Moreover, certificate validation procedures will use only PKIX
checks when no DANE information is available. An active attacker who is able to divert
user traffic could block DANE traffic, so that he can bypassed these additional verification.
Moreover, there are situations where DANE information could fail to get to the End Entity due
to server errors or broken intermediaries that fi lter DNSSEC errors. Under these circumstances,
the End Entity performing the validation could assume an attack is undergoing and terminate
the connection, or it could dismiss the error and proceed. The latter would mean that blocking
DNSSEC traffic could help to bypass the DANE-defined procedures. Thus, in order for
DANE to be effectively used to prevent MITM attacks, a deployment of DNSSEC in clients,
servers, DNS infrastructure and intermediaries (i.e., to avoid DNSSEC information filtering) is
required. Taking into account the traditional resilience of network operators and manufacturers,
we cannot rely solely on DANE to provide the kind of path validation we are looking for in
this work. Finally, the verification of a key would require several DNSSEC queries that would
introduce an undesired latency, unaffordable in some cases, e.g., SIP, XMPP.