Softpanorama (slightly skeptical) Open Source Software Educational Society

May the source be with you, but remember the KISS principle ;-) Softpanorama Search

Solaris route command

Routing is the process of forwarding a packet from one computer to another. It is based on an address in the message’s layer-3 header. But there’s more involved than just reading a layer-3 address, though.

The Solaris IP stack maintains a special database called a routing table. When Solaris forwards a packet, it first refers to the routing table to decided where to send the packet. If the host tries to connect to a host which is the same network it just sends packet to the destination (direct routing). If the target host is on another network than host's own, then it sends the data to the router (indirect routing). Often only one router is defined as the 'default gateway'.  In this case from the host’s point of view it outsources all the processing of "non-local" packets to this gateway. If there is no default gateway and the router doesn’t find an entry in its routing table that matches the destination network address, the router discards the packet.

The Solaris route command enables manual manipulation of the route table. The routing table contains a list of networks that the router knows it can send packets to, and states over which link to send the packets. These can be static routes, (entered by a system administrator) they can be default routes (a special static route to send unknown destination IP addresses to), or they can be dynamically discovered routes, found using a routing protocol. In case of dynamic routing routers talk to each other and share information about networks they are connected to. That sounds scary and it really is. It also creates the infinite possibilities of misunderstanding. At the same time, dynamic routing takes the building of the routing table out of your hands and puts it into the hands of specific routing protocol. Your job then becomes the monitoring of the routing tables to make sure that the routers are playing nicely with each other. See Routing for more details.

Route command gives the possibility to specify static routes. The syntax of the command is somewhat complicated:

route [-fn] add | delete | get [net|host|default] [destination] [gateway]

With keywords add and delete the default for optional [net|host|default] troika is host.

Here are some examples that might help to learn the intricacies of the command: 

• Add a route
  # route add net 128.50.3.0 bigbox
   
• Add a default route [host|net] dest. [ gateway]
  # route add default  maingw
   
• Delete a route
  # route delete net 128.50.3.0 sword-r
  
• Lookup and display the route for a destination:
  # route get 128.50.2.0
  
• Get routing reports continuously
  # route monitor
  
• Flush the routing table:
  # route flush
  
• Add the multicast path for 224.0.0.0:
  # route add 224.0.0.0 `uname -n` 0
  
• Use the "route add net" command with the -netmask option to make the route command to take the netmask specified on the command line
  
  # route add net 192.168.68.0 128.50.1.250 1 -netmask 255.255.255.192

Notes:

1. Netmask  is used to specify which part of the address is the network number part, and which is the host part. This is accomplished by a logical bitwise-AND between the Netmask and the IP address. The result specifies the network number. For Class C, the Netmask will always be 255.255.255.0; for Class B, the Netmask will always be 255.255.0.0; and so on.
2. The metric between two machines increases by one each time a new router (gateway) is encountered in the path. RIP routing protocol automatically chooses the path with the lowest metric. The metric information for a path is kept in the kernel's routing table in cache.
3. When deleting entries from or flushing the routing table, the process in.routed stops listening for RIP broadcasts but continues listening for RDISC advertisements. This freezes the current table. The appropriate process must be manually restarted to have it continue listening for RIP broadcasts.

Specifing default route with /etc/defaultrouter

As we already discussed there are two types of routing:

• Direct routing is used if the packet does not need to be forwarded, i.e. both the source and destination addresses have the same network number. In this case the packet is simply sent to destination as specified by the host part of IP address.
• Indirect routing is used when the network numbers of the source and destination do not match. This is the case where the packet must be forwarded by a node that supposedly knows how to reach the destination (a router). The simplest form of indirect routing is default routing. In this case any packet not suitable for direct routing is forwarded to the same router. This type of routing is often used for workstations and in cases where network segment has a single border router (choking point).

A default route is a route table entry that allows a host to define default routers to use if no other specific route is available. The default routers must be reliable. There is no need to define every reachable network. All indirectly connected packet destinations go to the default router.

A default router is identified by entries in /etc/defaultrouter.  It contains hostname or IP address entries that correspond to one or more routers. Upon rebooting, this prevents the startup script (/etc/rc2.d/S69inet) from launching the in.routed and in.rdisc dynamic router processes. Default route table entries may also be added by the in.rdisc daemon.

Advantages of default routing are:

• The /etc/defaultrouter file prevents additional routing daemons from starting.
  
• A default route is suitable when only one router is used to reach all indirectly connected destinations.
  
• A single default route entry results in a smaller routing table.
  
• Multiple default routers can be identified, which eliminate single points of failures within a network.
  
• A default route can run other routing protocols on the route.

Disadvantages of default routing:

• The default entry is always present, even when the default router is shut down. The host does not learn about other possible paths.
  
• All systems must have the /etc/defaultrouter file configured. This may be a problem on large, evolving networks.
  
• ICMP redirects occur if more than one router is available to a host.

Quiz

Q1. Which command manipulates the routing table ?

 A: route

 Q2: Command to add a route to server1 via router1:

 A: route add host server1 router1

Q3. Which command will add a route to the network 128.251.0.0 via 128.144.1.254 ?

 A: route add net 128.251.0.0 128.144.1.254

Q4. Which command would delete a route to host server1 via router1 ?

 A: route delete host server1 router1

Q5. Which command will delete the route to the network 128.251.0.0 via 128.144.1.254 ?

 A: route delete net 128.251.0.0 128.144.1.254

Q6. Which command will remove all entries from the routing table ?

 A: route flush

Q7. Which command allows you to monitor the routing requests that are not getting resolved from the routing table ?

 A7: route monitor

Q8. Which script starts the in.named daemon ?

A:  Depend on the Solaris version:

• /etc/init.d/inetsvc

Recommended Links

Internal links

• Solaris Routing
• Subnetting and VLSM

External links

• Routing with Solaris How to get your packets from point a to point b by Lance Spitzner
• Internetworking Technology Handbook - Routing Basics - Cisco Systems
• Cisco Router Configuration Tutorial
• Cisco router configuration tutorial