7. Network¶
The Network section of the administrative GUI contains these components for viewing and configuring network settings on the FreeNAS® system:
- Global Configuration: general network settings.
- Interfaces: settings for each network interface.
- IPMI: settings controlling connection to the appliance through the hardware side-band management interface if the graphical user interface becomes unavailable.
- Link Aggregations: settings for network link aggregation and link failover.
- Network Summary: display an overview of the current network settings.
- Static Routes: add static routes.
- VLANs: configure IEEE 802.1q tagging for virtual LANs.
Each of these is described in more detail in this section.
7.1. Global Configuration¶
Network → Global Configuration,
shown in
Figure 7.1.1,
is for general network settings that are not unique to any particular
network interface.
Fig. 7.1.1 Global Network Configuration
Table 7.1.1 summarizes the settings on the Global Configuration tab. Hostname and domain fields are pre-filled as shown in Figure 7.1.1, but can be changed to meet requirements of the local network.
| Setting | Value | Description |
|---|---|---|
| Hostname | string | system host name |
| Domain | string | system domain name |
| IPv4 Default Gateway | IP address | typically not set (see NOTE below) |
| IPv6 Default Gateway | IP address | typically not set (see NOTE below) |
| Nameserver 1 | IP address | primary DNS server (typically in Windows domain) |
| Nameserver 2 | IP address | secondary DNS server |
| Nameserver 3 | IP address | tertiary DNS server |
| HTTP Proxy | string | enter the proxy information for the network in the format http://my.proxy.server:3128 or http://user@password:my.proxy.server:3128 |
| Enable netwait feature | checkbox | if enabled, network services are not started at boot until the interface is able to ping the addresses listed in Netwait IP list |
| Netwait IP list | string | if Enable netwait feature is checked, list of IP addresses to ping; otherwise, ping the default gateway |
| Host name database | string | used to add one entry per line which will be appended to /etc/hosts; use the format
IP_address space hostname where multiple hostnames can be used if separated by a space |
When Active Directory is being used, set the IP address of the realm’s DNS server in the Nameserver 1 field.
If the network does not have a DNS server, or NFS, SSH, or FTP users are receiving “reverse DNS” or timeout errors, add an entry for the IP address of the FreeNAS® system in the Host name database field.
Note
In many cases, a FreeNAS® configuration does not include default gateway information as a way to make it more difficult for a remote attacker to communicate with the server. While this is a reasonable precaution, such a configuration does not restrict inbound traffic from sources within the local network. However, omitting a default gateway will prevent the FreeNAS® system from communicating with DNS servers, time servers, and mail servers that are located outside of the local network. In this case, it is recommended to add Static Routes to be able to reach external DNS, NTP, and mail servers which are configured with static IP addresses. When a gateway to the Internet is added, make sure that the FreeNAS® system is protected by a properly configured firewall.
7.2. Interfaces¶
Network → Interfaces
shows which interfaces have been manually configured and allows adding
or editing a manually configured interface.
Note
Typically, the interface used to access the FreeNAS® administrative GUI is configured by DHCP. This interface does not appear in this screen, even though it is already dynamically configured and in use.
Figure 7.2.1
shows the screen that opens on clicking
Interfaces → Add Interface.
Table 7.2.1
summarizes the configuration options shown when adding an interface or
editing an already configured interface. Note that if any changes to
this screen require a network restart, the screen will turn red when
the OK button is clicked and a pop-up message will point
out that network connectivity to the FreeNAS® system will be
interrupted while the changes are applied.
Fig. 7.2.1 Adding or Editing an Interface
| Setting | Value | Description |
|---|---|---|
| NIC | drop-down menu | the FreeBSD device name of the interface; a read-only field when editing an interface |
| Interface Name | string | description of interface |
| DHCP | checkbox | requires static IPv4 or IPv6 configuration if unchecked; only one interface can be configured for DHCP |
| IPv4 Address | IP address | enter a static IP address if DHCP is unchecked |
| IPv4 Netmask | drop-down menu | enter a netmask if DHCP is unchecked |
| Auto configure IPv6 | checkbox | only one interface can be configured for this option; if unchecked, manual configuration is required to use IPv6 |
| IPv6 Address | IPv6 address | must be unique on network |
| IPv6 Prefix Length | drop-down menu | match the prefix used on network |
| Options | string | additional parameters from ifconfig(8), separate multiple parameters with a space; for example: mtu 9000 increases the MTU for interfaces which support jumbo frames |
This screen also provides for the configuration of IP aliases, making it possible for a single interface to have multiple IP addresses. To set multiple aliases, click the Add extra alias link for each alias. Aliases are deleted by clicking the interface in the tree, clicking the Edit button, checking the Delete checkbox below the alias, then clicking the OK button.
Warning
Aliases are deleted by checking the Delete checkbox in the alias area, then clicking OK for the interface. Do not click the Delete button at the bottom of this screen, which deletes the entire interface.
Multiple interfaces cannot be members of the same subnet. See Multiple network interfaces on a single subnet for more information. Check the subnet mask if an error is shown when setting the IP addresses on multiple interfaces.
This screen will not allow an interface’s IPv4 and IPv6 addresses to both be set as primary addresses. An error is shown if both the IPv4 address and IPv6 address fields are filled in. Instead, set only one of these address fields and create an alias for the other address.
7.3. IPMI¶
Beginning with version 9.2.1, FreeNAS® provides a graphical screen for configuring an IPMI interface. This screen will only appear if the system hardware includes a Baseboard Management Controller (BMC).
IPMI provides side-band management if the graphical administrative interface becomes unresponsive. This allows for a few vital functions, such as checking the log, accessing the BIOS setup, and powering on the system without requiring physical access to the system. IPMI can also be used to allow another person remote access to the system to assist with a configuration or troubleshooting issue. Before configuring IPMI, ensure that the management interface is physically connected to the network. The IPMI device may share the primary Ethernet interface, or it may be a dedicated separate IPMI interface.
Warning
It is recommended to first ensure that the IPMI has been patched against the Remote Management Vulnerability before enabling IPMI. This article provides more information about the vulnerability and how to fix it.
IPMI is configured from
Network → IPMI.
The IPMI configuration screen, shown in
Figure 7.3.1,
provides a shortcut to the most basic IPMI configuration. Those
already familiar with IPMI management tools can use them instead.
Table 7.3.1
summarizes the options available when configuring IPMI with the
FreeNAS® GUI.
Fig. 7.3.1 IPMI Configuration
| Setting | Value | Description |
|---|---|---|
| Channel | drop-down menu | select the channel to use |
| Password | string | enter the password used to connect to the IPMI interface from a web browser |
| DHCP | checkbox | if left unchecked, the following three fields must be set |
| IPv4 Address | string | IP address used to connect to the IPMI web GUI |
| IPv4 Netmask | drop-down menu | subnet mask associated with the IP address |
| IPv4 Default Gateway | string | default gateway associated with the IP address |
| VLAN ID | string | enter the VLAN identifier if the IPMI out-of-band management interface is not on the same VLAN as management networking |
After configuration, the IPMI interface is accessed using a web browser and the IP address specified in the configuration. The management interface prompts for a username and the configured password. Refer to the IPMI device’s documentation to determine the default administrative username.
After logging in to the management interface, the default administrative username can be changed, and additional users created. The appearance of the IPMI utility and the functions that are available vary depending on the hardware.
7.4. Link Aggregations¶
FreeNAS® uses FreeBSD’s lagg(4) interface to provide link aggregation and link failover. The lagg interface allows aggregation of multiple network interfaces into a single virtual lagg interface, providing fault-tolerance and high-speed multi-link throughput. The aggregation protocols supported by lagg determine which ports are used for outgoing traffic and whether a specific port accepts incoming traffic. The link state of the lagg interface is used to validate whether the port is active.
Aggregation works best on switches supporting LACP, which distributes traffic bi-directionally while responding to failure of individual links. FreeNAS® also supports active/passive failover between pairs of links. The LACP, FEC, and load-balance modes select the output interface using a hash that includes the Ethernet source and destination address, VLAN tag (if available), IP source and destination address, and flow label (IPv6 only). The benefit can only be observed when multiple clients are transferring files from the NAS. The flow entering into the NAS depends on the Ethernet switch load-balance algorithm.
The lagg driver currently supports several aggregation protocols, although only Failover is recommended on network switches that do not support LACP:
Failover: the default protocol. Sends traffic only through the
active port. If the master port becomes unavailable, the next active
port is used. The first interface added is the master port; any
interfaces added after that are used as failover devices. By default,
received traffic is only accepted when received through the active
port. This constraint can be relaxed, which is useful for certain
bridged network setups, by creating a a tunable with a
Variable of net.link.lagg.failover_rx_all, a
Value of a non-zero integer, and a Type of
Sysctl in
System → Tunables → Add Tunable.
FEC: supports Cisco EtherChannel on older Cisco switches. This is a static setup and does not negotiate aggregation with the peer or exchange frames to monitor the link.
