RFC 5996 compatible implementations MAY send an INFORMATIONAL message
with an AUTHENTICATION_FAILED if the initiator failed to authenticate us.
Handle such a message like a DELETE for an IKE_SA.
According to RFC 5996, we MAY send an INFORMATIONAL message having an
AUTHENTICATION_FAILED. We don't do any retransmits, though, but just close
the IKE_SA after one message has been sent, avoiding the danger that an
unauthenticated IKE_SA stays alive.
While this was problematic in earlier releases, it seems that it works just
fine the way we handle compression now. So there is no need to disable it over
NATed connections or when using forceencaps.
The previous code did not properly check for the situation when the
DELETE for a redundant CHILD_SA created by a responder during a
CHILD_SA rekey collision arrives before the responder's answer to the
initiator's winning CREATE_CHILD_SA request.
Fixes the use of EAP methods in the non-first authentication round if the
initiator demands mutual EAP. Also mutual EAP can now be enforced when the
initiator sets rightauth=eap, not only with rightauth=any.
Fixes retransmit of the last Main Mode or IKE_AUTH message, and correctly
queues Main Mode messages when processing of the last message is still in
progress.
When the last request message of the initial tunnel setup is retransmitted,
we must retransmit the response instead of ignoring the request.
Fixes#295.
Setting the responder SPI to 0 can only be done while generating the
response, otherwise we'd fail to check in the IKE_SA again in case the
hash table is enabled. That's because we use the responder SPI as hash
value since 5.0.0.
If it is set on an auth config IDr will not be sent, and later the configured
identity will not only be checked against the returned IDr, but also
against other identities contained in the responder's certificate.
If the provider has access to the full pool list, it can enumerate
them twice, for example to search for existing leases first, and
only search for new leases in a second step.
Fixes lease enumeration in attr-sql using multiple pools.
With uniqueids=no the daemon still deletes any existing IKE_SA with the
same peer if an INITIAL_CONTACT notify is received. With this new option
it also ignores these notifies.
This allows listeners to migrate to the new IKE_SA with the
ike_reestablish event without having to worry about an ike_updown event
for the old IKE_SA.
Martin Willi