Pages

[Gregg]

This release is a bug fix & stability release.

This release includes:
APR Version:        1.7.0
APU Version:        1.6.1
Brotli Version:    1.0.9
Jansson Version:    2.13.1
Libcurl Version:   7.78.0
LibXML2 Version:    2.9.12
LUA Version:        5.2.4
NGHTTP2 Version:    1.44.0
OpenSSL Version:    1.1.1l or LibreSSL 3.3.4
PCRE Version:       8.44
SQLite3 Version:    3.36.0
ZLib Version:       1.2.11


You can get your copy of the new Apache HTTP Server from our download page.

OpenSSL changes:

  *) Fixed an SM2 Decryption Buffer Overflow.

     In order to decrypt SM2 encrypted data an application is expected to call the
     API function EVP_PKEY_decrypt(). Typically an application will call this
     function twice. The first time, on entry, the "out" parameter can be NULL and,
     on exit, the "outlen" parameter is populated with the buffer size required to
     hold the decrypted plaintext. The application can then allocate a sufficiently
     sized buffer and call EVP_PKEY_decrypt() again, but this time passing a non-NULL
     value for the "out" parameter.

     A bug in the implementation of the SM2 decryption code means that the
     calculation of the buffer size required to hold the plaintext returned by the
     first call to EVP_PKEY_decrypt() can be smaller than the actual size required by
     the second call. This can lead to a buffer overflow when EVP_PKEY_decrypt() is
     called by the application a second time with a buffer that is too small.

     A malicious attacker who is able present SM2 content for decryption to an
     application could cause attacker chosen data to overflow the buffer by up to a
     maximum of 62 bytes altering the contents of other data held after the
     buffer, possibly changing application behaviour or causing the application to
     crash. The location of the buffer is application dependent but is typically
     heap allocated.
     (CVE-2021-3711)
     [Matt Caswell]

  *) Fixed various read buffer overruns processing ASN.1 strings

     ASN.1 strings are represented internally within OpenSSL as an ASN1_STRING
     structure which contains a buffer holding the string data and a field holding
     the buffer length. This contrasts with normal C strings which are repesented as
     a buffer for the string data which is terminated with a NUL (0) byte.

     Although not a strict requirement, ASN.1 strings that are parsed using OpenSSL's
     own "d2i" functions (and other similar parsing functions) as well as any string
     whose value has been set with the ASN1_STRING_set() function will additionally
     NUL terminate the byte array in the ASN1_STRING structure.

     However, it is possible for applications to directly construct valid ASN1_STRING
     structures which do not NUL terminate the byte array by directly setting the
     "data" and "length" fields in the ASN1_STRING array. This can also happen by
     using the ASN1_STRING_set0() function.

     Numerous OpenSSL functions that print ASN.1 data have been found to assume that
     the ASN1_STRING byte array will be NUL terminated, even though this is not
     guaranteed for strings that have been directly constructed. Where an application
     requests an ASN.1 structure to be printed, and where that ASN.1 structure
     contains ASN1_STRINGs that have been directly constructed by the application
     without NUL terminating the "data" field, then a read buffer overrun can occur.

     The same thing can also occur during name constraints processing of certificates
     (for example if a certificate has been directly constructed by the application
     instead of loading it via the OpenSSL parsing functions, and the certificate
     contains non NUL terminated ASN1_STRING structures). It can also occur in the
     X509_get1_email(), X509_REQ_get1_email() and X509_get1_ocsp() functions.

     If a malicious actor can cause an application to directly construct an
     ASN1_STRING and then process it through one of the affected OpenSSL functions
     then this issue could be hit. This might result in a crash (causing a Denial of
     Service attack). It could also result in the disclosure of private memory
     contents (such as private keys, or sensitive plaintext).
     (CVE-2021-3712)
     [Matt Caswell]

Change Logs for LibreSSL

3.3.4