LACP: supports the IEEE 802.3ad Link Aggregation Control Protocol (LACP) and the Marker Protocol. LACP negotiates a set of aggregable links with the peer into one or more link aggregated groups (LAGs). Each LAG is composed of ports of the same speed, set to full-duplex operation. Traffic is balanced across the ports in the LAG with the greatest total speed; in most cases there will only be one LAG which contains all ports. In the event of changes in physical connectivity, link aggregation will quickly converge to a new configuration. LACP must be configured on the switch, and LACP does not support mixing interfaces of different speeds. Only interfaces that use the same driver, like two igb ports, are recommended for LACP. Using LACP for iSCSI is not recommended, as iSCSI has built-in multipath features which are more efficient.
Load Balance: balances outgoing traffic across the active ports based on hashed protocol header information and accepts incoming traffic from any active port. This is a static setup and does not negotiate aggregation with the peer or exchange frames to monitor the link. The hash includes the Ethernet source and destination address, VLAN tag (if available), and IP source and destination address. Requires a switch which supports IEEE 802.3ad static link aggregation.
Round Robin: distributes outgoing traffic using a round-robin scheduler through all active ports and accepts incoming traffic from any active port. This mode can cause unordered packet arrival at the client. This has a side effect of limiting throughput as reordering packets can be CPU intensive on the client. Requires a switch which supports IEEE 802.3ad static link aggregation.
None: this protocol disables any traffic without disabling the lagg interface itself.
Note
When using LACP, verify that the switch is configured for active LACP, as passive LACP is not supported.
7.4.1. LACP, MPIO, NFS, and ESXi¶
LACP bonds Ethernet connections to improve bandwidth. For example, four physical interfaces can be used to create one mega interface. However, it cannot increase the bandwidth for a single conversation. It is designed to increase bandwidth when multiple clients are simultaneously accessing the same system. It also assumes that quality Ethernet hardware is used and it will not make much difference when using inferior Ethernet chipsets such as a Realtek.
LACP reads the sender and receiver IP addresses and, if they are deemed to belong to the same TCP connection, always sends the packet over the same interface to ensure that TCP does not need to reorder packets. This makes LACP ideal for load balancing many simultaneous TCP connections, but does nothing for increasing the speed over one TCP connection.
MPIO operates at the iSCSI protocol level. For example, if four IP addresses are created and there are four simultaneous TCP connections, MPIO will send the data over all available links. When configuring MPIO, make sure that the IP addresses on the interfaces are configured to be on separate subnets with non-overlapping netmasks, or configure static routes to do point-to-point communication. Otherwise, all packets will pass through one interface.
LACP and other forms of link aggregation generally do not work well with virtualization solutions. In a virtualized environment, consider the use of iSCSI MPIO through the creation of an iSCSI Portal. This allows an iSCSI initiator to recognize multiple links to a target, utilizing them for increased bandwidth or redundancy. This how-to contains instructions for configuring MPIO on ESXi.
NFS does not understand MPIO. Therefore, one fast interface is needed, since creating an iSCSI portal will not improve bandwidth when using NFS. LACP does not work well to increase the bandwidth for point-to-point NFS (one server and one client). LACP is a good solution for link redundancy or for one server and many clients.
7.4.2. Creating a Link Aggregation¶
Before creating a link aggregation, double-check that no
interfaces have been manually configured in
Network → Interfaces → View Interfaces.
If any manually-configured interfaces exist, delete them as lagg creation fails if any interfaces are manually configured.
Note
Creating or editing link aggregations can disconnect clients using the FreeNAS® computer. Please verify that clients have saved their work and are not connected through the affected networks before making changes.
Figure 7.4.1
shows the configuration options when adding a lagg interface using
Network → Link Aggregations
→ Create Link Aggregation.
Fig. 7.4.1 Creating a lagg Interface
Note
If interfaces are installed but do not appear in the Physical NICs list, check that a FreeBSD driver for the interface exists here.
To create a link aggregation, select the desired Protocol Type. LACP is preferred. If the network switch does not support LACP, choose Failover. Highlight the interfaces to associate with the lagg device, and click the OK button.
Once the lagg device has been created, click its entry to enable its Edit, Delete, and Edit Members buttons.
Clicking the Edit button for a lagg opens the configuration screen shown in Figure 7.4.2. Table 7.4.1 describes the options in this screen.
If the network interface used to connect to the FreeNAS® web GUI is a member of the lagg, the network connection will be lost when the new lagg is created. The switch settings might also require changes to communicate through the new lagg interface.
The IP address of the new lagg can be set with DHCP or manually from the console setup menu. If the IP address is set manually, it might also be necessary to enter a default gateway to allow access to the GUI from the new lagg interface.
