Entity Tag

What is an ETag (Entity Tag)?

An ETag is an HTTP response header that provides a validator – a token representing the state of a specific representation (usually a specific body bytes).

ETag is a hash identifier the server attaches to a resource so the client can check if the resource has changed without downloading it again.

Think of it as a fingerprint for the response entity.

• Header name: ETag
• Example values:
  • ETag: “33a64df551425fcc55e4d42a148795d9f25f89d4”

Purpose

• Help caches and clients determine whether a resource has changed.
• Enable conditional requests (If-None-Match, If-Match) to avoid sending full payloads and to implement concurrency controls.

Why ETag Exists

Imagine your browser already downloaded a 5 MB image yesterday. You revisit the page today.

Should the browser download the whole file again?

No.

“Has the file changed since I last downloaded it?”

If not changed -> server responds 304 Not Modified (no download).

If not -> server sends new file -> new ETag.

How It Works

1 Server sends a resource with an ETag

HTTP/1.1 200 OK
ETag: "abc123xyz"
Content-Type: image/png

The ETag here “abc123xyz” uniquely identifies this version of the file.

2 Client stores the ETag

Browser caches:

• The file
• The ETag “abc123xyz”

3 Later, client makes a conditional request

Client asks the server:

If-None-Match: "abc123xyz"

Meaning:

“Send the file only if it does not match this version."

4 Server compares the ETag

Case A: File unchanged

Server returns:

HTTP/1.1 304 Not Modified

Case B: File changed

Server returns:

HTTP/1.1 200 OK
ETag: "newetag456"

Browser downloads new version and updates cache.

Benefits of ETag

Strong vs Weak ETags

Strong ETag

Exact match of bytes.

ETag: "abc123"

Means:

Bit-for-bit identical

Weak ETag

Only semantically identical.

ETag: W/"abc123"

Content is effectively the same, but not necessarily byte-for-byte.