Preparation for Fault Management
Introduction
Fault Management is one of the five network management categories defined by the International Standards Organization (ISO). The goal of Fault Management is to monitor and detect anomalous conditions on the network and, if a fault occurs, restore the network to a normal condition as rapidly as possible. A good Fault Management policy should be able to resolve a majority of network faults without human intervention.
You will utilize a local network that consists of six LANs with a total of 60 users. The users need to access the network database server and the network web server via a DS1 connection to the ISP. If that DS1 link should go down, the company employees will lose the ability to access the network. In addition, it is suspected that the DS1 link is being over-utilized and is causing database query response times and HTTP page response times to be slow. The Fault Management solution is to implement a redundant DS1 which will act as a backup for the primary DS1.
This implementation alone, however, would not increase response times. Therefore, we will implement load balancing along with the redundant DS1 solution such that if either DS1 goes down, the other can take over, but during normal operations they will share the network load. With this solution, we will have redundancy with load balancing and, as a result, response times for network services should improve.
Explanation and Background
When a serious fault occurs in the network, it usually means that a mission-critical link or device has gone down. This could result in network outages and user downtime. When this is the case, the company will lose money in any of several ways. First, the network outage may be causing your clients to be unable to access goods and the service from which you profit. Another reason is that employees may not be able to perform their jobs during the network outage because they cannot access the network resources that they require. Consequently, the organization is paying them a salary to wait for the network outage to be resolved. In this lab, the company must consider the fact that it is leasing the DS1 from a service provider. This adds a level of complexity to the Fault Management design. The DS1 might go down on the WAN side and cause a network outage for the organization without it being any fault of the firm’s whatsoever. A redundant link for such a mission-critical connection, therefore, would seem to be a wise decision.
Modus Operandi
Execute the steps as shown. In some instances, you may need to refer to the OPNET documentation to obtain additional information necessary to execute a step, solve a problem, or build the simulation. In other instances, you may need to do some research in your text or in online resources to understand and answer a question.
When you encounter the “ß 3.” symbol, you will be required to answer a question in the lab report. In the template for the lab report, answer the questions in the body of the text.
In this example, you would answer Question #3. Do the same each time you encounter this symbol followed by a number. There may be additional questions at the end of the lab.
When completed, hand in the Answers document according to your instructor. Be sure to retain an electronic copy.
The èsymbol denotes a note or a sidebar to which you should pay attention.
Procedure
Start OPNET Modeler
Double-click on the shortcut to the OPNET Modeler icon in the desktop or use the Start Menu.
Open the Scenario
1. Select File/Open .
2. Select the folder FaultManagement.project.
3. Choose “FaultManagement. ”
4. Click OK . The project FaultManagement and the Scenario no_redundancy should open.
1. The network has been created for you, but you will need to configure the elements.
Configure the Applications and Profiles
1. Right-click the Applications Config icon and select Edit Attributes .
2. Open the Applications Definitions tree and left-click on the 0 in the Rows column. You should get a pull-down menu. Select 2 . If you need to, you may simply edit the Rows to 2 .
3. Expand the trees on both of the new rows and enter Database as the Row 0 Application Name and HTTP as the Row 1 Application Name.
4. Open the Row 0 Description tree and, next to Database , left-click Off to see the dropdown menu. Select High Load .
5. Do the same for the Row 1 Description and, under HTTP , select Heavy Browsing .
You have used this object to define the way in which these two applications are being used on the network and, consequently, how the traffic they generate will be configured (type and volume).
6.Close the Application Config object by clicking OK . Save your project.
7.Right-click on the Profile Config object and select Edit Attributes .
8.Open the tree on Profile Configuration and add 1 row.
9.Next to Profile Name , enter Network User and press Enter .
10. Change Operation Mode from Serial (Ordered) to Simultaneous via the dropdown menu.
11. Open the Applications tree and add 2 rows. Remember that we created two applications.
12. Under Row 0 , click on the field next to Name . Note that you get a dropdown menu with the two applications we configured, Database andHTTP . Select Database for Row 0 and HTTP for Row 1 .
13. Be certain that your screen matches the one above and click OK . Save your project.
Configuring the Servers
1. Right-click the Web Server and select Edit Attributes.
2. Open the Applications tree and left-click the field next to Application: Supported Services .
3. You should get the (Applications: Supported Services) Table .
4. Left-click in the Rows field on the lower left and select 1 from the dropdown menu or edit the field to 1. You should get a row in the window.
5. Left-click on None and from the dropdown menu, select HTTP . You have just configured the web server to process HTTP. Click OK twice. Save your project.
6. Right-click on the Database Server and select Edit Attributes .
7. Follow the same steps that you just did on the web server to set the Applications: Support Services to Database . Click OK twice and save your project.
Configure the LANs
To this point, we created two applications that we will study: HTTP (web browsing) and Database. We created a Network User to use those two applications. At this point, we need to make this Network User “live” by configuring him or her in the LANs.
1. Left-click on LAN 1 (or any of the LANs) so that a circle appears around it. Right-click on the same LAN and choose Select Similar Nodes so that circles now appear around all of the LANs.
2. Right-click on the original LAN and select Edit Attributes . When the (LAN1) Attributes window appears, be sure that the box at the bottom,Apply changes to selected objects, is checked.
3. Open the tree on Applications: Supported Profiles and add 1 row .
4. Click on the field next to Profile Name and from the dropdown menu, select Network User . Leave the other fields the same. Your window should look like the one below.
5. Again, make sure that the Apply Changes… box is checked and click OK . You should receive a warning message alerting you to the fact that changes made to multiple objects cannot be undone by the Undo function but must be backed out of each individual node. Click Yes .
6. You might want to check to make sure that all of the LANs were changed. You have now configured all users on the six LANs (60 users) to use the applications that we have set up.
