either /etc/modules.conf or
1597 /etc/modprobe.conf (depending upon which is installed on the system).
1599 For example, given a modules.conf containing the following:
1601 alias bond0 bonding
1602 options bond0 mode=some-mode miimon=50
1603 alias eth0 tg3
1604 alias eth1 tg3
1605 alias eth2 e1000
1606 alias eth3 e1000
1608 If neither eth0 and eth1 are slaves to bond0, then when the
1609 bond0 interface comes up, the devices may end up reordered. This
1610 happens because bonding is loaded first, then its slave device's
1611 drivers are loaded next. Since no other drivers have been loaded,
1612 when the e1000 driver loads, it will receive eth0 and eth1 for its
1613 devices, but the bonding configuration tries to enslave eth2 and eth3
1614 (which may later be assigned to the tg3 devices).
1616 Adding the following:
1618 add above bonding e1000 tg3
1620 causes modprobe to load e1000 then tg3, in that order, when
1621 bonding is loaded. This command is fully documented in the
1622 modules.conf manual page.
1624 On systems utilizing modprobe.conf (or modprobe.conf.local),
1625 an equivalent problem can occur. In this case, the following can be
1626 added to modprobe.conf (or modprobe.conf.local, as appropriate), as
1627 follows (all on one line; it has been split here for clarity):
1629 install bonding /sbin/modprobe tg3; /sbin/modprobe e1000;
1630 /sbin/modprobe --ignore-install bonding
1632 This will, when loading the bonding module, rather than
1633 performing the normal action, instead execute the provided command.
1634 This command loads the device drivers in the order needed, then calls
1635 modprobe with --ignore-install to cause the normal action to then take
1636 place. Full documentation on this can be found in the modprobe.conf
1637 and modprobe manual pages.
1639 8.3. Painfully Slow Or No Failed Link Detection By Miimon
1640 ---------------------------------------------------------
1642 By default, bonding enables the use_carrier option, which
1643 instructs bonding to trust the driver to maintain carrier state.
1645 As discussed in the options section, above, some drivers do
1646 not support the netif_carrier_on/_off link state tracking system.
1647 With use_carrier enabled, bonding will always see these links as up,
1648 regardless of their actual state.
1650 Additionally, other drivers do support netif_carrier, but do
1651 not maintain it in real time, e.g., only polling the link state at
1652 some fixed interval. In this case, miimon will detect failures, but
1653 only after some long period of time has expired. If it appears that
1654 miimon is very slow in detecting link failures, try specifying
1655 use_carrier=0 to see if that improves the failure detection time. If
1656 it does, then it may be that the driver checks the carrier state at a
1657 fixed interval, but does not cache the MII register values (so the
1658 use_carrier=0 method of querying the registers directly works). If
1659 use_carrier=0 does not improve the failover, then the driver may cache
1660 the registers, or the problem may be elsewhere.
1662 Also, remember that miimon only checks for the device's
1663 carrier state. It has no way to determine the state of devices on or
1664 beyond other ports of a switch, or if a switch is refusing to pass
1665 traffic while still maintaining carrier on.
1667 9. SNMP agents
1668 ===============
1670 If running SNMP agents, the bonding driver should be loaded
1671 before any network drivers participating in a bond. This requirement
1672 is due to the interface index (ipAdEntIfIndex) being associated to
1673 the first interface found with a given IP address. That is, there is
1674 only one ipAdEntIfIndex for each IP address. For example, if eth0 and
1675 eth1 are slaves of bond0 and the driver for eth0 is loaded before the
1676 bonding driver, the interface for the IP address will be associated
1677 with the eth0 interface. This configuration is shown below, the IP
1678 address 192.168.1.1 has an interface index of 2 which indexes to eth0
1679 in the ifDescr table (ifDescr.2).
1681 interfaces.ifTable.ifEntry.ifDescr.1 = lo
1682 interfaces.ifTable.ifEntry.ifDescr.2 = eth0
1683 interfaces.ifTable.ifEntry.ifDescr.3 = eth1
1684 interfaces.ifTable.ifEntry.ifDescr.4 = eth2
1685 interfaces.ifTable.ifEntry.ifDescr.5 = eth3
1686 interfaces.ifTable.ifEntry.ifDescr.6 = bond0
1687 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.10.10.10.10 = 5
1688 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.192.168.1.1 = 2
1689 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.10.74.20.94 = 4
1690 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.127.0.0.1 = 1
1692 This problem is avoided by loading the bonding driver before
1693 any network drivers participating in a bond. Below is an example of
1694 loading the bonding driver first, the IP address 192.168.1.1 is
1695 correctly associated with ifDescr.2.
