(1) Physical interface settings
Set up physical interfaces. The physical interface settings vary depending on redundant network methods and operating system.
See the following Table 3.8 Physical interface settings.
Redundant network methods | Operating system | ||
---|---|---|---|
RHEL5/RHEL6 | |||
Tagged VLAN disabled | Tagged VLAN enabled | ||
Fast switching mode (IPv4) | |||
Fast switching mode (IPv6) | |||
Fast switching mode (Dual) | |||
NIC switching mode (IPv4) | Primary interface | ||
Secondary interface | |||
NIC switching mode (IPv6) | Primary interface | ||
Secondary interface | |||
NIC switching mode (Dual) | Primary interface | ||
Secondary interface | |||
Virtual NIC mode | |||
GS linkage mode | Not supported |
Note
When a operating system is RHEL5 or RHEL6, to specify "HWADDR=XX:XX:XX:XX:XX:XX" for the GLS physical interface settings (the "/etc/sysconfig/network-scripts/ifcfg-ethX" file), add "HOTPLUG=no" to the settings. When the system is RHEL5, edit /etc/udev/rules.d/60-net.rules file, /etc/hotplug/net.agent file or /etc/udev/rules.d/31-network.rules file to invalidate the network hotplug function of the interface (sha*,eth*.*).
Information
If "HOTPLUG=no" is set when the operating system is RHEL5 or RHEL6, it will not disable the PCI Hot Plug. Active maintenance of NIC (PCI cards) can be performed for the physical interface where "HOTPLUG=no" is set.
/etc/sysconfig/network-scripts/ifcfg-ethX
DEVICE=ethX BOOTPROTO=static HWADDR=XX:XX:XX:XX:XX:XX HOTPLUG=no BROADCAST=XXX.XXX.XXX.XXX IPADDR=XXX.XXX.XXX.XXX NETMASK=XXX.XXX.XXX.XXX NETWORK=XXX.XXX.XXX.XXX ONBOOT=yes TYPE=Ethernet |
/etc/sysconfig/network-scripts/ifcfg-ethX
DEVICE=ethX BOOTPROTO=static HWADDR=XX:XX:XX:XX:XX:XX HOTPLUG=no ONBOOT=yes TYPE=Ethernet |
/etc/sysconfig/network-scripts/ifcfg-ethX
DEVICE=ethX BOOTPROTO=static HWADDR=XX:XX:XX:XX:XX:XX HOTPLUG=no ONBOOT=yes TYPE=Ethernet |
/etc/sysconfig/network-scripts/ifcfg-ethX.Y
DEVICE=ethX.Y BOOTPROTO=static BROADCAST=XXX.XXX.XXX.XXX IPADDR=XXX.XXX.XXX.XXX NETMASK=XXX.XXX.XXX.XXX NETWORK=XXX.XXX.XXX.XXX ONBOOT=yes |
Information
If you want to use the NIC switching mode to share the same physical connection between two virtual interfaces, one bundles physical interfaces and the other bundles tagged VLAN interfaces, you need to set up ifcfg-ethX the same as Setup 1 using the same IP address (IPADDR=) and other values. For example, if sha0 bundles eth0 and eth1, and sha1 bundles eth0.2 and eth1.2, configure ifcfg-eth0 according to ifcfg-ethX in Setup 1, not according to that shown in Setup 3.
/etc/sysconfig/network-scripts/ifcfg-ethX
DEVICE=ethX BOOTPROTO=static HWADDR=XX:XX:XX:XX:XX:XX HOTPLUG=no ONBOOT=yes TYPE=Ethernet |
/etc/sysconfig/network-scripts/ifcfg-ethX.Y
DEVICE=ethX.Y BOOTPROTO=static ONBOOT=yes |
/etc/sysconfig/network-scripts/ifcfg-ethX
DEVICE=ethX BOOTPROTO=static HWADDR=XX:XX:XX:XX:XX:XX HOTPLUG=no ONBOOT=yes DEVICETYPE=hanet |
(2) Verification of the physical interface
Verify if the physical interface is inserted into the system using ifconfig command.
Also, if the physical interface is UP, check whether it is marked as "RUNNING". If "RUNNING" is not displayed, the links might be down on the interface. Check the cable switch and HUB speed settings. Use the "ethtool" command to check the link state.
