Skip navigation

Product Blog

3 Posts authored by: dexml
dexml

Routing Basics

Posted by dexml Jan 10, 2014

This blog is meant to provide a simple overview of routing in general, for those new to networking, and as an addendum to my other two routing related blogs:

  1. Routing Tools Overview for Network Performance Monitor 10.5
  2. TROUBLESHOOTING WITH ROUTING TOOLS FOR NETWORK PERFORMANCE MONITOR 10.5

 

While network and internet routing can be quite complex, the fundamentals are easy to comprehend.

 

  • If I am at “Location A” and want to get to “Location Z” 
    • I might be able to take a direct route and get there without stopping. 
    • In other cases I might have to use an alternate route, going through “Location B” or “Location X” first along the way.

               01_locations.jpg

Let’s take a look at a simple network topology map and see what a route might look like connecting networks together from across the world.

 

  • Sender 1 - on the left side of the map - is connected to Router 1 (R1) in California.
  • Receiver 2 - on the right side of the map - is connected to Router 4 (R4) in Paris
  • Router 2 (R2)- in the middle of the map - is connected to both Router 1 and Router 4.

     01_routes.jpg

As you can see, if you want to send packets from Sender 1 to Receiver 2 there is only one route to take. Any device connected to Router 1 must go through Router 2 in order to get data to any other device connected to Router 4. In this scenario:

 

  • The “Origin” is Sender 1
  • The “Next Hop” is Router 2
  • And the “Destination” is Receiver 2

 

I am sure you can imagine other possible scenarios where additional Routers could serve as the “Next Hop” similar to Router 2 connecting Router 1 to Router 4. For example there might be a “Next Hop” Router in Miami in addition to New York, in which case you would have multiple pathways (routes) that could make connections between the Origin and Destination.

 

To summarize, there can be multiple routes that packets can take through networks to get to any given destination. With the new Routing tools now provided by SolarWinds NPM, you know have a view of end to end routing for easier troubleshooting of network issues.

 

Protocol Overview


A routing protocol defines how routers communicate with one another. Routers need to communicate with one another to share information that will enable them to select routes between any two network nodes. Routers can only communicate directly with their neighbors but information is passed along, one to another, so that eventually a router can gain and store an entire network topology. There are several different routing protocols in use today for both internal networks and external networks such as the internet. By “Internal” network we really mean an “Autonomous System” or “Autonomous Network” which is defined as a network that is under the domain and control of a single Administrative entity such as a corporate network.  Protocols for internal use are referred to as “interior” protocols while protocols used on the internet are referred to as… you guessed it, “exterior” protocols.

 

The SolarWinds routing resources and tools contained in Network Performance Monitor version 10.5 and above provides support for the following popular network protocols:

 

  • OSPF
  • RIP
  • BGP

 

These are important networking protocols that allow network traffic to traverse vast internal networks and span a wide array of self-contained but interconnected networks such as the internet. The first two, OSPF and RIP, are interior routing protocols otherwise known as “Interior Gateway Protocols” or IGP. IGPs share information within only one routing domain and cannot span across separate autonomous networks, the biggest impact of this limitation is that these protocols cannot be used on the internet. The third in the list is BGP or Border Gateway Protocol which is currently the most commonly used exterior gateway protocol and which allows all manner of autonomous networks to communicate across the internet.

 

To sum it up internal networks can use OSPF, RIP (or other IGP protocols) and if any of these internal networks need to communicate with other networks they can and most likely will use BGP to send packets across the internet.

Note: There are a few other protocols in use today such as EIGRP and we are working on integrating these into NPM for a future release. 

 

Now if only I could find a way to insert myself into a network packet, I could have lunch in Paris and be back to work in Austin for my 1pm meeting!



Last week I provided an overview of the routing tools for Network Performance Monitor 10.5 and this week I will walk you through a troubleshooting scenario using the Interactive Online Demo to help illustrate how you can use these tools to troubleshoot real world issues in your own environment.

 

Obviously, routing is one of the keys to a properly functioning network. IP packets on your network must be routed, not only to their given destinations, but should also arrive in the fastest way possible. The new Routing tools for NPM have combined the Logical and Physical layers to help you get things done.

