Lines Matching +full:single +full:- +full:link
1 .. SPDX-License-Identifier: GPL-2.0
11 Corrections, HA extensions: 2000/10/03-15:
13 - Willy Tarreau <willy at meta-x.org>
14 - Constantine Gavrilov <const-g at xpert.com>
15 - Chad N. Tindel <ctindel at ieee dot org>
16 - Janice Girouard <girouard at us dot ibm dot com>
17 - Jay Vosburgh <fubar at us dot ibm dot com>
22 - Mitch Williams <mitch.a.williams at intel.com>
28 multiple network interfaces into a single logical "bonded" interface.
31 Additionally, link integrity monitoring may be performed.
35 the original tools from extreme-linux and beowulf sites will not work
69 7. Link Monitoring
77 8.3 Painfully Slow Or No Failed Link Detection By Miimon
84 11.1 High Availability in a Single Switch Topology
87 11.2.2 HA Link Monitoring for Multiple Switch Topology
90 12.1 Maximum Throughput in a Single Switch Topology
91 12.1.1 MT Bonding Mode Selection for Single Switch Topology
92 12.1.2 MT Link Monitoring for Single Switch Topology
95 12.2.2 MT Link Monitoring for Multiple Switch Topology
98 13.1 Link Establishment and Failover Delays
119 -----------------------------------------------
135 ---------------------------
148 ``/etc/modprobe.d/*.conf`` configuration files, or in a distro-specific
156 configuring a bond, it is recommended "tail -f /var/log/messages" be
161 degradation will occur during link failures. Very few devices do not
173 (active-backup, balance-alb and balance-tlb). Possible values
175 string. If a name is given, the slave and its link must be up in order
190 is 1 - 65535. If the value is not specified, it takes 65535 as the
198 In an AD system, this specifies the mac-address for the actor in
200 address. If the all-zeroes MAC is specified, bonding will internally
202 local-admin bit set for this mac but driver does not enforce it. If
230 - A slave is added to or removed from the bond
232 - Any slave's link state changes
234 - Any slave's 802.3ad association state changes
236 - The bond's administrative state changes to up
253 In an AD system, the port-key has three parts as shown below -
259 01-05 Speed
260 06-15 User-defined
264 from 0 - 1023. If not given, the system defaults to 0.
283 Specifies the ARP link monitoring frequency in milliseconds.
300 the same link which could cause the other team members to
309 sent to determine the health of the link to the targets.
320 sent to determine the health of the link to the targets.
331 non-ARP traffic should be filtered (disregarded) for link
389 beyond a common switch. Should the link between the switch and
400 packets for link availability purposes. Arriving packets that are
406 determining if a slave has received traffic for link availability
413 link availability purposes.
421 This option affects only active-backup mode for slaves with
445 The default value is 2, and the allowable range is 1 - 255.
450 a slave after a link failure has been detected. This option
451 is only valid for the miimon link monitor. The downdelay
458 Specifies whether active-backup mode should set all slaves to
468 bonding to set all slaves of an active-backup bond to
495 monitor, devices which assert link up prior to being
541 Option specifying the rate in which we'll ask our link partner
563 Specifies the MII link monitoring frequency in milliseconds.
564 This determines how often the link state of each slave is
565 inspected for link failures. A value of zero disables MII
566 link monitoring. A value of 100 is a good starting point.
567 The use_carrier option, below, affects how the link state is
574 asserting carrier. It is similar to the Cisco EtherChannel min-links
576 must be up (link-up state) before marking the bond device as up
585 aggregator cannot be active without at least one available link,
591 balance-rr (round robin). Possible values are:
593 balance-rr or 0
595 Round-robin policy: Transmit packets in sequential
600 active-backup or 1
602 Active-backup policy: Only one slave in the bond is
609 occurs in active-backup mode, bonding will issue one
621 balance-xor or 2
639 IEEE 802.3ad Dynamic link aggregation. Creates
649 regards to the packet mis-ordering requirements of
659 2. A switch that supports IEEE 802.3ad Dynamic link
665 balance-tlb or 5
687 balance-alb or 6
689 Adaptive load balancing: includes balance-tlb plus
717 and when an inactive slave is re-activated. The
721 When a link is reconnected or a new slave joins the
750 failover event. As soon as the link is up on the new slave
752 bonding device and each VLAN sub-device. This is repeated at
756 The valid range is 0 - 255; the default value is 1. These options
757 affect only the active-backup mode. These options were added for
770 The valid range is 0 - 65535; the default value is 1. This option
771 has effect only in balance-rr mode.
