NAV offers several ways to organize your seed data and collected information.
Organizing your seed data¶
NAV’s data model allows you to organize your IP device information in several ways:
Define physical or conceptual locations of your operation. These usually map to geographical domains, such as locations for each campus in a multi-campus university.
Define your rooms, such as server rooms and wiring closets, where your equipment is actually wired up, and organize these into each of your locations. Geographical point coordinates can be attached to each room, for display on a map.
Define an organizational hierarchy. This can be used to set which organizational unit is responsible for the operation of each IP device you monitor, and to map your subnets and VLANs to organizational units.
Define arbitrary groups of devices. Some examples: Grouping all your print servers, your web servers, or just grouping your most critical routers. These groups can later be used for filtering alerts.
All these definitions are entered into NAV through the SeedDB tool, where you added your devices in the getting started guide. If you alreadu have much of this information in electronic format, it too can be bulk imported into NAV using the text formats described in each SeedDB tab.
As seen in that guide, NAV ships with an example location, room and organizational unit. These are fine for just trying things out, but to truly take advantage of NAV’s capabilities and gain control of your network inventory, you should take spend some time getting organized.
Organizing your collected data¶
Once your routers and switches are being monitored, NAV will collect information about your subnets and VLAN assignments, how these relate to each other, and store them in its database.
VLANs and subnets¶
NAV supports two different conventions for router port descriptions; if adopted, these allow NAV to categorize each VLAN into a usage category and an owner (an organizational unit from the previously defined hierarchy).
Read the guidelines in our wiki to learn how to do this. Usage categories are also defined using the SeedDB tool.
Network prefixes and scopes¶
Each network prefix configured in your device’s interfaces will be collected and stored by NAV. To the best of its ability, NAV will correlate which VLAN corresponds to each prefix. Sometimes, VLANs will have multiple prefixes, in the case of secondary network addresses or when both IPv4 and IPv6 are deployed in a subnet.
NAV features a subnet matrix tool, which charts your subnet allocations and their utilization percentages. To take advantage of this, you must manually add one ore more scope prefixes through the SeedDB Prefix tab. Each scope prefix will usually correspond to an IP address block you have been assigned by your regional internet registry (in some cases, you may want to subdivide those further, for your own organizational purposes).
You can also add reserved prefixes using SeedDB. These are typically prefixes within your scope that aren’t routed by your equipment, but possibly reserved for third parties or some future use. The subnet matrix tool will highlight these address ranges accordingly.
Vendors and device types¶
NAV will automatically discover and assign device types to SNMP-enabled devices being monitored. Each device type corresponds to a unique sysObjectID. An SNMP-enabled device will usually report a vendor-specific and unique sysObjectID, which may map to some specific device model, type and/or software.
Each device type in NAV is mapped to a vendor ID, a moniker such as
juniper. You can edit your device types and vendors through
the SeedDB tool.
When NAV sees a previously unknown sysObjectID, it will automatically
register a new device type and attach it to the
unknown vendor id. You may
wish to later edit these auto-created device types using the SeedDB Type
tab, to set the correct vendor id and a more proper type name and description.
Cabling and patching¶
If desireable, you can also document your cabling plans and your patch panels, using SeedDB. This would enable NAV to tell you to which office each switch port is patched through to (unless you are already diligent and add this information to the switch port description when patching).