Fig. 7.4.2 Editing a lagg
| Setting | Value | Description |
|---|---|---|
| NIC | string | read-only; automatically assigned the next available numeric ID |
| Interface Name | string | by default same as device (NIC) name, can be changed to a more descriptive value |
| DHCP | checkbox | check if the lagg device will get IP address info from DHCP server |
| IPv4 Address | string | enter a static IP address if DHCP is left unchecked |
| IPv4 Netmask | drop-down menu | enter a netmask if DHCP is left unchecked |
| Auto configure IPv6 | checkbox | check only if DHCP server available to provide IPv6 address info |
| IPv6 Address | string | optional |
| IPv6 Prefix Length | drop-down menu | required if an IPv6 address is entered |
| Options | string | additional ifconfig(8) options |
This screen also allows the configuration of an alias for the lagg interface. Multiple aliases can be added with the Add extra Alias link.
Click the Edit Members button, click the entry for a member, then click its Edit button to see the configuration screen shown in Figure 7.4.3. The configurable options are summarized in Table 7.4.2.
Fig. 7.4.3 Editing a Member Interface
| Setting | Value | Description |
|---|---|---|
| LAGG Interface group | drop-down menu | select the member interface to configure |
| LAGG Priority Number | integer | order of selected interface within the lagg; configure a failover to set the master interface to 0 and the other interfaces to 1, 2, etc. |
| LAGG Physical NIC | drop-down menu | physical interface of the selected member |
| Options | string | additional parameters from ifconfig(8) |
Options can be set at the lagg level using the Edit button, or at the individual parent interface level using the Edit Members button. Changes are typically made at the lagg level (Figure 7.4.2) as each interface member will inherit from the lagg. To configure at the interface level (Figure 7.4.3) instead, the configuration must be repeated for each interface within the lagg. Some options can only be set on the parent interfaces and are inherited by the lagg interface. For example, to set the MTU on a lagg, use Edit Members to set the MTU for each parent interface.
Note
A reboot is required after changing the MTU to create a jumbo frame lagg.
To see if the link aggregation is properly load balancing, run this command from Shell:
systat -ifstat
More information about this command can be found at systat(1).
7.5. Network Summary¶
Network → Network Summary
shows a quick summary of the addressing information of every
configured interface. For each interface name, the configured IPv4 and
IPv6 addresses, DNS servers, and default gateway are displayed.
7.6. Static Routes¶
No static routes are defined on a default FreeNAS® system. If a static
route is required to reach portions of the network, add the route with
Network → Static Routes → Add Static Route,
shown in
Figure 7.6.1.
Fig. 7.6.1 Adding a Static Route
The available options are summarized in Table 7.6.1.
| Setting | Value | Description |
|---|---|---|
| Destination network | integer | use the format A.B.C.D/E where E is the CIDR mask |
| Gateway | integer | enter the IP address of the gateway |
| Description | string | optional |
Added static routes are shown in View Static Routes. Click a route’s entry to access the Edit and Delete buttons.
7.7. VLANs¶
FreeNAS® uses FreeBSD’s vlan(4) interface to demultiplex frames with IEEE 802.1q tags. This allows nodes on different VLANs to communicate through a layer 3 switch or router. A vlan interface must be assigned a parent interface and a numeric VLAN tag. A single parent can be assigned to multiple vlan interfaces provided they have different tags.
Note
VLAN tagging is the only 802.1q feature that is implemented. Additionally, not all Ethernet interfaces support full VLAN processing. See the HARDWARE section of vlan(4) for details.
Click
Network → VLANs → Add VLAN,
to see the screen shown in
Figure 7.7.1.
Fig. 7.7.1 Adding a VLAN
Table 7.7.1 summarizes the configurable fields.
| Setting | Value | Description |
|---|---|---|
| Virtual Interface | string | use the format vlanX where X is a number representing a vlan interface not currently being used as a parent |
| Parent Interface | drop-down menu | usually an Ethernet card connected to a properly configured switch port; note that newly created Link Aggregations will not appear in the drop-down until the system is rebooted |
| VLAN Tag | integer | number between 1 and 4095 which matches a numeric tag set up in the switched network |
| Description | string | optional |
The parent interface of a VLAN must be up, but it can have an IP
address or it can be unconfigured, depending upon the requirements of
the VLAN configuration. This makes it difficult for the GUI to do the
right thing without trampling the configuration. To remedy this, after
adding the VLAN, go to
Network → Interfaces → Add Interface.
Select the parent interface from the NIC drop-down menu
and in the Options field, type up. This will
bring up the parent interface. If an IP address is required, it can be
configured using the rest of the options in the
Add Interface screen.
Warning
Creating a vlan will cause network connectivity to be interrupted. Accordingly, the GUI will provide a warning and an opportunity to cancel the vlan creation.