UCEN 3133 Advanced Computer Networks - Tutorial 5 - Year 2012

Tutorial 5 – Routing and Traffic Engineering

1. Why is conventional IP routing mechanisms such as RIP incapable of supporting constraint based routing (CBR) ? Based on your answer:

     (i)            Explain why CBR routing is supported by a variant of link state routing protocols such as OSPF

   (ii)            Give two reasons why a MPLS network is particularly suited to support CBR

Constraint-based routing requires route calculation at the source as different sources may have different constraints for a path to the same destination, and the constraints associated with a particular source router are known only to that router, but not to any other router in a network. In IP routing such as RIP, every router in a network is involved in computation of a route is in a distributed fashion.

i) Link-State Routing Protocols floods the network with information about all links in the network (in addition to constraint-related information associated with that link), hence allowing the source router to perform the entire route calculation

ii) The LSRs within the core MPLS network only switch on the basis of these labels and no longer make any forwarding decisions of their own based on the IP header (which is what happens in conventional routing).

Mapping between an FEC and an LSP is completely confined to the edge LSR at the head end of the LSP. The decision as to which packets will take a particular explicit route is completely achieved by this edge LSR, noone else is involved.

2. What is the common additional feature used in extending RSVP and LDP to support CBR, and briefly explain how this feature is used as part of CR-LDP ?

The Explicit Route Object (ERO) contains the explicit route that the message has to take. Forwarding of a message containing a ERO by a router is determined not by the IP destination address, but the content of the ERO. The ERO consists of an ordered sequence of “hops,” where the sequence specifies an explicit route and each hop is represented by an “abstract node.”, which is a group of one or more routers

1. A LSR determines the explicit route to be established and constructs a ERO that contains this route.
2. This LSR then constructs a LDP LABEL REQUEST message and includes ERO in this message, and finds the first abstract node in ERO and forwards this message onwards to it.
3. The receiving LSR removes the abstract node associated with it and forwards it onwards to the next  LSR specified in the ERO
4. This process is repeated until it finally arrives at the destination LSR.
5. This LSR then constructs a LABEL MAPPING message and sends it back on the same route – each LSR along the way will use the label to populate its label forwarding message.
6. When it reaches the original LSR, the labels would have established a complete LSP from the original LSR to the destination.

3. List 5 common traffic parameters that are part of the CR-LDP specification and briefly explain their purpose.

• Peak data rate (PDR), Peak burst size (PBS)
• Committed data rate (CDR), Committed burst size (CBS)
• Excess burst size (EBS)

Peak data rate and burst size together define a token bucket, which characterizes the maximum rate of traffic that is expected to be sent down this LSP. Committed data rate and burst size define a token bucket characterizing the average rate at which traffic is expected to be sent on this LSP. Excess burst size defines another token bucket that can be used to characterize the amount by which bursts may exceed the committed burst size.

4. Resource reservation for QOS purposes using RSVP may cause problems in a MPLS network that uses link state routing protocols such as CSPF. Explain briefly the cause of this problem, and how it is addressed.

Establishing a route for a particular traffic flow may require resource reservation along the route using RSVP. Once resources (such as bandwidth) of a link are reserved, the attributes of these links will change (since bandwidth is usually one of the attributes). In link state routing protocols such as CSPF, when the attribute of a given link changes, the node to which the link is connected floods this information throughout the network. Since resource reservation happens quite frequently, flooding also happens frequently which results in high network overhead.

This can be addressed by establishing an upper bound on the frequency of flooding of information when a link attribute changes.

5. List the drawbacks of using RSVP and CR-LDP respectively in a MPLS network.

RSVP makes reservations for individual microflows between single applications, which does not scale well as  the number of microflows in a large IP network is very large. RSVP’s soft state also requires constant refreshing which consumes bandwidth and processing resources.

CR-LDP, by contrast, runs on top of TCP. TCP’s congestion avoidance may limit the transfer of information between LSRs. There is overhead involved in establishing an adjacency between two LSRs because they must go through TCP’s handshake sequence before initiating an LDP session.

6. What is meant by a routing transient and name 2 factors that influence the duration of a routing transient within a network based on conventional IP routing

Routing transients refer to episodes in a network where routing information across a network is changing, mainly due to failures of links or routers or both. At such times, the routing information stored at different routers may be temporarily inconsistent.

The duration of routing transients depends on two factors: - The time it takes for a router adjacent to the failed link (or node) to detect the failure, and the time it takes to distribute this information among all the routers and for them to recompute their forwarding tables based on this information (converge on the failure).

7. What is the motivation behind the use of a protection LSP in a MPLS ? Briefly explain how it is set up and used in that context.

Protection LSPs are used to handle link failures in an MPLS, so that when a link fails, the LSR attached to that link can channel all traffic destined for that link on to the alternative protection LSP.

When a link between two LSRs fail, the information about this failure will be distributed (via OSPF or IS-IS) to all the LSR. Once the original LSR gets this information, it can use constraint-based routing to compute a new route (the protection LSR). Or it may already have a precomputed alternative route

To route traffic onto the new route, the LSR pushes a new label corresponding to the protection LSP into the label stack of all incoming packets, allowing them to be switched over the protection link.

8. The Constrained Shortest Path First (CSPF) algorithm is typically employed in a MPLS network for TE purposes during the construction of routing tables. In the process of constructing a routing table based on CSPF, what are the tiebreaking factors used to arbitrate between two paths of equal hop cost ?                    

For tie break

1. Take the path with the largest minimum available bandwidth.
2. If there is still a tie, take the path with the lowest hop count (the number of routers in the path).
3. If there is still a tie, take one path at random.

9. Consider a MPLS network of routers (A-E) as shown below. The parenthesis pair that labels each link between the routers is used to denote the hop cost and bandwidth available, respectively. For example, the link between A and B has a hop cost of 2 and a bandwidth of 90 Mbps. CSPF is now used to calculate the best path to router D from router A, given a constraining bandwidth of 70 Mbps. Show all steps involved in constructing router A’s table to determine this best path. In your working, show all tentative routes possible and mark them as cancelled if they do not qualify to be used.

Configured constraint 70 Mbps

PATH list	TENT List
{A, 0, self, N/A}

PATH list	TENT List
{A, 0, self, N/A}	{B, 2, B, 90} {C, 8, C, 80} {D, 10, D, 100}

PATH list	TENT List
{A, 0, self, N/A} {B, 2, B, 90}	{E, 3, B, 90} {C, 7, B, 90}  [1] {C, 8, C, 80} -> cancel {D, 10, D, 100} {D, 10, B, 80} -> cancel

PATH list	TENT List
{A, 0, self, N/A} {B, 2, B, 90} {E, 3, B, 90}	{C, 7, B, 90} {D, 10, D, 100} {D, 10, B, 90} -> cancel   [2]

PATH list	TENT List
{A, 0, self, N/A} {B, 2, B, 90} {E, 3, B, 90} {C, 7, B, 90}	{D, 10, D, 100} {D, 8, B, 50} -> cancel    [3]

PATH list	TENT List
{A, 0, self, N/A} {B, 2, B, 90} {E, 3, B, 90} {C, 7, B, 90} {D, 10, D, 100}

[1] {C, 7, B, 90} – The required format is {destination, cost, next hop, minimum bandwidth}. If we take the newly added node to the PATH list {B, 2, B, 90}, this means that the destination is B, cost = 2, next hop to get to B is also B, and minimum bandwidth on all the routes to B is 90. Now when we consider connection to C from B, the entry for the TENT list becomes {C, 7, B, 90}. 90 because if we consider from A->B->C, the minimum bandwidth on all the links encountered is 90. The logic of taking the minimum bandwidth is that the delay along any given path is most affected by the link on that path with the smallest bandwidth. So if a packet was going from A -> C, the slowest  travel time would be from A to B. Therefore, if we have two paths to C with exactly the same cost, then as a tie breaker we would pick the path with the highest bandwidth (remember the rule).