1697 interfaces.ifTable.ifEntry.ifDescr.1 = lo
1698 interfaces.ifTable.ifEntry.ifDescr.2 = bond0
1699 interfaces.ifTable.ifEntry.ifDescr.3 = eth0
1700 interfaces.ifTable.ifEntry.ifDescr.4 = eth1
1701 interfaces.ifTable.ifEntry.ifDescr.5 = eth2
1702 interfaces.ifTable.ifEntry.ifDescr.6 = eth3
1703 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.10.10.10.10 = 6
1704 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.192.168.1.1 = 2
1705 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.10.74.20.94 = 5
1706 ip.ipAddrTable.ipAddrEntry.ipAdEntIfIndex.127.0.0.1 = 1
1708 While some distributions may not report the interface name in
1709 ifDescr, the association between the IP address and IfIndex remains
1710 and SNMP functions such as Interface_Scan_Next will report that
1711 association.
1713 10. Promiscuous mode
1714 ====================
1716 When running network monitoring tools, e.g., tcpdump, it is
1717 common to enable promiscuous mode on the device, so that all traffic
1718 is seen (instead of seeing only traffic destined for the local host).
1719 The bonding driver handles promiscuous mode changes to the bonding
1720 master device (e.g., bond0), and propagates the setting to the slave
1721 devices.
1723 For the balance-rr, balance-xor, broadcast, and 802.3ad modes,
1724 the promiscuous mode setting is propagated to all slaves.
1726 For the active-backup, balance-tlb and balance-alb modes, the
1727 promiscuous mode setting is propagated only to the active slave.
1729 For balance-tlb mode, the active slave is the slave currently
1730 receiving inbound traffic.
1732 For balance-alb mode, the active slave is the slave used as a
1733 "primary." This slave is used for mode-specific control traffic, for
1734 sending to peers that are unassigned or if the load is unbalanced.
1736 For the active-backup, balance-tlb and balance-alb modes, when
1737 the active slave changes (e.g., due to a link failure), the
1738 promiscuous setting will be propagated to the new active slave.
1740 11. Configuring Bonding for High Availability
1741 =============================================
1743 High Availability refers to configurations that provide
1744 maximum network availability by having redundant or backup devices,
1745 links or switches between the host and the rest of the world. The
1746 goal is to provide the maximum availability of network connectivity
1747 (i.e., the network always works), even though other configurations
1748 could provide higher throughput.
1750 11.1 High Availability in a Single Switch Topology
1751 --------------------------------------------------
1753 If two hosts (or a host and a single switch) are directly
1754 connected via multiple physical links, then there is no availability
1755 penalty to optimizing for maximum bandwidth. In this case, there is
1756 only one switch (or peer), so if it fails, there is no alternative
1757 access to fail over to. Additionally, the bonding load balance modes
1758 support link monitoring of their members, so if individual links fail,
1759 the load will be rebalanced across the remaining devices.
1761 See Section 13, "Configuring Bonding for Maximum Throughput"
1762 for information on configuring bonding with one peer device.
1764 11.2 High Availability in a Multiple Switch Topology
1765 ----------------------------------------------------
1767 With multiple switches, the configuration of bonding and the
1768 network changes dramatically. In multiple switch topologies, there is
1769 a trade off between network availability and usable bandwidth.
1771 Below is a sample network, configured to maximize the
1772 availability of the network:
1774 | |
1775 |port3 port3|
1776 +-----+----+ +-----+----+
1777 | |port2 ISL port2| |
1778 | switch A +--------------------------+ switch B |
1779 | | | |
1780 +-----+----+ +-----++---+
1781 |port1 port1|
1782 | +-------+ |
1783 +-------------+ host1 +---------------+
1784 eth0 +-------+ eth1
1786 In this configuration, there is a link between the two
1787 switches (ISL, or inter switch link), and multiple ports connecting to
1788 the outside world ("port3" on each switch). There is no technical
1789 reason that this could not be extended to a third switch.
1791 11.2.1 HA Bonding Mode Selection for Multiple Switch Topology
1792 -------------------------------------------------------------
1794 In a topology such as the example above, the active-backup and
1795 broadcast modes are the only useful bonding modes when optimizing for
1796 availability; the other modes require all links to terminate on the
1797 same peer for them to behave rationally.
