Question

Is it possible that a packet can be transferred in Internet forever? If yes, please give an example; If not, please explain why a packet cannot remain “valid” in Internet forever?

Answer 1

Solution:-

The direct answer to your question is NO as it is not possible that a packet can be transferred on the internet forever.

Let us understand with a little example. I would rather like to do it with the main reasons why packets get lost and parallels understand why the above statement is not possible.

Reason 1 ( Physical Layer issue): As cable, wireless networks have the signal problems, it is common that the packet is lost somewhere in between the transit. As noise and poor cabling could be the reason for the packet not delivered. In this case we can see that due to the physical interference the packet gets lost. Now, let us see a real-time scenario where you with one of your friends is trying to share some data over the wireless medium. While the transfer was in progress your phone restarts or get switched off due to various possible reasons. If you try to find an option to resume the data from the previous transfer status, that would not be possible. Now please understand that while your friend’s phone had sent one packet successfully and before reaching your phone due to restart, where does that packet go. It simply gets lost due to the above-said reason. It could not be traced back.

Reason 2 ( Congestion): Here there must be one node present which is full in its potential and thus discards every other incoming packet as it cannot handle more. So it has two options left with it. Either to throw away (right option) or to simply collapse (wrong option). So what is been done is not been known to the sender and it is waiting for the acknowledgment or fixed time. It transmits again by wondering that the packet has been lost somewhere in transit, in this scenario we see that the packet was being discarded by the receiver and that was not known to the sender. You as having the question where does that packet goes, the answer will be it will simply drop. To overcome these sorts of scenarios where the overall throughput gets down, TCP/IP was used. Today even the discarded data packets information are been shared with the sender automatically or slow down the transfer speed.

Reason 3 (Human Error): A routing loop error is one that causes any particular router on the path to the receiver, doesn't have a clue to what to be done with the packet it just received. In this scenario as to create space the router will simply delete the packet. This cannot be fixed by itself and thus need human work. With one router could be fixed easily but with multiple routers, eventually, slow in rectification will slow down the network and lowers the throughput value.

So with the above explanation we can say that each packet has sender information, route it would take, and destination information within it to be successfully transmitted to the receiver. Within the transit many packets can get lost, damaged, and eventually dropped in some cases. While the same is been lost, damaged, or dropped. We can conclude that if it is lost, there is no path known to the packet to trace the route and get back even if we know that packets don't have the mind to do so. If it is damaged, then maybe the destination information is lost and eventually, it will not land anywhere and will get lost within and if it is dropped by the destination then, as packet reached to the destination it was mean to and was rejected from the receiver it will round for some time and will eventually be lost and with zero chances of recovery. TCP/IP still gives a chance to get the packet retransmitted over as without receiving acknowledgment or limited waiting time but with UDP, simply forget the packet.

We can also say that we only concern about the transmission and successfully receipt by the receiver of the message rather than find out where the packet gets lost. We resend the data again and again till it delivers successfully with positive acknowledgment or till a valid number of attempts after which we only stop sending too in some scenario.