MySQL身份认证漏洞 升级到5.5.24可修正

这个BUG实际上早在4月份就被发现了，今年5月7号，MySQL发布5.5.24的时候，修正了这个BUG。

漏洞分析：

出问题的代码如下

/*
    Check that scrambled message corresponds to the password; the function
    is used by server to check that recieved reply is authentic.
    This function does not check lengths of given strings: message must be
    null-terminated, reply and hash_stage2 must be at least SHA1_HASH_SIZE
    long (if not, something fishy is going on).
  SYNOPSIS
    check_scramble()
    scramble     clients' reply, presumably produced by scramble()
    message      original random string, previously sent to client
                 (presumably second argument of scramble()), must be
                 exactly SCRAMBLE_LENGTH long and NULL-terminated.
    hash_stage2  hex2octet-decoded database entry
    All params are IN.

  RETURN VALUE
    0  password is correct
    !0  password is invalid
*/

my_bool
check_scramble(const uchar *scramble_arg, const char *message,
               const uint8 *hash_stage2)
{
  SHA1_CONTEXT sha1_context;
  uint8 buf[SHA1_HASH_SIZE];
  uint8 hash_stage2_reassured[SHA1_HASH_SIZE];

  mysql_sha1_reset(&sha1_context);
  /* create key to encrypt scramble */
  mysql_sha1_input(&sha1_context, (const uint8 *) message, SCRAMBLE_LENGTH);
  mysql_sha1_input(&sha1_context, hash_stage2, SHA1_HASH_SIZE);
  mysql_sha1_result(&sha1_context, buf);
  /* encrypt scramble */
    my_crypt((char *) buf, buf, scramble_arg, SCRAMBLE_LENGTH);
  /* now buf supposedly contains hash_stage1: so we can get hash_stage2 */
  mysql_sha1_reset(&sha1_context);
  mysql_sha1_input(&sha1_context, buf, SHA1_HASH_SIZE);
  mysql_sha1_result(&sha1_context, hash_stage2_reassured);
  return memcmp(hash_stage2, hash_stage2_reassured, SHA1_HASH_SIZE);
}

memcmp的返回值实际上是int，而my_bool实际上是char。那么在把int转换成char的时候，就有可能发生截断。比如，memcmp返回0×200，截断后变成了0，调用check_scramble函数的就误以为“password is correct“。

但是一般来说，memcmp的返回值都在[127,-128]之内。glibc的经SSE优化后的代码，不是如此。所以这个BUG只在特定的编译环境下才会触发：即编译MySQL的时候加了-fno-builtin，并且所使用的glibc是经SSE优化后的（一般系统自带的都是如此）。这里所说的 glibc是指Linux的glibc，FreeBSD的libc不受影响。

总的来说这个BUG还是比较严重的，上次MySQL出现这样的BUG还是在3.23/4.0时代。