Index Chain Is Expected That The Algorithm To Attack The Password Hash

To attack password hash, a traditional brute-force method based on exhaustive search need a lot of computing power or a lot of time to complete.In this dissertation, the author put forward the index chain pre-computation method. While a traditional brute force cracker try all possible plaintexts one by one in cracking time, index chain method works in another way. It precomputes all possible plaintext - ciphertext pairs in advance and convert and store them in the file so called "index chain table". It may take a long time to precompute the tables, but once the one time precomputation is finished, you will always be able to crack the ciphertext covered by the index chain tables in seconds.The index chain tables produced by the index chain pre-computation method do not contain the same data as the "plaintext-hash" pairs which are calculated by the traditional brute-force method in its attacking process. The index chain table contains index chains, each of which contains two indexes, the first is called index head, and the other is index tail.Both index head and tail index are indexes corresponding to plaintexts, in other words, they are actually the plaintext' s ordinal numbers in the total plaintext combination space.The index head is an 8-byte number created by random method, and it should be kept within the range of total plaintext combination space.The index tail is computed from the index head by repeating the following steps several times:1. convert index to plaintext2. convert plaintext to hash using related hash algorithm3. convert hash to another index using reduction functionIt's evident that using "index head-index tail" pair to replace the "plaintext-hash* pair can significantly decrease the disk space needed.To crack a password hash, we first convert it to an index, then search it in the index chain table by comparing it with the index tails. If it can not be located, try next index computed by using 3 steps stated above. After the index is located on one index chain, we count the times repeated to do the 3 steps, say N times, then walk backward N+l times from the index tail, so that we position the index whose plaintext may be responsible for the hash we want to crack. If that index' s plaintext is proved to be not associated with the hash we are cracking, then continue the searching process.