[Swan-commit] Changes to ref refs/heads/master

Andrew Cagney cagney at vault.libreswan.fi
Sun Dec 23 01:33:48 UTC 2018


New commits:
commit 1de260eb41ef20066c05e3938cb9e89db07b2c12
Merge: 3a9c4ef 9cc5b65
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Dec 22 20:28:40 2018 -0500

    demux: respond to messages from the future using IKEv2
    
    Merge commit '9cc5b653e1ec214cb282d5025440cbaf5e8ba4d5'

commit 9cc5b653e1ec214cb282d5025440cbaf5e8ba4d5
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Dec 22 10:02:33 2018 -0500

    ikev2: respond to an IKEv++ message with v2N_INVALID_MAJOR_VERSION
    
    Not IKEv1.  Also ignore version=0.  Add notes on how dodgy all
    this is.  For instance, have to assume that MD is an initial
    request.

commit a82b6f4aed542c4780d2fd9c5333f01e17495372
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Dec 22 20:25:47 2018 -0500

    ikev2: in send_v2N_response_from_md() don't pexpect() inconsistencies
    
    Code needs to handle an MD from the future - IKEv++.

commit e146eaf73689c39e4fe2a32033a6e2708e1aa7d3
Author: Andrew Cagney <cagney at gnu.org>
Date:   Sat Dec 22 09:59:02 2018 -0500

    ikev2: in open_v2_message() don't call v2_msg_role() to sanity check it is a response
    
    v2_msg_role() expects an IKEv2 message which isn't true when
    responding to IKEv++.  Instead just trust MD!=NULL and assume
    caller knows best.



More information about the Swan-commit mailing list