Basic Management
The configuration of all network adapters can be viewed using the following command:
# ifconfig -a
Using the example addresses defined in TABLE 1, the ifconfig -a command yields the following output:
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 qfe0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2 inet 192.168.49.42 netmask ffffff00 broadcast 192.168.49.255 groupname production ether 8:0:20:c7:6e:bc qfe0:1: flags=9040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4, NOFAILOVER> mtu 1500 index 2 inet 192.168.49.105 netmask ffffff00 broadcast 192.168.49.255 qfe4: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3 inet 192.168.49.7 netmask ffffff00 broadcast 192.168.49.255 groupname production ether 8:0:20:b3:e6:f7 qfe4:1: flags=9040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4, NOFAILOVER> mtu 1500 index 3 inet 192.168.49.106 netmask ffffff00 broadcast 192.168.49.255
The output in the preceding codebox shows that four addresses have been defined. The two IP network multipathing test addresses, qfe0:1 and qfe4:1, are tied to each interface; that is, they are marked NOFAILOVER, and will not migrate to the surviving interface during a failure. The purpose of the addresses being marked NOFAILOVER is to detect failure and recovery of an interface.
Interface Failure
To test that IP network multipathing is functioning correctly (using the example configuration defined in TABLE 1), unplug the network cable attached to qfe0. This causes the following error messages to be displayed on the console:
Dec 11 16:32:49 camelot qfe: NOTICE: SUNW,qfe0: No response from Ethernet network: Link Down - cable problem? Dec 11 16:32:57 camelot in.mpathd[36]: NIC failure detected on qfe0 Dec 11 16:32:57 camelot in.mpathd[36]: Successfully failed over from NIC qfe0 to NIC qfe4 Dec 11 16:33:01 camelot qfe: NOTICE: SUNW,qfe0: No response from Ethernet network: Link Down - cable problem?
NOTE
It takes approximately 10 seconds to detect and recover from a failure with the default configuration. The configuration of the IP Networking Multipathing daemon is set in the /etc/default/mpathd file.
Executing the ifconfig -a command produces the following output:
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 qfe0: flags=19000842<BROADCAST,RUNNING,MULTICAST,IPv4,NOFAILOVER, FAILED> mtu 1500 index 3 inet 0.0.0.0 netmask 0 groupname production ether 8:0:20:b3:e6:f7 qfe0:1 flags=19040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED, IPv4,NOFAILOVER,FAILED> mtu 1500 index 2 inet 192.168.49.105 netmask ffffff00 broadcast 192.168.49.255 groupname production ether 8:0:20:c7:6e:bc qfe4: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3 inet 192.168.49.7 netmask ffffff00 broadcast 192.168.49.255 groupname production ether 8:0:20:b3:e6:f7 qfe4:1: flags=9040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4 NOFAILOVER> mtu 1500 index 3 inet 192.168.49.106 netmask ffffff00 broadcast 192.168.49.255 qfe4:2: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3 inet 192.168.49.42 netmask ffffff00 broadcast 192.168.49.255
Notice in the preceding codebox the output, qfe0 has been marked as FAILED and the IP address 192.168.49.42 has been moved from qfe0 to qfe4:2; thus, clients can still reach the host at this address.
NOTE
To detect errors on a single network interface, use IP network multipathing configured with a group containing one interface and a data address marked NOFAILOVER. However, be aware that there is no resilience in such a configuration. Further information on this feature can be found in the IP Network Multipathing Administration Guide section of the Solaris 8 OE documentation. Go to: http://docs.sun.com/