1799 active-backup: This is generally the preferred mode, particularly if
1800 the switches have an ISL and play together well. If the
1801 network configuration is such that one switch is specifically
1802 a backup switch (e.g., has lower capacity, higher cost, etc),
1803 then the primary option can be used to insure that the
1804 preferred link is always used when it is available.
1806 broadcast: This mode is really a special purpose mode, and is suitable
1807 only for very specific needs. For example, if the two
1808 switches are not connected (no ISL), and the networks beyond
1809 them are totally independent. In this case, if it is
1810 necessary for some specific one-way traffic to reach both
1811 independent networks, then the broadcast mode may be suitable.
1813 11.2.2 HA Link Monitoring Selection for Multiple Switch Topology
1814 ----------------------------------------------------------------
1816 The choice of link monitoring ultimately depends upon your
1817 switch. If the switch can reliably fail ports in response to other
1818 failures, then either the MII or ARP monitors should work. For
1819 example, in the above example, if the "port3" link fails at the remote
1820 end, the MII monitor has no direct means to detect this. The ARP
1821 monitor could be configured with a target at the remote end of port3,
1822 thus detecting that failure without switch support.
1824 In general, however, in a multiple switch topology, the ARP
1825 monitor can provide a higher level of reliability in detecting end to
1826 end connectivity failures (which may be caused by the failure of any
1827 individual component to pass traffic for any reason). Additionally,
1828 the ARP monitor should be configured with multiple targets (at least
1829 one for each switch in the network). This will insure that,
1830 regardless of which switch is active, the ARP monitor has a suitable
1831 target to query.
1833 Note, also, that of late many switches now support a functionality
1834 generally referred to as "trunk failover." This is a feature of the
1835 switch that causes the link state of a particular switch port to be set
1836 down (or up) when the state of another switch port goes down (or up).
1837 Its purpose is to propagate link failures from logically "exterior" ports
1838 to the logically "interior" ports that bonding is able to monitor via
1839 miimon. Availability and configuration for trunk failover varies by
1840 switch, but this can be a viable alternative to the ARP monitor when using
1841 suitable switches.
1843 12. Configuring Bonding for Maximum Throughput
1844 ==============================================
1846 12.1 Maximizing Throughput in a Single Switch Topology
1847 ------------------------------------------------------
1849 In a single switch configuration, the best method to maximize
1850 throughput depends upon the application and network environment. The
1851 various load balancing modes each have strengths and weaknesses in
1852 different environments, as detailed below.
1854 For this discussion, we will break down the topologies into
1855 two categories. Depending upon the destination of most traffic, we
1856 categorize them into either "gatewayed" or "local" configurations.
1858 In a gatewayed configuration, the "switch" is acting primarily
1859 as a router, and the majority of traffic passes through this router to
1860 other networks. An example would be the following:
1863 +----------+ +----------+
1864 | |eth0 port1| | to other networks
1865 | Host A +---------------------+ router +------------------->
1866 | +---------------------+ | Hosts B and C are out
1867 | |eth1 port2| | here somewhere
1868 +----------+ +----------+
1870 The router may be a dedicated router device, or another host
1871 acting as a gateway. For our discussion, the important point is that
1872 the majority of traffic from Host A will pass through the router to
1873 some other network before reaching its final destination.
1875 In a gatewayed network configuration, although Host A may
1876 communicate with many other systems, all of its traffic will be sent
1877 and received via one other peer on the local network, the router.
1879 Note that the case of two systems connected directly via
1880 multiple physical links is, for purposes of configuring bonding, the
1881 same as a gatewayed configuration. In that case, it happens that all
1882 traffic is destined for the "gateway" itself, not some other network
1883 beyond the gateway.
1885 In a local configuration, the "switch" is acting primarily as
1886 a switch, and the majority of traffic passes through this switch to
1887 reach other stations on the same network. An example would be the
1888 following:
1890 +----------+ +----------+ +--------+
1891 | |eth0 port1| +-------+ Host B |
1892 | Host A +------------+ switch |port3 +--------+
1893 | +------------+ | +--------+
1894 | |eth1 port2| +------------------+ Host C |
1895 +----------+ +----------+port4 +--------+
1898 Again, the switch may be a dedicated switch device, or another
1899 host acting as a gateway. For our discussion, the important point is
1900 that the majority of traffic from Host A is destined for other hosts
1901 on the same local network (Hosts B and C in the above example).
