A method and system for detect SQL sentence security of agricultural database

Abstract

The invention discloses an agricultural database SQL sentence safety detection method, comprising the following steps: (1) extracting a source SQL sentence in the database, carrying out syntax analysis, if the syntax analysis is wrong, terminating; Otherwise, it is judged whether the source SQL statement is in the white list or not, and if the source SQL statement is already in the white list, themethod ends; Otherwise, lexical analysis is carried out; (2) if the lexical analysis is incorrect, it ends; Otherwise, the source SQL statement is added to the whitelist, and the word structure withreference count under the source SQL statement is generated. (3) Collect multiple SQL statements which contain the same table structure as the table accessed by the source SQL statement in the currentwhitelist, compare the word structure of the source SQL statement under different length with the word structure of multiple SQL statements extracted from the whitelist, and then judge whether the source SQL statement is a safe access statement or not. The invention can not only be safe but also can not block dynamically spliced SQL sentences, so as to ensure the normal use and safety of the background agricultural database.

Description

technical field [0001] The invention relates to the field of database security, in particular to an agricultural database SQL statement security detection method and system. Background technique [0002] The security of agricultural background data has always attracted much attention. The data interacts with the external network and the database, and many security products have been derived. The database firewall is one of the important security products. There are many types of database firewall implementations, most of which implement high-risk operations such as signature database, whitelist, blacklist, SQL injection defense, limited update and deletion, limited update and deletion without where, and limited create and drop. [0003] However, there are always two states that cannot be exhaustive for the whitelist: [0004] 1. There is no guarantee that all SQL statements involved in complex business systems can be recorded into a whitelist within a certain period of time...

AI Technical Summary

Problems solved by technology

[0009] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a method and system for detecting the security of SQL statements in agricultural databases, which solves the problem that the white list cannot be exhaustive and the potential safety hazards caused by the release of unallowed statements , Problems that affect database performance

Images