5

Is there any other difference except the session cache?

Will only Session ID based Session Resumption update symmetric TLS session key ?

4
  • 1
    I recommend the "Session Resumption" section of the "High Performance Browser Networking" book by Ilya Grigorik. Jan 3, 2017 at 8:16
  • Thanks @StackzOfZtuff, I want to know about, Will only session ID based session resumption update symmetric TLS session key ?, "High Performance Browser Networking" book didn't talk about that. Jan 3, 2017 at 8:21
  • I think NEITHER of the schemes will do that. (But I don't have a good source.) What makes you think otherwise? Jan 3, 2017 at 8:56
  • Have you sample code in which session ID based Session resumption is used ? I want deep understanding of the both methods...? according to memory constrain as well as security parameters. @StackzOfZtuff Jan 3, 2017 at 9:13

1 Answer 1

6

Update: Warning: my understanding of this is botchy. See Dave Thompson’s comment.

——————————————

Will only Session ID based Session Resumption update symmetric TLS session key ?

Neither method will do that.

To quote Adam Langley: (line breaks mine)

TLS offers two session resumption mechanisms: session IDs (where the server and client each store their own secret state) and session tickets (where the client stores the server's state, encrypted by the server).

If an attacker can obtain the session resumption information for a connection then they can decrypt the connection.

(This needn't be completely true, but it is for TLS because of the way that TLS is designed.)

Source: How to botch TLS forward secrecy (27 Jun 2013) (Archived here.)

Further reading

8
  • Thanks. But, still i have lots of question regarding to TLS Session Resumption. i will go through it first... Thanks for support. @StackzOfZtuff Jan 3, 2017 at 12:41
  • 1
    @DevangKubavat: See also section 2.2 in this PDF: dl.acm.org/citation.cfm?id=2987480 Jan 4, 2017 at 13:50
  • Yes, @StackzOfZtuff. After session resume, It is not updated Session key.. Same Previous session key is used in both the methods as per the paper. Jan 5, 2017 at 6:33
  • security.stackexchange.com/questions/147322/…. posted new question.Can u give me answer if you have an idea.. @StackzOfZtuff Jan 5, 2017 at 9:45
  • 2
    To be exact, the master secret is reused; the actual keys, plural (at least 2, often 4) are re-derived with new nonces, which does help with things like sweet32 and RC4 bias (or did before 3DES and RC4 were obsoleted), but (as you describe) does not provide forward secrecy. Apr 5, 2018 at 5:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .