What errors does the data link layer detect?

The data link layer detects errors related to data corruption while a frame is transmitted over the physical medium.

Briefly and to the point: L2 checks whether the frame arrived without bit errors.

---

#Exactly which errors the data link layer detects

#1. Bit errors

The most fundamental type.

They occur when:

• the signal is distorted
• interference flips a 0 to a 1 or vice versa

The data link layer:

• recalculates the checksum (FCS)
• compares it with the received one
• if they do not match, it considers the frame corrupted

---

#2. Frame corruption

If:

• part of the frame is lost
• the data is altered
• the frame arrived incomplete

The checksum:

• will not match
• the frame will be discarded

---

#3. Loss of frame integrity

The data link layer can detect:

• errors in the payload
• errors in the frame header

Any change to the bits:

• results in an FCS error

---

#4. Errors related to collisions (classic Ethernet)

On networks with CSMA/CD:

• collisions corrupted frames
• such frames failed the FCS check

---

#Which errors the data link layer does NOT detect

Important for an interview:

The data link layer:

• does not understand logical errors (wrong address, wrong port, wrong protocol)
• does not check delivery order
• does not guarantee delivery
• does not correct errors (it only detects them)

Correction and retransmission are handled by:

• the transport layer (TCP)
• or logic above it

---

#How error detection actually happens

Ethernet uses:

• CRC (Cyclic Redundancy Check)
• the value is recorded in the FCS field

The receiver:

1. recalculates the CRC
2. compares it with the FCS
3. if they do not match: discards the frame