# ifconfig -a eth0 Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet addr:192.168.70.2 Bcast:192.168.70.255 Mask:255.255.255.0 inet6 addr: fe80::xxx:xxxx:xxxx:xxxx/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2140 errors:0 dropped:0 overruns:0 frame:0 TX packets:2140 errors:0 dropped:0 overruns:0 carrier:0 collisions:22 txqueuelen:1000 RX bytes:278285 (271.7 KiB) TX bytes:273656 (267.2 KiB) Base address:0xec80 Memory:d2fc0000-d2fe0000 eth1 Link encap:Ethernet HWaddr 0xx:xx:xx:xx:xx:xx inet addr:192.168.71.2 Bcast:192.168.71.255 Mask:255.255.255.0 inet6 addr: fe80::xxx:xxxx:xxxx:xxxx/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2138 errors:0 dropped:0 overruns:0 frame:0 TX packets:2118 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:278492 (271.9 KiB) TX bytes:273666 (267.2 KiB) Base address:0xecc0 Memory:d2fe0000-d3000000 |
In the above example, it is possible to use eth0 and eth1. For details regarding ifconfig command, refer to the Linux manual.
Information
When using Tagged VLAN, ensure that the NIC supports tagged VLAN functionality (IEEE 802.1Q). In addition, in a Redundant Line Control function, the effective range of VLAN-ID which can be specified is from 1 to 4094.
(3) Checking the name service
When using name services such as DNS or NIS, define keywords such as hosts in /etc/nsswitch.conf file to first refer to the local file. This allows to solve the address even if the DNS, NIS or LDAP sever is unreachable. The following is an example of /etc/nsswitch.conf.
# # /etc/nsswitch.conf # # An example Name Service Switch config file. This file should be # sorted with the most-used services at the beginning. # # The entry '[NOTFOUND=return]' means that the search for an # entry should stop if the search in the previous entry turned # up nothing. Note that if the search failed due to some other reason # (like no NIS server responding) then the search continues with the # next entry. # # Legal entries are: # # nisplus or nis+ Use NIS+ (NIS version 3) # nis or yp Use NIS (NIS version 2), also called YP # dns Use DNS (Domain Name Service) # files Use the local files # db Use the local database (.db) files # compat Use NIS on compat mode # hesiod Use Hesiod for user lookups # [NOTFOUND=return] Stop searching if not found so far # # To use db, put the "db" in front of "files" for entries you want to be # looked up first in the databases # # Example: #passwd: db files nisplus nis #shadow: db files nisplus nis #group: db files nisplus nis passwd: files shadow: files group: files #hosts: db files nisplus nis dns hosts: files dns ..... |
Information
If the host name rather than the IP address is used in setting GLS, enable the hostname resolution function (set by hanetparam -h), which allows you to change the host name to the IP address using only the /etc/hosts file without depending on the /etc/nsswitch.conf file setting.
(4) IPv6 RA daemon configuration
GLS supports radvd(router advertisement daemon for IPv6) for RA (router advertisement) daemon. To use IPv6(dual) on Fast switching mode, you must start RA daemon on the host running GLS in order to transmit RA from virtual interfaces. Other than this purpose, RA daemon is not necessary. The following describes configuration procedure.
radvd configuration
Define the configuration in /etc/radvd.conf.
When transmitting network information (network fec0:1::, prefix length 64) from sha0 with RA(router advertisement), define the configuration description as shown below.
In addition, the difference in the version of radvd needs to define the following kernel parameters (net.ipv6.conf.all.forwarding=1) in /etc/sysctl.conf.
For details on radvd configuration, refer to radvd manual.
interface sha0 { AdvSendAdvert on; # Sending router advertisements MinRtrAdvInterval 3; MaxRtrAdvInterval 10; prefix fec0:1::0/64 # Sending Prefix fec0:1::0/64 from sha0 { AdvOnLink on; AdvAutonomous on; AdvRouterAddr on; }; }; |
Configure radvd to startup during system startup (when run level is 2, 3, or 5).
# chkconfig --level 235 radvd on |
Verify radvd is configured to startup on run level 2, 3, 5.
# chkconfig --list radvd radvd 0:off 1:off 2:on 3:on 4:off 5:on 6:off |
(5) Route configuration
Route configuration of IPv4 or IPv6 is described below.
Default gateway configuration
Define the default gateway address (GATEWAY) in the "/etc/sysconfig/network" file.