1903 In summary, in a gatewayed configuration, traffic to and from
1904 the bonded device will be to the same MAC level peer on the network
1905 (the gateway itself, i.e., the router), regardless of its final
1906 destination. In a local configuration, traffic flows directly to and
1907 from the final destinations, thus, each destination (Host B, Host C)
1908 will be addressed directly by their individual MAC addresses.
1910 This distinction between a gatewayed and a local network
1911 configuration is important because many of the load balancing modes
1912 available use the MAC addresses of the local network source and
1913 destination to make load balancing decisions. The behavior of each
1914 mode is described below.
1917 12.1.1 MT Bonding Mode Selection for Single Switch Topology
1918 -----------------------------------------------------------
1920 This configuration is the easiest to set up and to understand,
1921 although you will have to decide which bonding mode best suits your
1922 needs. The trade offs for each mode are detailed below:
1924 balance-rr: This mode is the only mode that will permit a single
1925 TCP/IP connection to stripe traffic across multiple
1926 interfaces. It is therefore the only mode that will allow a
1927 single TCP/IP stream to utilize more than one interface's
1928 worth of throughput. This comes at a cost, however: the
1929 striping generally results in peer systems receiving packets out
1930 of order, causing TCP/IP's congestion control system to kick
1931 in, often by retransmitting segments.
1933 It is possible to adjust TCP/IP's congestion limits by
1934 altering the net.ipv4.tcp_reordering sysctl parameter. The
1935 usual default value is 3, and the maximum useful value is 127.
1936 For a four interface balance-rr bond, expect that a single
1937 TCP/IP stream will utilize no more than approximately 2.3
1938 interface's worth of throughput, even after adjusting
1939 tcp_reordering.
1941 Note that the fraction of packets that will be delivered out of
1942 order is highly variable, and is unlikely to be zero. The level
1943 of reordering depends upon a variety of factors, including the
1944 networking interfaces, the switch, and the topology of the
1945 configuration. Speaking in general terms, higher speed network
1946 cards produce more reordering (due to factors such as packet
1947 coalescing), and a "many to many" topology will reorder at a
1948 higher rate than a "many slow to one fast" configuration.
1950 Many switches do not support any modes that stripe traffic
1951 (instead choosing a port based upon IP or MAC level addresses);
1952 for those devices, traffic for a particular connection flowing
1953 through the switch to a balance-rr bond will not utilize greater
1954 than one interface's worth of bandwidth.
1956 If you are utilizing protocols other than TCP/IP, UDP for
1957 example, and your application can tolerate out of order
1958 delivery, then this mode can allow for single stream datagram
1959 performance that scales near linearly as interfaces are added
1960 to the bond.
1962 This mode requires the switch to have the appropriate ports
1963 configured for "etherchannel" or "trunking."
1965 active-backup: There is not much advantage in this network topology to
1966 the active-backup mode, as the inactive backup devices are all
1967 connected to the same peer as the primary. In this case, a
1968 load balancing mode (with link monitoring) will provide the
1969 same level of network availability, but with increased
1970 available bandwidth. On the plus side, active-backup mode
1971 does not require any configuration of the switch, so it may
1972 have value if the hardware available does not support any of
1973 the load balance modes.
1975 balance-xor: This mode will limit traffic such that packets destined
1976 for specific peers will always be sent over the same
1977 interface. Since the destination is determined by the MAC
1978 addresses involved, this mode works best in a "local" network
1979 configuration (as described above), with destinations all on
1980 the same local network. This mode is likely to be suboptimal
1981 if all your traffic is passed through a single router (i.e., a
1982 "gatewayed" network configuration, as described above).
1984 As with balance-rr, the switch ports need to be configured for
1985 "etherchannel" or "trunking."
1987 broadcast: Like active-backup, there is not much advantage to this
1988 mode in this type of network topology.
1990 802.3ad: This mode can be a good choice for this type of network
1991 topology. The 802.3ad mode is an IEEE standard, so all peers
1992 that implement 802.3ad should interoperate well. The 802.3ad
1993 protocol includes automatic configuration of the aggregates,
1994 so minimal manual configuration of the switch is needed
1995 (typically only to designate that some set of devices is
1996 available for 802.3ad). The 802.3ad standard also mandates
1997 that frames be delivered in order (within certain limits), so
1998 in general single connections will not see misordering of
1999 packets. The 802.3ad mode does have some drawbacks: the
2000 standard mandates that all devices in the aggregate operate at
2001 the same speed and duplex. Also, as with all bonding load
2002 balance modes other than balance-rr, no single connection will
2003 be able to utilize more than a single interface's worth of
2004 bandwidth.
2006 Additionally, the linux bonding 802.3ad implementation
2007 distributes traffic by peer (using an XOR of MAC addresses),
2008 so in a "gatewayed" configuration, all outgoing traffic will
2009 generally use the same device. Incoming traffic may also end
2010 up on a single device, but that is dependent upon the
2011 balancing policy of the peer's 8023.ad implementation. In a
2012 "local" configuration, traffic will be distributed across the
2013 devices in the bond.
2015 Finally, the 802.3ad mode mandates the use of the MII monitor,
2016 therefore, the ARP monitor is not available in this mode.
2018 balance-tlb: The balance-tlb mode balances outgoing traffic by peer.
2019 Since the balancing is done according to MAC address, in a
2020 "gatewayed" configuration (as described above), this mode will
2021 send all traffic across a single device. However, in a
2022 "local" network configuration, this mode balances multiple
2023 local network peers across devices in a vaguely intelligent
2024 manner (not a simple XOR as in balance-xor or 802.3ad mode),
2025 so that mathematically unlucky MAC addresses (i.e., ones that
2026 XOR to the same value) will not all "bunch up" on a single
2027 interface.
2029 Unlike 802.3ad, interfaces may be of differing speeds, and no
2030 special switch configuration is required. On the down side,
2031 in this mode all incoming traffic arrives over a single
2032 interface, this mode requires certain ethtool support in the
2033 network device driver of the slave interfaces, and the ARP
2034 monitor is not available.
2036 balance-alb: This mode is everything that balance-tlb is, and more.
2037 It has all of the features (and restrictions) of balance-tlb,
2038 and will also balance incoming traffic from local network
2039 peers (as described in the Bonding Module Options section,
2040 above).
2042 The only additional down side to this mode is that the network
2043 device driver must support changing the hardware address while
2044 the device is open.
2046 12.1.2 MT Link Monitoring for Single Switch Topology
2047 ----------------------------------------------------
2049 The choice of link monitoring may largely depend upon which
2050 mode you choose to use. The more advanced load balancing modes do not
2051 support the use of the ARP monitor, and are thus restricted to using
2052 the MII monitor (which does not provide as high a level of end to end
2053 assurance as the ARP monitor).
2055 12.2 Maximum Throughput in a Multiple Switch Topology
2056 -----------------------------------------------------
2058 Multiple switches may be utilized to optimize for throughput
2059 when they are configured in parallel as part of an isolated network
2060 between two or more systems, for example:
2062 +-----------+
2063 | Host A |
2064 +-+---+---+-+
2065 | | |
2066 +--------+ | +---------+
2067 | | |
2068 +------+---+ +-----+----+ +-----+----+
2069 | Switch A | | Switch B | | Switch C |
2070 +------+---+ +-----+----+ +-----+----+
2071 | | |
2072 +--------+ | +---------+
2073 | | |
2074 +-+---+---+-+
2075 | Host B |
2076 +-----------+
2078 In this configuration, the switches are isolated from one
2079 another. One reason to employ a topology such as this is for an
2080 isolated network with many hosts (a cluster configured for high
2081 performance, for example), using multiple smaller switches can be more
2082 cost effective than a single larger switch, e.g., on a network with 24
2083 hosts, three 24 port switches can be significantly less expensive than
2084 a single 72 port switch.
2086 If access beyond the network is required, an individual host
2087 can be equipped with an additional network device connected to an
2088 external network; this host then additionally acts as a gateway.
2090 12.2.1 MT Bonding Mode Selection for Multiple Switch Topology
2091 -------------------------------------------------------------
2093 In actual practice, the bonding mode typically employed in
2094 configurations of this type is balance-rr. Historically, in this
2095 network configuration, the usual caveats about out of order packet
2096 delivery are mitigated by the use of network adapters that do not do
2097 any kind of packet coalescing (via the use of NAPI, or because the
2098 device itself does not generate interrupts until some number of
2099 packets has arrived). When employed in this fashion, the balance-rr
2100 mode allows individual connections between two hosts to effectively
2101 utilize greater than one interface's bandwidth.
2103 12.2.2 MT Link Monitoring for Multiple Switch Topology
2104 ------------------------------------------------------
2106 Again, in actual practice, the MII monitor is most often used
2107 in this configuration, as performance is given preference over
2108 availability. The ARP monitor will function in this topology, but its
2109 advantages over the MII monitor are mitigated by the volume of probes
2110 needed as the number of systems involved grows (remember that each
2111 host in the network is configured with bonding).
2113 13. Switch Behavior Issues
2114 ==========================
2116 13.1 Link Establishment and Failover Delays
2117 -------------------------------------------
2119 Some switches exhibit undesirable behavior with regard to the
2120 timing of link up and down reporting by the switch.
2122 First, when a link comes up, some switches may indicate that
2123 the link is up (carrier available), but not pass traffic over the
2124 interface for some period of time. This delay is typically due to
2125 some type of autonegotiation or routing protocol, but may also occur
2126 during switch initialization (e.g., during recovery after a switch
2127 failure). If you find this to be a problem, specify an appropriate
2128 value to the updelay bonding module option to delay the use of the
2129 relevant interface(s).
2131 Second, some switches may "bounce" the link state one or more
2132 times while a link is changing state. This occurs most commonly while
2133 the switch is initializing. Again, an appropriate updelay value may
2134 help.
2136 Note that when a bonding interface has no active links, the
2137 driver will immediately reuse the first link that goes up, even if the
2138 updelay parameter has been specified (the updelay is ignored in this
2139 case). If there are slave interfaces waiting for the updelay timeout
2140 to expire, the interface that first went into that state will be
2141 immediately reused. This reduces down time of the network if the
2142 value of updelay has been overestimated, and since this occurs only in
2143 cases with no connectivity, there is no additional penalty for
2144 ignoring the updelay.
2146 In addition to the concerns about switch timings, if your
2147 switches take a long time to go into backup mode, it may be desirable
2148 to not activate a backup interface immediately after a link goes down.
2149 Failover may be delayed via the downdelay bonding module option.
2151 13.2 Duplicated Incoming Packets
2152 --------------------------------
2154 NOTE: Starting with version 3.0.2, the bonding driver has logic to
2155 suppress duplicate packets, which should largely eliminate this problem.
2156 The following description is kept for reference.
2158 It is not uncommon to observe a short burst of duplicated
2159 traffic when the bonding device is first used, or after it has been
2160 idle for some period of time. This is most easily observed by issuing
2161 a "ping" to some other host on the network, and noticing that the
2162 output from ping flags duplicates (typically one per slave).
2164 For example, on a bond in active-backup mode with five slaves
2165 all connected to one switch, the output may appear as follows:
2167 # ping -n 10.0.4.2
2168 PING 10.0.4.2 (10.0.4.2) from 10.0.3.10 : 56(84) bytes of data.
2169 64 bytes from 10.0.4.2: icmp_seq=1 ttl=64 time=13.7 ms
2170 64 bytes from 10.0.4.2: icmp_seq=1 ttl=64 time=13.8 ms (DUP!)
2171 64 bytes from 10.0.4.2: icmp_seq=1 ttl=64 time=13.8 ms (DUP!)
2172 64 bytes from 10.0.4.2: icmp_seq=1 ttl=64 time=13.8 ms (DUP!)
2173 64 bytes from 10.0.4.2: icmp_seq=1 ttl=64 time=13.8 ms (DUP!)
2174 64 bytes from 10.0.4.2: icmp_seq=2 ttl=64 time=0.216 ms
2175 64 bytes from 10.0.4.2: icmp_seq=3 ttl=64 time=0.267 ms
2176 64 bytes from 10.0.4.2: icmp_seq=4 ttl=64 time=0.222 ms
2178 This is not due to an error in the bonding driver, rather, it
2179 is a side effect of how many switches update their MAC forwarding
2180 tables. Initially, the switch does not associate the MAC address in
2181 the packet with a particular switch port, and so it may send the
2182 traffic to all ports until its MAC forwarding table is updated. Since
2183 the interfaces attached to the bond may occupy multiple ports on a
2184 single switch, when the switch (temporarily) floods the traffic to all
2185 ports, the bond device receives multiple copies of the same packet
2186 (one per slave device).
2188 The duplicated packet behavior is switch dependent, some
2189 switches exhibit this, and some do not. On switches that display this
2190 behavior, it can be induced by clearing the MAC forwarding table (on
2191 most Cisco switches, the privileged command "clear mac address-table
2192 dynamic" will accomplish this).
2194 14. Hardware Specific Considerations
2195 ====================================
2197 This section contains additional information for configuring
2198 bonding on specific hardware platforms, or for interfacing bonding
2199 with particular switches or other devices.
2201 14.1 IBM BladeCenter
2202 --------------------
2204 This applies to the JS20 and similar systems.
2206 On the JS20 blades, the bonding driver supports only
2207 balance-rr, active-backup, balance-tlb and balance-alb modes. This is
2208 largely due to the network topology inside the BladeCenter, detailed
2209 below.
2211 JS20 network adapter information
2212 --------------------------------
2214 All JS20s come with two Broadcom Gigabit Ethernet ports
2215 integrated on the planar (that's "motherboard" in IBM-speak). In the
2216 BladeCenter chassis, the eth0 port of all JS20 blades is hard wired to
2217 I/O Module #1; similarly, all eth1 ports are wired to I/O Module #2.
2218 An add-on Broadcom daughter card can be installed on a JS20 to provide
2219 two more Gigabit Ethernet ports. These ports, eth2 and eth3, are
2220 wired to I/O Modules 3 and 4, respectively.
2222 Each I/O Module may contain either a switch or a passthrough
2223 module (which allows ports to be directly connected to an external
2224 switch). Some bonding modes require a specific BladeCenter internal
2225 network topology in order to function; these are detailed below.
2227 Additional BladeCenter-specific networking information can be
2228 found in two IBM Redbooks (www.ibm.com/redbooks):
2230 "IBM eServer BladeCenter Networking Options"
2231 "IBM eServer BladeCenter Layer 2-7 Network Switching"
2233 BladeCenter networking configuration
2234 ------------------------------------
2236 Because a BladeCenter can be configured in a very large number
2237 of ways, this discussion will be confined to describing basic
2238 configurations.
2240 Normally, Ethernet Switch Modules (ESMs) are used in I/O
2241 modules 1 and 2. In this configuration, the eth0 and eth1 ports of a
2242 JS20 will be connected to different internal switches (in the
2243 respective I/O modules).
2245 A passthrough module (OPM or CPM, optical or copper,
2246 passthrough module) connects the I/O module directly to an external
2247 switch. By using PMs in I/O module #1 and #2, the eth0 and eth1
2248 interfaces of a JS20 can be redirected to the outside world and
2249 connected to a common external switch.
2251 Depending upon the mix of ESMs and PMs, the network will
2252 appear to bonding as either a single switch topology (all PMs) or as a
2253 multiple switch topology (one or more ESMs, zero or more PMs). It is
2254 also possible to connect ESMs together, resulting in a configuration
2255 much like the example in "High Availability in a Multiple Switch
2256 Topology," above.
2258 Requirements for specific modes
2259 -------------------------------
2261 The balance-rr mode requires the use of passthrough modules
2262 for devices in the bond, all connected to an common external switch.
2263 That switch must be configured for "etherchannel" or "trunking" on the
2264 appropriate ports, as is usual for balance-rr.
2266 The balance-alb and balance-tlb modes will function with
2267 either switch modules or passthrough modules (or a mix). The only
2268 specific requirement for these modes is that all network interfaces
2269 must be able to reach all destinations for traffic sent over the
2270 bonding device (i.e., the network must converge at some point outside
2271 the BladeCenter).
2273 The active-backup mode has no additional requirements.
2275 Link monitoring issues
2276 ----------------------
2278 When an Ethernet Switch Module is in place, only the ARP
2279 monitor will reliably detect link loss to an external switch. This is
2280 nothing unusual, but examination of the BladeCenter cabinet would
2281 suggest that the "external" network ports are the ethernet ports for
2282 the system, when it fact there is a switch between these "external"
2283 ports and the devices on the JS20 system itself. The MII monitor is
2284 only able to detect link failures between the ESM and the JS20 system.
2286 When a passthrough module is in place, the MII monitor does
2287 detect failures to the "external" port, which is then directly
2288 connected to the JS20 system.
2290 Other concerns
2291 --------------
2293 The Serial Over LAN (SoL) link is established over the primary
2294 ethernet (eth0) only, therefore, any loss of link to eth0 will result
2295 in losing your SoL connection. It will not fail over with other
2296 network traffic, as the SoL system is beyond the control of the
2297 bonding driver.
2299 It may be desirable to disable spanning tree on the switch
2300 (either the internal Ethernet Switch Module, or an external switch) to
2301 avoid fail-over delay issues when using bonding.
2304 15. Frequently Asked Questions
2305 ==============================
2307 1. Is it SMP safe?
2309 Yes. The old 2.0.xx channel bonding patch was not SMP safe.
2310 The new driver was designed to be SMP safe from the start.
2312 2. What type of cards will work with it?
2314 Any Ethernet type cards (you can even mix cards - a Intel
2315 EtherExpress PRO/100 and a 3com 3c905b, for example). For most modes,
2316 devices need not be of the same speed.