[2] {E, 3, B, 90} -> This means the shortest path to E from A has a total cost of 3, with next hop B, and minimum bandwidth among all the links so far is 90. E has a connection with metric {7,100} to D. So the total cost from A to D now becomes 10. The next hop is still B. The minimum bandwidth is still 90 (because the E-D bandwidth is 100). So, the correct entry becomes {D, 10, B, 90}, which is cancelled because its bandwidth is lower than the existing entry of {D, 10, D, 100} -> remember, in tie breaker we choose the one with the highest bandwidth

[3] {C, 7, B, 90} -> This means the shortest path to C from A has a total cost of 7, with next hop B, and minimum bandwidth among all the links so far is 90. C  has a connection with metric {1,50} to D. So the total cost from A to D now becomes 8. The next hop is still B. The minimum bandwidth becomes 50 (because this is lower than the previous lowest of 90). So, the correct entry becomes {D, 8, B, 50}, which is cancelled because the bandwidth of 50 is less than the initial configured constraint of 70 as given in the question.

10. A multimedia network that provides QOS guarantees uses a leaky bucket policer in one of its routers to ensure that the incoming packet traffic does not exceed the TSpec specification agreed upon during an initial session of Integrated Services (IntServ). The following are the features of this policer:

·         The token buffer can hold at most three (3) tokens, and is initially filled with two (2) tokens at time slot t = 0.

·         New tokens arrive into the bucket at a rate of two (2) tokens per time slot. Packets arrive at the beginning of a time slot and enter the packet queue, where they are processed and transferred to the output link in a First In First Out (FIFO) manner.

·         The size of the packet queue is four  (i.e. it can queue a maximum of 4 packets at any given time slot); any extra arriving packets are dropped.

·         Packets that obtain available tokens in a given time slot go together on the same time slot in the output link.             

Time slot	Incoming Packets
0	A B C D
1	E F
2	G
3	-
4	-
5	H I J K
6	L M N O
7	P Q
8	-
9	R S T

The table shows incoming packets from the network into the router with the policer, from time slot t = 0 to time slot t = 9. Based on this information, construct a new table with columns showing the packets in queue, tokens in bucket and packets on output link from time slot t = 0 to t = 9.                                            

Time slot	Packets in queue	Tokens in bucket	Packets at output
0	A B C D	2	A B
1	C D E F	2	C D
2	E F G	2	E F
3	G	2	G
4	-	3	-
5	H I J K	3	H I J
6	K L M N	2	K L
7	M N P Q	2	M N
8	P Q	2	P Q
9	R S T	2	R S

UCEN 3133 Advanced Computer Networks - Tutorial 3 - Year 2012

Tutorial 3 – MPLS Operation

1. A smaller autonomous system (AS) is defined inside the boundaries of a larger AS, and both of them employ MPLS to forward packets. A label switched path (LSP) has already been established from end-to-end between edge Label Switched Routers (LSRs) from the larger AS. This LSP tunnels through the smaller AS. In this scenario, what mechanism does MPLS provide in order to minimize routing information maintenance in the internal LSRs of the smaller AS ? Describe in detail how this mechanism works as a labelled packet is forwarded through the smaller AS.

MPLS uses a stack of labels.

1. When a transit packet arrives with its own label corresponding to an external LSP, the ingress LSR of the smaller AS pushes a new label onto the label stack.
2. This new label corresponds to an internal LSP within the smaller AS
3. The internal LSRs of the smaller AS switch the packet through to the egress LSR, working only on the topmost label (the new label)
4. At the egress LSR of the smaller AS, the new label is popped off the stack restoring original label.
5. The packet resumes its travel using the original external LSP for the larger AS.

2. What is the motivation behind the use of Penultimate Hop Popping in an MPLS network ? Describe briefly how it works.

It provides a form of optimization to minimize the performance penalty incurred when the egress LSR of a MPLS network has to perform two lookup operations when processing a label stack. When a packet with a stack of two labels arrives at the penultimate LSR (the LSR before the egress LSR), it pops the top stack before forwarding it onwards to the last stack.

3. Distinguish between the independent and ordered control approach to exchanging labels in the Label Distribution Protocol (LDP).

Independent control - Each LSR advertises label mappings at will. In an unsolicited mode, LSRs will generally advertise new labels whenever the IP routing information changes.

Ordered control - In this mode,  all path advertisements originate at the egress router, and the label advertisement messages travel upstream to the ingress router. For advertisement on demand, a request from the ingress router triggers a chain of requests downstream to the egress router

4. Explain, with the aid of a diagram, how the liberal approach to label retention works in the LDP ? What advantage does the liberal approach offer over the conservative approach, and what is its corresponding disadvantage ?

It retains all label mappings even if they don’t correspond to an existing LSP.

• Advantage: If the next hop path changes for the LSP, the given LSR  may already have a new LSP ready, so it can react to the routing change almost immediately. Whereas in conservative approach, the LSR will need to request label mappings first from alternative downstream LSRs in the neighbourhood.
• Disadvantage: Larger mapping table space required to keep all the labels, as well as network overhead in label exchange when the number of neighbouring LSRs is very large.

5. Explain two approaches that could be employed to address the potential problem of loops in the label distribution protocol (LDP)

·         The hop count field in the label request message is incremented by all LSRs that the message passes through. If the hop count ever reaches a maximum value,  a loop has occurred and the LSR sends a LDP notification message to the originator of the message.

• Path vectors - Each router inserts its own LSR ID in the path vector field of the message, and if it ever finds its own ID here in a newly received message, it breaks the loop

6. Name the two situations in which a LSR in the middle of a LSP may need to return an Internet Control Message Protocol (ICMP) error message to a datagram’s source. What is the problem with using ICMP in a Virtual Private Network (VPN) deployed using MPLS, and how is this addressed by the LSR that wishes to create the ICMP message ?

Two situations:

• The datagrams hop limit may run out
• The packet may encounter a problem in a link between routers (link broken, or datagram too large to be fragmented and sent over link)

In a VPN, only the egress router at the boundary of the VPN is able to access the source IP address. Other intermediate routers cannot access the source IP address (which is encrypted using IPSec), and hence cannot route the appropriate ICMP message back to the source.

This is addressed by creating a ICMP message and adding the original stack label from the packet that caused the error. This ICMP datagram will follow the route of the original packet to the egress router, who can then successfully route it back to its source.

7. How might a session initialization deadlock occur during the LDP of a MPLS ? Explain how this can be overcome using the LDP hello message.

As part of LDP, two LSRs may not be able to agree on parameters during session initialization. The propose / reject cycle could then continue indefinitely, causing a deadlock. This can be overcome by having the network administrator intervene and make the passive LSR accept the proposed session. To make the active LSR propose again, the passive LSR increments the configuration sequence number in its hello messages to the active LSR. 

Subnet Calculator for CCNA Exam

Online Subnet Calculator

The IP Subnet Mask Calculator enables subnet network calculations using network class, IP address, subnet mask, subnet bits, mask bits, maximum required IP subnets and maximum required hosts per subnet.

http://www.subnet-calculator.com/

http://www.subnetmask.info/

http://www.warriorsofthe.net/utils/index.html

http://www.vlsm-calc.net/

Using Proxy To Surf Internet in Firefox

Why need proxy?

1) To hide your online privacy and anonymous web surfing.

2) To bypass some of the company or school network rules (e.g. block facebook site. etc)

3) To view some website that restrict geographical area access (e.g. some sites from other countries are not viewable in Malaysia using Malaysia ip)

4) To access some site with smooth view experience without lagging.

5) To spoof your identity on the internet. (e.g. in a forum, spoof that you are not a malaysian using ip)

So, how to use proxy in order to archive above functions?

First of all, you will need a browser. In this tutorial, I will recommeded using Mozila Firefox.

1) Get a free proxy website that provide free proxy server.
e.g. http://proxy-list.org/en/
You can get more if go google search for free proxy server list.

input the confirmation code to get a list of free proxy servers.

Then, you will get a list of proxy server with ip and port provided

Please note that, there are total of 3 types of proxy server: elite, transp, and anony.

Transparent Proxy – IP address can be caught by sever using HTTP_X_FORWARDED_FOR variable, such proxies usually used to increase internet speed and decrease amount of abroad traffic (some internet providers grant their user free local traffic).
Anonymous Proxy – IP address can not be caught by server but server knows that proxy was used during connection. It is very useful to be anonymous surfing in internet .
Elite Proxy – IP address as well as fact that proxy was used during connection can not be caught by server. Best way to hide any information about you presence in internet.

 copy the ip and port of proxy server to your firefox network setting

Configure the firefox network setting: Tools > Options > Advanced > Network > Settings

Please note that don't forget to check the option "Use this proxy server for all protocols".


Well done, you are now can surf the net using proxy server!!
Please leave comment if you need some help, we love feedback!

Some advice when using proxy:
1) Do not log in any account that is very important to you when you are using proxy, since all connection will be pass through the server, that is no guarantee of confidentiality!
2) Don't ever think that no one can trace you when you are using proxy, although it is hard but it is possible too! So, please do not use proxy to do bad things.
3) Some proxy may have some restriction (i.e. require to enter confirmation code while surfing to a new site), so choose your proxy wisely.

-bluefiz- 

How to Auto Start Thunder at Any Time 自动启动迅雷在任何时刻

maybe you are a collague who is stayed with a group of people or blah blah blah
and the most suck thing you are sharing with  most the streamyx or blah blah blah broadband.
So, they have their rule to ensure they can surf the net without any lag.
Their rule: Only can download from 12:00AM to 6:00AM..

You said this is a great rule, it can ensure the line at home smooth before 12:00AM
you can still googling or blah blah blah before 12:00AM
but...
you want download something but it can only start from 12:00AM, are you gonna to wait at there til 12:00 AM and finally can open the thunder or other download application? Ow man, it definitely wasting my time!!!

As a result, I will show this "auto way" to all of you. If you using my method, you can go sleep at 8:00pm also but the thunder application will automatic start and the download progress will continue on anytime. Sound great, right?Let's start our tutorial :)

ll let your download progress continue when your thunder application is start.

Open your command prompt(Start->Run->cmd). Enter the command above. Dun forget to get your own thunder application path! The time also can be change as you want.

After execute the command(press enter), you will get the red border message which show you are successfully added one job.

Well.. you can now go to sleep and wait your thunder start automatically. ^o^

How to automatic shutdown your computer??

the red border show the time you want before shutdown. For instance, 3600 mean after one hour your computer will auto shutdown. the time unit is in second.

*all above method are tested in window xp only. I am not sure other OS is compatible.

You are welcome to ask me question ^o^

Disable USB Autorun to Save PC from USB Viruses

First Way:

• Browse to the following key HKEY_CURRENT_USER\Software\Microsoft\Windows\ CurrentVersion\Policies\Explorer
  
• Modify the value of NoDriveTypeAutoRun to df (hexadecimal)

Second Way:

1.START ->RUN ->Key in 'regedit' to open registry.

2. HKEY_CURRENT_USER\ Software\Microsoft\ Windows\CurrentVersion \Explorer\MountPoints2,

3.Right click 'mountpoints2' and select 'permission'

4.Then click 'Advance',uncheck 'inherit from parent the permission entires that apply to child objects.Include these with entires explicity defined here'.

5.Click 'remove'.'Yes' and 'ok'.

Turn ON/OFF your Remote Desktop

Here is the batch :

@ECHO OFF
:Prompt
::Prompting technician for machine name/IP which to connect
SET /P TargetMachine=Please enter the machine name or IP address:
CLS
ECHO Remote Desktop Check on \\%TargetMachine%
:Menu
ECHO.
ECHO 1. Remote Desktop Check (Enabled 0x0 or disabled 0x1)
ECHO 2. Remote Desktop On (0x0)
ECHO 3. Remote Desktop Off (0x1)

set choice=
set /p choice=Please type in your choice and press enter.
if not '%choice%'==' set choice=%choice:~0,1%
if '%choice%'=='1' goto RemoteDesktopCheck
if '%choice%'=='2' goto RemoteDesktopOn
if '%choice%'=='3' goto RemoteDesktopOff
ECHO "%choice%" is not valid please try again
ECHO.
goto Menu

::Listing 3: CheckRD.cmd
:RemoteDesktopCheck
@reg query "\\%TargetMachine%\HKLM\SYSTEM\CurrentControlSet\Control\Terminal Server" /v fDenyTSConnections
GOTO Prompt

::Listing 4: RDOn.cmd
:RemoteDesktopOn
@reg add "\\%TargetMachine%\HKLM\SYSTEM\CurrentControlSet\Control\Terminal Server" /v fDenyTSConnections /t REG_DWORD /d 0 /f
GOTO Prompt

::Listing 5: RDOff.cmd
:RemoteDesktopOff
@reg add "\\%TargetMachine%\HKLM\SYSTEM\CurrentControlSet\Control\Terminal Server" /v fDenyTSConnections /t REG_DWORD /d 1 /f
GOTO Prompt

Copy and save it the name as you like.

This batch file can use to check for your remote desktop status either on or off. On the other hand, you can also use it to turn on or off of your remote desktop.

For more details of Remote Desktop : Please visit here .

Access Command Prompt from UTAR

There are many method to unlock the command prompt in UTAR
Below are those method that found out by the people :

Method 1:
Create a batch file. First open the notepad. After that..
---------------(copy after this line)
command
---------------(copy end)
Save the word above into xxx.bat and run it then it will pop up a shell which is cmd.


Method 2:
Using gpedit.msc. gpedit.msc can be found in the %systemroot% normally is 'winnt' or 'windows' folder. When you click on it, choose the User Configuration -> Administrative Template -> System.



Then you will find 'Prevent access to the command prompt', click it.

Choose 'Disabled'
After that, you can start your command prompt in your program accessories.


Method 3:
Modify the registry key of the command prompt in the regedit.

If any of UTARIAN found any method, you are welcome to share your experince and method with us to bncteam4u@gmail.com

Regards,

Bluefiz

Chapter 1 - Winrar EXE

Hi to all :)

Today find something interesting to blog !
For those who are interested , hope that really helpful to them :P

Sometime, when you download some files from the internet
After completed, you go to click the file
But unfortunately, it did not give you any response.
Then we can conclude that : Congratulations, your computer might be infected by virus and Trojan . :)

Mostly those file which can spread Trojan or virus is exe file.
Exe file is kind of executable file , it can use many kind of method to bulid it .
One of the easiest way is using the WINRAR.

Winrar, I assume that most of the people know what it is. =.=
Let's start !



First , you must have winrar install on your computer.
Then, point to the file that you wish to compile into exe file , you can either select only one or more than one.

---->Press the mouse right button and select 'add to archive'.


After that, a window will pop up


---->Then check the 'Create SFX archive' box, you can see the .rar become .exe



---->Select 'SFX options'



---->There are three way to extract the exe file
------->Create in program files ( This one everybody should know)
------->Create in current folder ( This one means it extract the file in where your exe file locate)
------->Absolute path (This option let you can choose where you wan to extract the exe file)
----------->eg: %systemroot%, c:\windows, %userprofile%\Desktop ... etc.


---->Setup Program( Let you choose wat to run when extract the exe file)
------->Run after extraction, eg:trojan.bat
------->Run before extraction eg:girl.jpg
Please note that the file that you wish to run must be include in the archive.

#This is the part how the trojan affect your system! When you click it, nothing happen but what really happen is the trojan had inject something into your system. So, this part are important if you wanna made a trojan. :)



---->You can also delete the file after execution of the exe file.
------->eg: your file include before compile to exe :girl.jpg, start.bat, source.bat. after the extraction you wish to delete source.bat to avoid people see your source code, so this delete file option will help you



---->The temporary mode can ignore, less use
---->Slient mode
------->Hide all: this mean you run the exe file without anything pop-up just like wat i already said on above.



---->This is the coolest part.
---->You can change the icon of your exe file. That mean you can really made the file like rar file then other people will treat it as rar file, then they will go to click. Conclusion: If you really wan made a good torjan to bluff people to click accidentally, find a icon that is normally can find in windows default, such as :winrar icon, zip icon, folder icon, txt icon...etc.




---->This part I also less use, unless you really wan to create a software program if not ignore it.


---->This part not so important , ignore it.:)



So, we had done !
Do you success made your exe executable file from winrar ?
Any question dont hesitate to ask !
I am willing to help :)