/etc/sysconfig/network
GATEWAY=192.168.1.254 |
For IPv6 and NIC switching mode, configure the setting by using the user command execution function instead of the network configuration file for the operating system. In the configuration file of the user command execution function, define the operating system command assigning the route of IPv6 to be executed after activation of the IP address. For details, refer to the Linux manual (ip(6), route(8) and so on).
Information
The default gateway device (GATEWAYDEV) can not be configured for a physical interface bound with NIC switching mode.
When defining the default gateway (GATEWAY) in the "/etc/sysconfig/network-scripts/ifcfg-ethX" file in the NIC switching mode, add the same configuration of GATEWAY in the configuration files of all NICs bound by GLS. Note that if different configurations of GATEWAY are defined in the "/etc/sysconfig/network" file and the "/etc/sysconfig/network-scripts/ifcfg-ethX" file, the configuration in the "/etc/sysconfig/network-scripts/ifcfg-ethX" file has a priority.
# /opt/FJSVhanet/usr/sbin/hanetconfig print [IPv4,Patrol / Virtual NIC] Name Hostname Mode Physical ipaddr Interface List +-----------+---------------+----+-----------------+-------------- sha0 192.168.1.10 e eth1,eth2 # cat /etc/sysconfig/network-scripts/ifcfg-eth1 DEVICE=eth1 (omitted) GATEWAY=192.168.1.254 # cat /etc/sysconfig/network-scripts/ifcfg-eth2 DEVICE=eth2 (omitted) GATEWAY=192.168.1.254 |
If you do not use the "/etc/sysconfig/network" file in the environment where Virtual NIC mode is used, configure the route in the "/etc/sysconfig/network-scripts/ifcfg-shaX" file. You do not need to configure it in the "/etc/sysconfig/network-scripts/ifcfg-ethX" file in the same way as NIC switching mode. For details, see "3.3.3 Virtual NIC mode".
Static route configuration
To configure a static route on a routing table, define the configuration on the "/etc/sysconfig/network-scripts/route-Interface name" file.
NIC switching mode
Apply the same setting for both physical interfaces (route-ethX, route-ethY) bundled by NIC switching mode.
/etc/sysconfig/network-scripts/route-ethX
GATEWAY0=192.168.40.10 NETMASK0=255.255.255.0 ADDRESS0=192.168.100.0 |
/etc/sysconfig/network-scripts/route-ethY
GATEWAY0=192.168.40.10 NETMASK0=255.255.255.0 ADDRESS0=192.168.100.0 |
Virtual NIC mode and GS linkage mode
Configure it for the virtual interface (route-shaX).
/etc/sysconfig/network-scripts/route-shaX
GATEWAY0=192.168.40.10 NETMASK0=255.255.255.0 ADDRESS0=192.168.100.0 |
Information
When activating the interface on the standby side where the static route was set on startup of the operating system, the message of "RTNETLINK answers:" may be output. This is because the IP address is not set to the setting file (ifcfg-ethY) of the interface on the standby side. Ignore this message.
Virtual interface mode and GS linkage mode
Apply the setting to virtual interfaces (route-shaX).
/etc/sysconfig/network-scripts/route-shaX
GATEWAY0=192.168.40.10 NETMASK0=255.255.255.0 ADDRESS0=192.168.100.0 |
(6) IPv6 module configuration
Supporting IPv6(dual) on GLS Fast switching mode or NIC switching mode, or performing the IPv6 communication on Virtual NIC mode, it is required to load IPv6 module on Linux. See the following for configuring IPv6 module.
Loading ipv6 module
Configure /etc/sysconfig/network to use IPv6. Ignore the tunneling configuration, since GLS does not support tunneling feature.
NETWORKING_IPV6=yes IPV6TO4INIT=no |
In the environment with RHEL5 or earlier, when the setting which disables IPv6 is defined in /etc/modprobe.conf, delete the setting or comment it out.
# alias net-pf-10 off # alias ipv6 off # options ipv6 disable=1 |
Reboot the system.
# /sbin/shutdown -r now |
Verify IPv6 module is loaded.
# lsmod | grep ipv6 ipv6 662756 10 |
(7) Tagged VLAN configuration
To use tagged VLAN interfaces in Fast switching mode, NIC switching mode, or Virtual NIC mode, set up the tagged VLANs as follows:
Loading tagged VLAN module
Configure /etc/sysconfig/network to use tagged VLAN.
VLAN=yes |
Reboot the system.
# /sbin/shutdown -r now |
Verify tagged VLAN module is loaded.