 

Troubleshooting Walk-Through


Troubleshooting StepsImage
  1. Let’s first take a look at the High Errors and Discards Today resource located about halfway down the page on the NPM Summary Tab in Orion.
    1. Notice the node labeled “Router5.lab.local” has a significant amount of discards.
    2. Click on “Router5.lab.local” (hereafter referred to as “Router 5”) to drill down and see the Node Details Page for that Router.
walkthrough_01.jpg

2. Once on the Node Details page, click on the Network Tab.

 

The Network Sub View is where all of the Routing Resources are located.

walkthrough_02.jpg

3. As it so happens, “Router 5” has some issues transmitting packets.


The "STATUS" column of the Current Percent Utilization of Each Interface resource (middle of the page) shows two “Down” interfaces.

walkthrough_03.jpg

4. Checking the Routing Table, we can see that the Down interfaces ( FastEthernet0/0 - Fa0/0 and FastEthernet0/1 - Fa0/1 ) do not actually appear in the Routing Table.


(hint all of the interfaces in the routing table shown are "1/0" and the two down interfaces are 0/0 and 0/1 respectively).


This is good news in this case, since the Down status of those particular interfaces is not actually affecting routing.

 

The fact that the Routing Table shows all GREEN and the subnets are properly routed to the other interfaces on this device is a quick way to make that determination.

walkthrough_04.jpg

So not only does NPM provide the fundamental set of information needed to define routing – showing protocol, destination, subnet size, next hop and interface information -  it also combines node information allowing visibility into the real availability of nodes and interfaces. Hovering over elements, such as interfaces, activate Pop Ups (image on the right) providing more details and a performance overview.

 

With these resources available at our fingertips, NPM is now doing all the heavy lifting. No longer do we have to manually connect to routers via command line to pull in this data and we do not have to share a router’s credentials with the guy in IT three floors down!

 

Everything is real time and polled frequently, by default every 2 minutes but configurable to meet your particular needs.

walkthrough_05.jpg

5. We have determined our Routing Table is clean, so now let’s take a look at the Routing Neighbors resource which shows you the status of the routers that are directly “next” to the one you are viewing. In this instance we have:

 

  • Router 2 as one of our neighbors using the OSPF protocol.
  • And the second neighbor 74.115.12.1 which is using the BGP Protocol.

 

The fact that the second neighbor, running BGP shows up as IP Address 74.115.12.1 actually tells us that Orion is not monitoring this device as a node. If it were being monitored in NPM we would could drill down and view its Node Details Page as we can with Router2.lab.local.

 

The point here is that if you notice something like this in your own environment you would want to have NPM monitor it so all of that extra information is available to you.

walkthrough_06.jpg

Note: To fully monitor the router 74.115.12.1 - or any other device - you can go to the Admin section of NPM and either:

 

  1. Run a manual discovery via "Add a Node"
  2. Run an automated discovery via Sonar Network Discovery
walkthrough_06b.jpg

6. Now let’s check out the Top 10 Flapping Routes resource.

 

Flapping details are usually something that cannot be obtained directly from the router, and some other kind of analysis tool is needed to get this info. But now with NPM all of this is available in one place.

 

By providing the number of “Flaps” (how many times the path was announced for this particular router) insight can be gained into serious issues that need attention right now or emerging issues which might become more serious down the road.

 

In the image to the right, notice that for Destination Network 10.0.0.0 we can see 8 Flaps to Router2.lab.local.

 

From here, we could drill down into Router 2 for details about the interfaces, the statistics and historical data to find out what is going on to prevent further Flapping from occurring.

walkthrough_07.jpg

7. Finally, we might check the Default Route Change resource to learn if the default route has been added/removed/changed and if so when.

walkthrough_08.jpg


“Wonder Twin Powers… Activate!”

One final thought. Once you find out the source of a given Routing issue the next step you need to make, more often than not, is connecting to one or many routers and make configuration changes to finally resolve the issue.

