RFR: 8366454: TLS1.3 server fails with bad_record_mac when receiving encrypted records with empty body

[Daniel Jeliński]

On Mon, 22 Sep 2025 20:20:19 GMT, Alice Pellegrini <duke at openjdk.org> wrote:

> According to RFC 8446 section 5.4, third paragraph 
>> Application Data records may contain a zero-length
>>    TLSInnerPlaintext.content if the sender desires.  This permits
>>    generation of plausibly sized cover traffic in contexts where the
>>    presence or absence of activity may be sensitive.  Implementations
>>    MUST NOT send Handshake and Alert records that have a zero-length
>>    TLSInnerPlaintext.content; if such a message is received, the
>>    receiving implementation MUST terminate the connection with an
>>    "unexpected_message" alert.
> 
> 
> The proposed change removes an off by 1 error in the SSLCipher implementation, forces the correct Alert message to be sent in response to zero-length Alert fragments, as well as updating some tests which detected the BadPaddingException but now detect a SSLProtocolException, which is thrown by `TransportContext.fatal`

Looks reasonable. Would it be possible to add a test for the empty plaintext case? Or do we never generate empty plaintexts?

Please update the copyright years.

src/java.base/share/classes/sun/security/ssl/SSLCipher.java line 1926:

> 1924:                 // remove inner plaintext padding
> 1925:                 int i = pt.limit() - 1;
> 1926:                 for (; i > 0 && pt.get(i) == 0; i--);

Suggestion:

                for (; i >= pos && pt.get(i) == 0; i--);

src/java.base/share/classes/sun/security/ssl/SSLCipher.java line 2444:

> 2442:                 // remove inner plaintext padding
> 2443:                 int i = pt.limit() - 1;
> 2444:                 for (; i > 0 && pt.get(i) == 0; i--) {

Suggestion:

                for (; i >= pos && pt.get(i) == 0; i--) {

-------------

PR Review: https://git.openjdk.org/jdk/pull/27438#pullrequestreview-3262902812
PR Review Comment: https://git.openjdk.org/jdk/pull/27438#discussion_r2375791852
PR Review Comment: https://git.openjdk.org/jdk/pull/27438#discussion_r2375792997