# lsmod | grep 8021q 8021q 18760 1 |
(8) Network hotplug configuration
To use GLS in RHEL5, edit the following file to deactivate the hotplug function for the virtual interface (sha). You do not need to edit the file below in RHEL6.
/etc/udev/rules.d/60-net.rules
SUBSYSTEM=="net", ENV{INTERFACE}=="sha*", GOTO="skipgls" ACTION=="add", SUBSYSTEM=="net", IMPORT{program}="/lib/udev/rename_device" SUBSYSTEM=="net", RUN+="/etc/sysconfig/network-scripts/net.hotplug" LABEL="skipgls" |
Note
When PTF of RHEL is applied, the "/etc/udev/rules.d/60-net.rules" file may return to the content before it is edited. Edit the file after application.
Common settings between IPv4 and IPv6
To create backup of the physical interface settings (the "/etc/sysconfig/network-scripts/ifcfg-ethX" or "/etc/sysconfig/network/ifcfg-ethX" file), the file name must begin with names other than "ifcfg-".
(e.g. bak_ifcfg-ethX)
If the file name begins with "ifcfg-", OS might recognize the interface as an interface to be activated during system startup.
When using an IPv4 address
Define the IPv4 address (virtual IP address, physical IP address, logical virtual interface, takeover virtual IP address) and a host name in /etc/hosts file. These host names must be specified in the /etc/hosts file even if no host names but IP addresses are directly specified in environment definitions.
Before defining a virtual interface, the physical interface you are going to apply must be in active state and be sure the IPv4 address is assigned. (When the system is RHEL, in the file /etc/sysconfig/network-scripts/ifcfg-ethX, define "ONBOOT=yes" and "IPADDR=X.X.X.X" then reboot the system.)
When using an IPv6 address
Define the IPv6 address (logical virtual interface, takeover virtual IP address) and a host name in /etc/hosts file.
Before defining a virtual interface, the physical interface you are going to apply must be in active state and be sure the IPv6 link-local address is assigned. (When the system is RHEL, in the file /etc/sysconfig/network-scripts/ifcfg-ethX, define "ONBOOT=yes" and in the /etc/sysconfig/network file, define "NETWORKING_IPV6=yes" then reboot the system.)
You must start the radvd daemon on 2 or more servers running as Fast switching mode in order to set the stateless address auto-configuration. Note that when starting up radvd on multiple servers, synchronize the prefix data of the virtual interfaces defined in /etc/radvd.conf between the servers. An example of setting a /etc/radvd.conf file when using a Linux server as an IPv6 router is shown below. In addition, depending on the version of radvd, kernel parameter(net.ipv6.conf.all.forwarding=1) must be defined in /etc/sysctl.conf file. For details on this topic, see the manual of radvd.conf(5) and radvd(8).
/etc/radvd.conf
interface sha0 { AdvSendAdvert on; # Sending router advertisements MinRtrAdvInterval 3; MaxRtrAdvInterval 10; prefix fec0:1::0/64 # Sending Prefix fec0:1::0/64 from sha0 { AdvOnLink on; AdvAutonomous on; AdvRouterAddr on; }; }; |
When setting stateless address auto-configuration against the physical interface (ethX) using Fast switching mode, start up radvd on either of the servers or provide IPv6 router on the same network. Additionally, stateless address auto-configuration on the physical interface (ethX) does not apply on the server that has kernel configuration (net.ipv6.conf.all.forwarding=1) for starting up radvd. In such case, after starting radvd, use sysctl command to reconfigure the kernel parameter. For details, refer to sysctl(8) manual.
# sysctl -w net.ipv6.conf.all.forwarding=0 |
Common settings between IPv4 and IPv6
To create backup of the physical interface settings (the "/etc/sysconfig/network-scripts/ifcfg-ethX" or "/etc/sysconfig/network/ifcfg-ethX" file), the file name must begin with names other than "ifcfg-".
(e.g. bak_ifcfg-ethX)
If the file name begins with "ifcfg-", OS might recognize the interface as an interface to be activated during system startup.
When using an IPv4 address
Define the IPv4 address (virtual IP address, physical IP address, monitored IP addresses to be specified in monitoring destination information) and a host name in /etc/hosts file. These host names must be specified in the /etc/hosts file even if no host names but IP addresses are directly specified in environment definitions.