 

If you have ever had to do this, you know that connecting to routers one by one via command line can be a major pain. So, it is worth pointing out that if you own SolarWinds Network Configuration Manager (NCM) you can fix problems such as duplicate subnet routing or add missing subnets more easily by pushing out new configurations through NCM directly to your routers without having to go through the slow and tedious process of manually connecting to each of your devices as you would have to do without using NCM.

 

These two tools used together, NPM and NCM, can save you a lot of time and frustration while helping you keep your network running smoothly. Oh, and if you were wondering about the “Wonder Twin Powers” reference, it is an old Saturday morning cartoon and

this video explains it quite well.

Ever wonder how the ancient merchants like Marco Polo figured out the best routes to travel around the world? Well, I don’t. But I do wonder how to get my data across the globe in the quickest way possible, and the routing tools in Network Performance Monitor (NPM) can help you do that.

 

There are five resources found in NPM 10.5 and later, that can help you troubleshoot routing issues to get your network back on track and optimized for maximum performance.

 

  • Routing Table
  • Top 10 Flapping Routes
  • Routing Neighbors
  • Default Route Changes
  • Routing Details

Routing_Network_Tab.jpg

Note: The easiest way to see all of the Routing Resources together by default, is to click the "Network" sub-view on a Router’s Node Details page.

 

The table below describes each of the aforementioned routing resources, providing a basic understanding of the feature set.

 

Resource Description

Routing Table  

routing_table.jpg       

Each Router has its own Routing Table and each Routing Table is potentially different for any given router in your network. As you can see the Routing Table information is displayed across 6 columns:

 

Destination Network - a list of networks you can reach from the Router you are on.


CIDR - The Classless Inter-Domain Routing for the given Destination Network.

 

Next Hop – The next router, or “next hop”, you need to go through to get to the given Destination Network.

 

Interface – The actual interface on the Next Hop router that the packets are sent through.

 

Metric – Routing Tables only keeps the best routes available and each protocol has its own of set metrics used to determine what the best route is. In general the lower the number the faster the route.

 

Source – The Source column shows the protocol being used.

Top 10 Flapping Routes

flapping-routes.jpg

The term “flapping” refers to a condition where a Router advertises a destination network via a particular route and then quickly sends another advertisement for a different route. When interfaces on a router go up and down unexpectedly or more frequently than they should, this causes a recalculation of routes in your Routing Tables. This actually slows down routing, and the slow network speed may result in outages or other connectivity issues.

 

The Top 10 Flapping Routes resource shares some columns in common with the Routing Table, so we will just focus on what is different and unique here:

 

  • Flaps – Shows you the number of Flaps that have occurred in the last selected time period (7 Days in the image shown). Notice that Yellow is a warning state whereas Red indicates a more severe issue.

 

  • Last Change – This gives you an indication of when a change was made.

 

  • Protocol - Shows the routing protocol used.

Routing Neighbors

routing-neighbors.jpg

The Routing Neighbors resource shows you which routers are directly connected to the router you are currently viewing and it provides status on this relationship. This table can be useful in the case where neighboring routers might be down or display other issues that might hinder the ability to route packets through your network.

 

The Routing Neighbors resources shows several more columns of information useful for troubleshooting:

 

  • Node Name - This is a clickable link that will take you directly to the Node shown, which makes it is to getting further information about a specific node.

 

  • Status - Give you information about a given router's status, helping with communication issues.

Default Route Changes

default-route_changes.jpg

The Default Route Changes resource provides a quick view of any changes made to default routes and help you narrow down when changes have occurred and help you correlate with other known data to help with your troubleshooting. You can set the view from the last 24 hours, the last 3 days, last 7 days, last 14 days and the last month.

Routing Details

routing-details.jpg

The Routing Details resource provides you a quick glance at when Protocols were last polled so you will know how fresh your routing data is. Special note, the row which says “Routing Table poller” is just showing you when the routing table was last polled by NPM for this device.

 

Summary

Customer feedback for the new Routing Tools in Network Performance Monitor has been overwhelmingly positive and the tools are doing a great job at what they are designed to do, speed up troubleshooting of routing issues and make your work life much easier!


Up Next

In my upcoming blog post we will walk through some troubleshooting scenarios using the Interactive Online Demo to give you a better sense of how to actually use the routing tools.

Filter Blog

By date: By tag: