Error checking in Data Link or Transport ?

walid97

Hey,

I am confused here, I read the technotes:

Data link: "requests retransmission of frames in case an error occurs. Provides error checking by adding a CRC to the frame"

Transport: "Provides services such as error checking and flow control"

and I read in another post that error checking happens in the Network layer too.

Can you guys explain further about error checking please?
thanks !

bwolfe

Well TCP, which is the transport layer protocol half of TCP/IP, is 'connection oriented'. Simply put TCP cares about which packets have made it safely to their destination and which ones need to be retransmitted. UDP, a connectionless transport protocol, on the other hand doesn't give a hoot about what gets where.

So that's one kind of error checking...

Then there must also be some data integrity check involved.

As for the comment re: the datalink layer providing error checking I don't know what was intended but perhaps the datalink layer plays a part in the process by providing a section of the frame that contains a checksum for data integrity while TCP is more concerned with lost/late packets. That's just a guess on my part. I'm still learning myself so I look forward to seeing some further responses.

Drakonblayde

They both provide error checking depending on what type of error you're looking for.... TCP makes sure the packets got there, and if anything got dropped, retransmits it. Data Link is just concerned with frame integrity, it ensures that nothing got corrupted in transit.

Think of TCP's error checking like sending certified mail and the data link error checking like the mail carrier making sure the box hasn't been opened before he hands it over to the owner