è We configured discrete traffic on the LANs by setting up the profiles and the users. In contrast, we could also set up Background traffic. This type of traffic would represent the loads generated by DNS, broadcasts, DHCP, and so on.
Configure the Stats to be Gathered
We have now configured the network to the state that we wish to test. Now we must configure the statistics that we need to see what is happening.The statistics we need are those that tell us how long it takes for the network users to download network resources. We also would like to look at the utilization of the T1 link.
1. Right-click on a blank area of the workspace and Choose Individual Statistics.
Expand Global Statistics.
Expand DB Query – select Response Time (sec) and Traffic Received (bytes/sec) .
Expand HTTP – select Page Response Time (seconds) and Traffic Received (bytes/sec).
2. Click OK.
3. Left-click on the DS1 link that connects Router A to the IP Cloud . Note that a dot appears in the middle of the link and the link becomes dashed.
4. Now, right-click on the dot and select Choose Individual DES Statistics . Expand point-to-point and select utilization <- .
5. Click OK . Save your project.
Configuring the Simulation Run
1. Select the Configuration/Run tab from the tool bar.
2. After clicking on it, the following screen will open.
3. Change the Duration to 15 and use the dropdown arrow to change the hour(s) to minute(s) .
4. Click Apply and then Cancel and save your project.
Run Simulation
(Method 1)
1. Select the Scenarios tab over the toolbar and select Manage Scenarios.
2. Change the Results from uncollected to collect , and then click OK to run the scenario.
3. After a Please Wait window, the Simulation Progress window will open and the scenario will run. When the simulation is complete, theClose button will be visible. Click Close.
(Method 2)
1. At the Configure/Run DES window where we changed the Duration and minutes () , click Run .
2. The Simulation Execution window will open and display the run stats.
3. When it completes, click Close .
Method 1 of running a sim is especially useful when we want to run two or more sims in one run to compare statistics between them. Method 2 is the more common method since we are usually dealing with only one scenario.
Results Analysis
1. To view the results of our simulation, click DES > Results > View Results OR click the View Results icon on the toolbar.
2. The Results Browser will open.Use the dropdown arrow to change the As Is in the bottom of the lower right-hand side of the screen toaverage . This way, we can look at the average values for all of our results.
Expand DB Query -> Select Response Time (sec)
Expand HTTP -> select Page Response Time (seconds)
4. Then, click the Show tab in the lower right-hand corner. You should see a graph that is similar to the following. Note that it may differ slightly due to node placement.
èNote that the graph panel has been labeled, as is necessary for any graph or graphs that you import to your lab report. The title contains your initials. To do this, simply right-click anywhere in the window and select Edit Attributes for the Panel Operations window.
ß 1. Copy this panel to your answer sheet. After the initial oscillations, what is the average database response time and what is the average HTTP page response time?
4. Close this graph by clicking on the red x in the upper right-hand corner. You will always get a message asking you if you are sure you want to delete it. Just click Delete because we can always bring the graph up again later.
5. Unselect the DB Response Time and the HTTP page response time. (Click the Unselect All button.)
6. Use the dropdown arrow to change average back to AS IS.
7. Expand the Object Statistics >Campus Network > Router A <-> IP Cloud > point-to-point > utilization <-
8. Select Show. You should see a graph similar to the one below. Note that this is in the AS IS view, not the average view. This view allows us to understand the link utilization behavior better.
ß 2. Copy this panel to your answer sheet. What does the panel tell us about the utilization on this link?
9. Close the panels and the Results Browser . You do not need to save any results at the moment, but do save your project.
Configuring Load Balancing with Redundancy
1. We now want to duplicate the scenario. For the new scenario, we want to implement a redundant DS1 link with load balancing. SelectScenarios>Duplicate Scenario .
2. Note that OPNET has automatically assigned a new name for our scenario that is an increment of 1 over the previous scenario. Backspace out the name OPNET has given to this scenario and input a name similar to Redundancy_Loadbalance.
3. Click ->OK
Configuring the Solution
1. While holding down the Shift key on your keyboard, left-click on the following 10BaseT links:
LAN4 – to – Switch
2. Tap the Delete key on your keyboard to delete the links.
The links should now be deleted.
3. Now open the Object PaletteSet Name .
(1) ethernet4_slip8_gtwy Name: Router B
(1) ethernet16_switch Name: Switch 2
4. Position Router B under Router A . Position Switch 2 under Switch .
5. From the Object Palette , obtain a 10BaseT link
LAN4 – to – Switch 2 6. Now from the Object Palette , obtain a PPP_DS1 link
Router B – to – IP Cloud
7. Close the Object Palette . Your Redundancy_LoadBalance scenario now should look as follows:
Configure OSPF
We now need to configure Router A and Router B with OSPF so that they will support load-balancing for same-cost-routes.
1. Left-click on Router A and then right-click and choose Select Similar Nodes . Both Routers A and B should be selected now, which is indicated by the black circles.
2. From the toolbar, select Protocols > IP > Routing > Configure Routing Protocols.
3. Configure the Routing Protocol Configurations in the order shown:
Check OSPF
Uncheck RIP
Make sure the Apply the above selection to subinterfaces is checked
Make sure the All interfaces (including loopback) is selected
Uncheck Visualize Routing Domains
Click OK . (Note: Do not bother to go back and look at your configuration because OPNET defaults back to the RIP view setting every time you view this window.)
3. While both routers are still selected, we need to activate the Load Balancing option. Select Protocols > IP > Routing > Configure Load Balancing Options .
Click OK . This will activate Load Balancing for the routers using OSPF with the same-cost routes.
4. Look in the lower left-hand corner of your screen; you should see the following:
All that we need to do now is add the results of one additio
