Amazon AWS Certified Advanced Networking Specialty – Networking & AWS Primer Part 3

7. IP Fragmentation

Hey everyone and welcome back to the KP Labs course. So in today’s lecture we will be discussing about the IP fragmentation. So this is one of the questions that might arise a lot of students that if an IP packet, let’s assume that you want to send a packet to a destination and you do not know the MTU. So if we what happens if the IEP packet is larger than the MTU of the interface? So the MTU of a router is 1500 bytes and you are trying to send the packet size of 1800 bytes. So what can happen? So, when the router receives the packet which is larger than the MTU, then there are two options. The first is it can fragment the packet into smaller PCs and send it forward. That is one fragment is dividing the packet into smaller chunks so that it can fit the empty you. And the second is it can discard the packet and send the ICMP packet to large message. So let’s understand both of these options. So you have the case one, you have the case two. Case one, where the router is good, it fragments the packet into smaller PCs and send it to the destination.

So you have the client, the client is sending a bigger packet which is larger than MTU. You have the router, router will take care of the fragmentation and sends it to the server. Now, you might think that this seems to be the best idea, but it is not. Packet fragmentation is not a good idea. Not only it brings the overhead to the router, but it will create a lot of issues in the reliability. So ideally fragmentation should not be done. That is the generic best rule which should be followed. Now, this is the case one case to discard the packet and send the ICMP packet to large message. So in the case two, when client tries to send the big packet to the router, a router will not fragment it because there is a DF bit set and it will send the ICMP packet to big.

So generally the difference is whenever a router receives a message which is larger than the MTU and if it does not have the DF bit set, DF basically means do not fragment, then the router will go ahead and fragment. However, if the IP has the specific bitcoin as DF which states do not fragment at all, then router will not fragment it and it will send the ICMP packet to bit message back to the client. So let’s look into what I mean by this. So I have a simple ping and let me try and send 1800 bytes of data to a Kplab in server. And if you’ll see it seems to be working perfectly. Now the question is why? The reason why it is working perfectly is because currently there is no do not fragment bit which is explicitly set. So router it does not see the Do Not Fragment bit and it will go ahead and fragment the 1800 bytes of data and send it to the server.

Now, in the second case, what we will do is we will explicitly define this Do Not Fragment bit. Let’s do it. So you can specify with hyphen f. So hyphen f will add the do not fragment bit. And now if you press Enter, you see it is saying that the packet needs to be fragmented, but the DF is set. So if you just want to see what exactly DF is all about. So if I just open up any packet within the IP protocol, within the flags, there is a flag called Don’t Fragment. And ideally in AWS many hosting providers, the dome fragment bit is set for all the packets. So Idly this fragmentation is not recommended. And this is the reason why various hosting provider, the packets by default have the Don’t Fragment bit set so that the client knows that it is actually trying to send the data packet which is larger than the MTU.

Now, one important thing to remember is that this is the ICMP based reply and if you block ICMP at your host level, it will lead to that you will not get any reply back. And that is also called as the ICMP sync host. So just avoid blindly blocking all the ICMP messages within your instance because many times you will actually miss certain important information there. So this is it. About the IP fragmentation in the a very high level overview and about the Do Not Fragment bit in the IP headers. This is it. I hope this has been informative for you and I look forward to seeing you in the next lecture.

8. Understanding Numeric Systems

Hey everyone, and welcome to the KP Labs course. So in today’s lecture, we will have a highlevel overview about the need of a numeric system. So let’s understand this with a simple use case. So since the start of the days, the days humans are born, we always have been looking into the ways to keep track of the count for the things, for our better understanding. So let’s understand this with a simple example. So let’s assume that you are an early human being and you want to keep track of the time it last reigned. So you want to keep track of number of days it has last rain. So the day it did not train, you write a symbol as I. So this is the I as the symbol. So the first day it did not rain.

So you put the symbol as I, second day it did not train, you type one more symbol and so on. So now you might say that why don’t I write number five? So number five is something which came later in the numeric systems. But this is like older times where numeric systems were not really introduced. So now if someone wants, like now someone asks you like how many days it did not train, so you will tell him this are the days it did not rain. So he’ll see, okay, these are the days it did not train. And he’ll try and analyze this particular scenario. Now, this is definitely a valid solution, but it cannot scale. So let’s assume that it did not rain for two years. So you will keep on writing and it might actually fill five or six pages. And the reader who wants to understand on how many days it did not rain, he’ll have to read through each and every page to understand the count.