Before booting an OS, the primary interface (physical interface) you are going to apply must in active state and make sure IPv4 address is assigned. (When the system is RHEL, in the file /etc/sysconfig/network-scripts/ifcfg-ethX, define "ONBOOT=yes" and in the /etc/sysconfig/network file, define "NETWORKING_IPV6=yes".)
Also, make sure the secondary interface (physical interface) is in inactive state.
For Redundant Line Control Function, the path information must be initialized and the routing daemon must be restarted. If path information is statically specified, the static paths must be described in a configuration file for routing daemon.
When using an IPv6 address
Define the IPv6 address (takeover virtual IP address, monitored IP addresses to be specified in monitoring destination information) and a host name in /etc/hosts file.
For communication using IPv6, use the virtual IP address defined for GLS. When using the IPv6 address assigned by stateless address auto-configuration, it will be changed according to the change of the link-local address assigned for the active NIC around the time of NIC switching.
Before booting an OS, the primary interface (physical interface) you are going to apply must in active state and be sure IPv6 address is assigned. (When the system is RHEL, in the file /etc/sysconfig/network-scripts/ifcfg-ethX, define "ONBOOT=yes" and in the file /etc/sysconfig/network, set "NETWORKING_IPV6=yes")
Also, make sure the secondary interface (physical interface) is in inactive state.
Do not set the server running NIC switching mode as an IPv6 router.
When using IPv6 virtual interfaces in the environment where stateless address auto-configuration by an IPv6 router is not set, assign the link-local address of monitored HUB to the monitored IP.
Edit the setting (/etc/sysconfig/network-scripts/ifcfg-ethX file) for the physical interface of the GLS bundles as follows:
Item | Value (Example) | Description |
---|---|---|
DEVICE | ethX | Specify the device name. Set to "ethX". |
BOOTPROTO | static | Specify the protocol when getting the IP address. Set to "static" or "none". |
HWADDR | XX:XX:XX:XX:XX:XX | Specify the MAC address of the device. |
HOTPLUG | no | Specify use of hotplug. Set to "no". Set to "no". |
ONBOOT | yes | Select whether to start the physical interface on startup of the operating system. Set to "yes". |
DEVICETYPE | hanet | Specify the type of the device. Set to "hanet". |
An example is shown below.
Example of /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0 BOOTPROTO=static HWADDR=AA:AA:AA:AA:AA:AA HOTPLUG=no ONBOOT=yes DEVICETYPE=hanet |
To create backup of the physical interface settings (the "/etc/sysconfig/network-scripts/ifcfg-ethX" file), the file name must begin with names other than "ifcfg-".
(e.g. bak_ifcfg-ethX)
If the file name begins with "ifcfg-", OS might recognize the interface as an interface to be activated during system startup.
Point
For the physical interfaces bundled by GLS, do not specify "TYPE=Ethernet". Otherwise, GLS will not work properly.
IPADDR and other addresses are not required, so do not make any settings.
Specify the HWADDR and "ONBOOT=yes" to avoid renaming the device on startup of the operating system.
Set HWADDR to the unique MAC address of the NIC.
For the ifcfg-ethX configuration of the physical interfaces bundled in the GLS virtual interface, add the item "DEVICETYPE=hanet".
If you use "HWADDR=XX:XX:XX:XX:XX:XX" in the setting (the /etc/sysconfig/network-scripts/ifcfg-ethX file) for the physical interface of the GLS bundles, add the "HOTPLUG=no" setting.
To create backup of the physical interface settings (the "/etc/sysconfig/network-scripts/ifcfg-ethX" or "/etc/sysconfig/network/ifcfg-ethX" file), the file name must begin with names other than "ifcfg-".
(e.g. bak_ifcfg-ethX)
If the file name begins with "ifcfg-", OS might recognize the interface as an interface to be activated during system startup.
Define the IPv4 address (virtual IP address, physical IP address, logical virtual interface, takeover virtual IP address) and a host name in /etc/hosts file. These host names must be specified in the /etc/hosts file even if no host names but IP addresses are directly specified in environment definitions.
Before defining a virtual interface, the physical interface you are going to apply must be in active state and be sure the IPv4 address is assigned. (When the system is RHEL, in the file /etc/sysconfig/network-scripts/ifcfg-ethX, define "ONBOOT=yes" and "IPADDR=X.X.X.X" then reboot the system.)
Be sure to define the virtual gateway in the /etc/sysconfig/network-scripts/route-shaX file to set the static route information.
You do not need to configure the routing daemon for the network setting when using this method.