2318 Starting with version 3.2.1, bonding also supports Infiniband
2319 slaves in active-backup mode.
2321 3. How many bonding devices can I have?
2323 There is no limit.
2325 4. How many slaves can a bonding device have?
2327 This is limited only by the number of network interfaces Linux
2328 supports and/or the number of network cards you can place in your
2329 system.
2331 5. What happens when a slave link dies?
2333 If link monitoring is enabled, then the failing device will be
2334 disabled. The active-backup mode will fail over to a backup link, and
2335 other modes will ignore the failed link. The link will continue to be
2336 monitored, and should it recover, it will rejoin the bond (in whatever
2337 manner is appropriate for the mode). See the sections on High
2338 Availability and the documentation for each mode for additional
2339 information.
2341 Link monitoring can be enabled via either the miimon or
2342 arp_interval parameters (described in the module parameters section,
2343 above). In general, miimon monitors the carrier state as sensed by
2344 the underlying network device, and the arp monitor (arp_interval)
2345 monitors connectivity to another host on the local network.
2347 If no link monitoring is configured, the bonding driver will
2348 be unable to detect link failures, and will assume that all links are
2349 always available. This will likely result in lost packets, and a
2350 resulting degradation of performance. The precise performance loss
2351 depends upon the bonding mode and network configuration.
2353 6. Can bonding be used for High Availability?
2355 Yes. See the section on High Availability for details.
2357 7. Which switches/systems does it work with?
2359 The full answer to this depends upon the desired mode.
2361 In the basic balance modes (balance-rr and balance-xor), it
2362 works with any system that supports etherchannel (also called
2363 trunking). Most managed switches currently available have such
2364 support, and many unmanaged switches as well.
2366 The advanced balance modes (balance-tlb and balance-alb) do
2367 not have special switch requirements, but do need device drivers that
2368 support specific features (described in the appropriate section under
2369 module parameters, above).
2371 In 802.3ad mode, it works with systems that support IEEE
2372 802.3ad Dynamic Link Aggregation. Most managed and many unmanaged
2373 switches currently available support 802.3ad.
2375 The active-backup mode should work with any Layer-II switch.
2377 8. Where does a bonding device get its MAC address from?
2379 When using slave devices that have fixed MAC addresses, or when
2380 the fail_over_mac option is enabled, the bonding device's MAC address is
2381 the MAC address of the active slave.
2383 For other configurations, if not explicitly configured (with
2384 ifconfig or ip link), the MAC address of the bonding device is taken from
2385 its first slave device. This MAC address is then passed to all following
2386 slaves and remains persistent (even if the first slave is removed) until
2387 the bonding device is brought down or reconfigured.
2389 If you wish to change the MAC address, you can set it with
2390 ifconfig or ip link:
2392 # ifconfig bond0 hw ether 00:11:22:33:44:55
2394 # ip link set bond0 address 66:77:88:99:aa:bb
2396 The MAC address can be also changed by bringing down/up the
2397 device and then changing its slaves (or their order):
2399 # ifconfig bond0 down ; modprobe -r bonding
2400 # ifconfig bond0 .... up
2401 # ifenslave bond0 eth...
2403 This method will automatically take the address from the next
2404 slave that is added.
2406 To restore your slaves' MAC addresses, you need to detach them
2407 from the bond (`ifenslave -d bond0 eth0'). The bonding driver will
2408 then restore the MAC addresses that the slaves had before they were
2409 enslaved.
2411 16. Resources and Links
2412 =======================
2414 The latest version of the bonding driver can be found in the latest
2415 version of the linux kernel, found on http://kernel.org
2417 The latest version of this document can be found in either the latest
2418 kernel source (named Documentation/networking/bonding.txt), or on the
2419 bonding sourceforge site:
2421 http://www.sourceforge.net/projects/bonding
2423 Discussions regarding the bonding driver take place primarily on the
2424 bonding-devel mailing list, hosted at sourceforge.net. If you have
2425 questions or problems, post them to the list. The list address is:
2427 bonding-devel@lists.sourceforge.net
2429 The administrative interface (to subscribe or unsubscribe) can
2430 be found at:
2432 https://lists.sourceforge.net/lists/listinfo/bonding-devel
2434 Donald Becker's Ethernet Drivers and diag programs may be found at :
2435 - http://www.scyld.com/network/
2437 You will also find a lot of information regarding Ethernet, NWay, MII,
2438 etc. at www.scyld.com.
2440 -- END --