So this is the older times. And in order to solve this use case, you had the numeric systems which were introduced. So two of the very popular ones are the base ten, also called as the decimal, and the base two, which is called as the binary. Now the base ten number system is something that most of us will be familiar with and it is also called as the decimal systems. So one of the easiest way in which you can understand the base ten is that there will be ten symbols. So let me give you one example. So this is one symbol. So you have another symbol, 3rd, 4th, 5th, 6th, 7th, 8th and 9th. So if you count from here, there are ten symbols which are used and this is the reason why it is called as the base ten system. Now, when you talk about the base two system, which is also called as the binary, since the base two, there are only two symbols, you have symbol one and symbol two.

So there are only two symbols which are used within the base two systems. So base two systems. Definitely. It powers the underlying computing system. So your calculator, your mobiles, your laptops, your computers, everything, they understand the binary system, which is the base two system. So there are only two symbols. You have zeros and ones. Now you can definitely convert because we are not that good with the binary ones, we are good with the base and also call it the decimal. So there can always be various conversion methods in which you can do so. The 192 number can be converted to binary and this is the converted equivalent. So you might write 192, but the computer underline will convert that to the binary for which it can understand. So this is what the numeric systems are all about in a very high level overview. So taking the example of the older times, this can be considered as the base one system because there is only one symbol which is used for counting.

9. Decimal to Binary Conversions

Hey everyone, and welcome back to the KP Labs course. So in today’s lecture, we look into the decimal to binary and how exactly the conversions would really look like. Now you might be wondering, is ZEEL alright? He’s teaching us numeric systems, is teaching us decimal to binary and all. So one thing I wanted to share is that these things are very important as far as the networking is concerned, concern, and you will find it to be very useful when the time comes. So let’s look into the decimal to binary and how exactly we can have a conversion. So let’s convert. So before we convert, we have certain things here. So you have two raised to zero, two raised to one, two raised to two, two raised to three, and so on. So this is two raised to seven and it can go on. Now what we do is we convert this to a equivalent decimal number. So two raised to zero is one.

So two raised to one will be two. Two raised to two will be four. So we convert this portion into the equivalent decimal number. Now let me quickly show you. So if you have two raised to two, so that would be you have to type two into two. So if you have two raised to four. So this part, two raised to four, that means type two, four times two into two into two into two. So two into 244288 into 216. So two raise to four would be 16 and so on. So let’s come back to the slide. So now we understood on how these decimal numbers arrived. Now you want to convert this into binary equivalent. So let’s look into how we can achieve that. So let’s assume we want to represent the number 50. This is a decimal number 50 into the equivalent binary. So the way in which you can achieve this is this. So this is the equivalent number 50 for the binary. So how you achieve this is for 50, you cannot use 128. So you put zero, you cannot use 64. So you put zero.

Here again, you can use 32. So I put one year, I can use 16, I can put one year. So 32 plus 16 is 48. So I just need two more. So then I search for two and this is where the two arises. So I put one year again. So you have 32 plus 16 plus two is equal to 50. So this is how the number 50 is represented in binary. So let’s take one more example of 95. So for 95 you cannot use 128, I am sure. So in the 128 field it will be zero. Can you use 64? Yes, you can use 64. So in the 64 field I’ll put one. So if you look into the answer, so 128, I have zero. 64, I have put it as one. Now 64 plus 32 would be 96. So 96 is not equal to 95. So in the 32 column I put it as zero. So 64 plus 16 plus eight plus four plus two plus one would represent as 95. And this is the reason why I have this specific task. So this is how you can actually convert the decimal into the equivalent binary representation.

So generally, when it comes to IP addresses, we already discussed that IP address we refer to as a decimal system. So 107, 231325. Now, computer will see it as a binary. So for each octave, so 172 again is converted into the binary equivalent with the same table. So you can try it out, you can convert 172 into binary, you can convert 32 into binary, 32 into binary and 50 into binary. So you will get this exact same output. So computer will see it as this. However, human being will see this as the IP address. So it becomes easier for us to understand as well as write. So during the subnetting aspect, the knowledge of decimal to binary conversion is extremely important and it will really help you understand things better. And this is the reason why we are actually discussing the entire conversion.

• Category: Uncategorized