778 This delay should be a multiple of the link monitor interval
780 value is 0 which means to match the value of the link monitor
789 for active-backup(1), balance-tlb (5) and balance-alb (6) mode.
799 off-line will alternate devices be used. This is useful when
803 The primary option is only valid for active-backup(1),
804 balance-tlb (5) and balance-alb (6) mode.
811 occurs. This option is designed to prevent flip-flopping between
853 characteristics but can cause packet reordering. If re-ordering is
856 xmit-hash-policy can be used to select the appropriate hashing for
871 slave after a link recovery has been detected. This option is
872 only valid for the miimon link monitor. The updelay value
879 ioctls vs. netif_carrier_ok() to determine the link
886 If bonding insists that the link is up when it should not be,
890 it will appear as if the link is always up. In this case,
892 MII / ETHTOOL ioctl method to determine the link state.
901 balance-xor, 802.3ad, and tlb modes. Possible values are:
934 network peer on the same slave. For non-IP traffic,
950 slaves, although a single connection will not span
966 information is omitted. For non-IP traffic, the
971 single TCP or UDP conversation containing both
1003 hash to load-balance traffic per-vlan, with failover
1006 use their own vlan, to give lacp-like functionality
1007 without requiring lacp-capable switching hardware.
1024 The valid range is 0 - 255; the default value is 1. A value of 0
1028 This option is useful for bonding modes balance-rr (0), active-backup
1029 (1), balance-tlb (5) and balance-alb (6), in which a failover can
1041 The valid range is 1 - 0x7fffffff; the default value is 1. This Option
1042 has effect only in balance-tlb and balance-alb modes.
1070 $ rpm -qf /sbin/ifup
1085 ----------------------------------------
1098 ifcfg-id file for each slave device. The simplest way to accomplish
1100 file ifcfg-id file created; see below for some issues with DHCP). The
1103 ifcfg-id-xx:xx:xx:xx:xx:xx
1108 Once the set of ifcfg-id-xx:xx:xx:xx:xx:xx files has been
1118 _nm_name='bus-pci-0001:61:01.0'
1128 Once the ifcfg-id-xx:xx:xx:xx:xx:xx files have been modified,
1130 itself. This file is named ifcfg-bondX, where X is the number of the
1132 ifcfg-bond0, the second is ifcfg-bond1, and so on. The sysconfig
1136 The contents of the ifcfg-bondX file is as follows::
1146 BONDING_MODULE_OPTS="mode=active-backup miimon=100"
1148 BONDING_SLAVE1="bus-pci-0000:06:08.1"
1177 for the bonding mode, link monitoring, and so on here. Do not include
1187 (bus-pci-0000:06:08.1 in the example above) specify the physical
1218 -------------------------------
1228 -----------------------------------------------
1232 bonding instance to have an appropriately configured ifcfg-bondX file
1236 ifcfg-bondX files.
1239 options in the ifcfg-bondX file, it is not necessary to add them to
1243 ------------------------------------------
1256 network-script file for all physical adapters that will be members of
1257 a bondX link. Network script files are located in the directory:
1259 /etc/sysconfig/network-scripts
1261 The file name must be prefixed with "ifcfg-eth" and suffixed
1263 for eth0 would be named /etc/sysconfig/network-scripts/ifcfg-eth0.
1274 must correspond with the name of the file, i.e., ifcfg-eth1 must have
1282 script will be /etc/sysconfig/network-scripts/ifcfg-bondX where X is
1283 the number of the bond. For bond0 the file is named "ifcfg-bond0",
1284 for bond1 it is named "ifcfg-bond1", and so on. Within that file,
1301 and, indeed, preferable, to specify the bonding options in the ifcfg-bond0
1304 BONDING_OPTS="mode=active-backup arp_interval=60 arp_ip_target=192.168.1.254"
1327 options bond0 mode=balance-alb miimon=100
1333 will restart the networking subsystem and your bond link should be now
1337 ---------------------------------
1350 -------------------------------------------------
1354 specifying the appropriate BONDING_OPTS= in ifcfg-bondX where X is the
1362 -----------------------------------------------
1372 `ip link` commands to the system's global init script. The name of
1381 modprobe bonding mode=balance-alb miimon=100
1384 ip link set eth0 master bond0
1385 ip link set eth1 master bond0
1404 enabled without re-running the entire global init script.
1420 -----------------------------------------
1449 options bond0 -o bond0 mode=balance-rr miimon=100
1452 options bond1 -o bond1 mode=balance-alb miimon=50
1455 named "bond0" and creates the bond0 device in balance-rr mode with an
1457 bond1 device in balance-alb mode with an miimon of 50.
1464 install bond1 /sbin/modprobe --ignore-install bonding -o bond1 \
1465 mode=balance-alb miimon=50
1471 to rename modules at load time (the "-o bond1" part). Attempts to pass
1479 ------------------------------------------
1499 -----------------------------
1506 # echo -bar > /sys/class/net/bonding_masters
1519 --------------------------
1531 # echo -eth0 > /sys/class/net/bond0/bonding/slaves
1540 # echo -eth0 > /sys/class/net/eth0/master/bonding/slaves
1545 -------------------------------
1549 The names of these files correspond directly with the command-
1558 To configure bond0 for balance-alb mode::
1562 - or -
1563 # echo balance-alb > /sys/class/net/bond0/bonding/mode
1576 monitoring is enabled, and vice-versa.
1589 # echo -192.168.0.100 > /sys/class/net/bond0/bonding/arp_ip_target
1602 ---------------------
1613 echo balance-alb > /sys/class/net/bond0/bonding/mode
1620 active-backup mode, using ARP monitoring, add the following lines to
1625 echo active-backup > /sys/class/net/bond1/bonding/mode
1633 -----------------------------------------
1640 the box. The ifenslave-2.6 package should be installed to provide bonding
1641 support. Once installed, this package will provide ``bond-*`` options
1644 Note that ifenslave-2.6 package will load the bonding module and use
1648 ----------------------
1651 active-backup mode, with eth0 and eth1 as slaves::
1655 bond-slaves eth0 eth1
1656 bond-mode active-backup
1657 bond-miimon 100
1658 bond-primary eth0 eth1
1667 bond-slaves none
1668 bond-mode active-backup
1669 bond-miimon 100
1673 bond-master bond0
1674 bond-primary eth0 eth1
1678 bond-master bond0
1679 bond-primary eth0 eth1
1681 For a full list of ``bond-*`` supported options in /etc/network/interfaces and
1683 /usr/share/doc/ifenslave-2.6.
1686 ----------------------------------------------
1709 Bonding Mode: fault-tolerance (active-backup)
1719 Link Failure Count: 0
1725 Link Failure Count: 0
1760 a pass-through for selecting output queues on the slave device rather than
1764 output slave selection was limited to round-robin and active-backup modes.
1767 ----------------------------------------------------------
1771 destined to link local mac addresses (which switches/bridges are not
1783 (a) ad_actor_system : You can set a random mac-address that can be used for
1785 Also it's preferable to set the local-admin bit. Following shell code
1786 generates a random mac-address as described above::
1798 is 65535, but system can take the value from 1 - 65535. Following shell
1804 (c) ad_user_port_key : Use the user portion of the port-key. The default
1805 keeps this empty. These are the upper 10 bits of the port-key and value
1806 ranges from 0 - 1023. Following shell code generates these 10 bits and
1817 -------------------------
1819 Each bonding device has a read-only file residing in the
1828 Bonding Mode: load balancing (round-robin)
1837 Link Failure Count: 1
1841 Link Failure Count: 1
1847 -------------------------
1860 bond0 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1867 eth0 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1874 eth1 Link encap:Ethernet HWaddr 00:C0:F0:1F:37:B4
1890 The active-backup, balance-tlb and balance-alb modes do not
1897 grouped together in a single etherchannel instance, then that
1901 The balance-rr, balance-xor and broadcast modes generally
1933 "un-accelerated" by the bonding driver so the VLAN tag sits in the
1940 would pick up the all-zeroes hardware address. Once the first slave
1961 underlying device -- i.e. the bonding interface -- to promiscuous
1965 7. Link Monitoring
1969 monitoring a slave device's link state: the ARP monitor and the MII
1977 -------------------------
1981 uses the response as an indication that the link is operating. This
1986 ------------------------------------
2001 For just a single target the options would resemble::
2009 -------------------------
2026 device's (via ioctl) MII registers and check the link state. If that
2031 and ethtool requests), then the MII monitor will assume the link is
2038 -------------------------
2073 ----------------------------
2084 options bond0 mode=some-mode miimon=50
2116 8.3. Painfully Slow Or No Failed Link Detection By Miimon
2117 ---------------------------------------------------------
2123 not support the netif_carrier_on/_off link state tracking system.
2128 not maintain it in real time, e.g., only polling the link state at
2131 miimon is very slow in detecting link failures, try specifying
2202 For the balance-rr, balance-xor, broadcast, and 802.3ad modes,
2205 For the active-backup, balance-tlb and balance-alb modes, the
2208 For balance-tlb mode, the active slave is the slave currently
2211 For balance-alb mode, the active slave is the slave used as a
2212 "primary." This slave is used for mode-specific control traffic, for
2215 For the active-backup, balance-tlb and balance-alb modes, when
2216 the active slave changes (e.g., due to a link failure), the
2229 11.1 High Availability in a Single Switch Topology
2230 --------------------------------------------------
2232 If two hosts (or a host and a single switch) are directly
2237 support link monitoring of their members, so if individual links fail,
2244 ----------------------------------------------------
2255 +-----+----+ +-----+----+
2257 | switch A +--------------------------+ switch B |
2259 +-----+----+ +-----++---+
2261 | +-------+ |
2262 +-------------+ host1 +---------------+
2263 eth0 +-------+ eth1
2265 In this configuration, there is a link between the two
2266 switches (ISL, or inter switch link), and multiple ports connecting to
2271 -------------------------------------------------------------
2273 In a topology such as the example above, the active-backup and
2278 active-backup:
2284 preferred link is always used when it is available.
2291 necessary for some specific one-way traffic to reach both
2294 11.2.2 HA Link Monitoring Selection for Multiple Switch Topology
2295 ----------------------------------------------------------------
2297 The choice of link monitoring ultimately depends upon your
2300 example, in the above example, if the "port3" link fails at the remote
2316 switch that causes the link state of a particular switch port to be set
2318 Its purpose is to propagate link failures from logically "exterior" ports
2327 12.1 Maximizing Throughput in a Single Switch Topology
2328 ------------------------------------------------------
2330 In a single switch configuration, the best method to maximize
2344 +----------+ +----------+
2346 | Host A +---------------------+ router +------------------->
2347 | +---------------------+ | Hosts B and C are out
2349 +----------+ +----------+
2371 +----------+ +----------+ +--------+
2372 | |eth0 port1| +-------+ Host B |
2373 | Host A +------------+ switch |port3 +--------+
2374 | +------------+ | +--------+
2375 | |eth1 port2| +------------------+ Host C |
2376 +----------+ +----------+port4 +--------+
2398 12.1.1 MT Bonding Mode Selection for Single Switch Topology
2399 -----------------------------------------------------------
2405 balance-rr:
2406 This mode is the only mode that will permit a single
2409 single TCP/IP stream to utilize more than one interface's
2432 through the switch to a balance-rr bond will not utilize greater
2437 delivery, then this mode can allow for single stream datagram
2444 active-backup:
2446 the active-backup mode, as the inactive backup devices are all
2448 load balancing mode (with link monitoring) will provide the
2450 available bandwidth. On the plus side, active-backup mode
2455 balance-xor:
2462 if all your traffic is passed through a single router (i.e., a
2465 As with balance-rr, the switch ports need to be configured for
2469 Like active-backup, there is not much advantage to this
2481 in general single connections will not see misordering of
2485 balance modes other than balance-rr, no single connection will
2486 be able to utilize more than a single interface's worth of
2493 traffic may also end up on a single device, but that is
2501 balance-tlb:
2502 The balance-tlb mode balances outgoing traffic by peer.
2505 send all traffic across a single device. However, in a
2508 manner (not a simple XOR as in balance-xor or 802.3ad mode),
2510 XOR to the same value) will not all "bunch up" on a single
2515 in this mode all incoming traffic arrives over a single
2520 balance-alb:
2521 This mode is everything that balance-tlb is, and more.
2522 It has all of the features (and restrictions) of balance-tlb,
2531 12.1.2 MT Link Monitoring for Single Switch Topology
2532 ----------------------------------------------------
2534 The choice of link monitoring may largely depend upon which
2541 -----------------------------------------------------
2547 +-----------+
2549 +-+---+---+-+
2551 +--------+ | +---------+
2553 +------+---+ +-----+----+ +-----+----+
2555 +------+---+ +-----+----+ +-----+----+
2557 +--------+ | +---------+
2559 +-+---+---+-+
2561 +-----------+
2567 cost effective than a single larger switch, e.g., on a network with 24
2569 a single 72 port switch.
2576 -------------------------------------------------------------
2579 configurations of this type is balance-rr. Historically, in this
2584 packets has arrived). When employed in this fashion, the balance-rr
2588 12.2.2 MT Link Monitoring for Multiple Switch Topology
2589 ------------------------------------------------------
2601 13.1 Link Establishment and Failover Delays
2602 -------------------------------------------
2605 timing of link up and down reporting by the switch.
2607 First, when a link comes up, some switches may indicate that
2608 the link is up (carrier available), but not pass traffic over the
2616 Second, some switches may "bounce" the link state one or more
2617 times while a link is changing state. This occurs most commonly while
2622 driver will immediately reuse the first link that goes up, even if the
2633 to not activate a backup interface immediately after a link goes down.
2637 --------------------------------
2649 For example, on a bond in active-backup mode with five slaves
2652 # ping -n 10.0.4.2
2669 single switch, when the switch (temporarily) floods the traffic to all
2676 most Cisco switches, the privileged command "clear mac address-table
2687 --------------------
2692 balance-rr, active-backup, balance-tlb and balance-alb modes. This is
2697 --------------------------------
2700 integrated on the planar (that's "motherboard" in IBM-speak). In the
2703 An add-on Broadcom daughter card can be installed on a JS20 to provide
2712 Additional BladeCenter-specific networking information can be
2715 - "IBM eServer BladeCenter Networking Options"
2716 - "IBM eServer BladeCenter Layer 2-7 Network Switching"
2719 ------------------------------------
2737 appear to bonding as either a single switch topology (all PMs) or as a
2744 -------------------------------
2746 The balance-rr mode requires the use of passthrough modules
2749 appropriate ports, as is usual for balance-rr.
2751 The balance-alb and balance-tlb modes will function with
2758 The active-backup mode has no additional requirements.
2760 Link monitoring issues
2761 ----------------------
2764 monitor will reliably detect link loss to an external switch. This is
2769 only able to detect link failures between the ESM and the JS20 system.
2776 --------------
2778 The Serial Over LAN (SoL) link is established over the primary
2779 ethernet (eth0) only, therefore, any loss of link to eth0 will result
2786 avoid fail-over delay issues when using bonding.
2793 -------------------
2799 -----------------------------------------
2801 Any Ethernet type cards (you can even mix cards - a Intel
2806 slaves in active-backup mode.
2809 ----------------------------------------
2814 ----------------------------------------------
2820 5. What happens when a slave link dies?
2821 ----------------------------------------
2823 If link monitoring is enabled, then the failing device will be
2824 disabled. The active-backup mode will fail over to a backup link, and
2825 other modes will ignore the failed link. The link will continue to be
2831 Link monitoring can be enabled via either the miimon or
2837 If no link monitoring is configured, the bonding driver will
2838 be unable to detect link failures, and will assume that all links are
2844 ----------------------------------------------
2849 ---------------------------------------------
2853 In the basic balance modes (balance-rr and balance-xor), it
2858 The advanced balance modes (balance-tlb and balance-alb) do
2864 802.3ad Dynamic Link Aggregation. Most managed and many unmanaged
2867 The active-backup mode should work with any Layer-II switch.
2870 ---------------------------------------------------------
2877 ifconfig or ip link), the MAC address of the bonding device is taken from
2883 ifconfig or ip link::
2887 # ip link set bond0 address 66:77:88:99:aa:bb
2892 # ifconfig bond0 down ; modprobe -r bonding
2900 from the bond (``ifenslave -d bond0 eth0``). The bonding driver will
2922 http://vger.kernel.org/vger-lists.html#netdev