A Novel Approach for Distributed Password Cracking Based on

Command Line

Kaushal Shah

and Dhaval Vasava

School of Technology,Pandit Deendayal Energy University, Gandhinagar, India

Keywords: Password, Hash Cracking, Distributed Cracking, Dictionary Attack, Brute Force Attack, Online Attack

Abstract: Password cracking process still continues to remain prevalent during a pentest or a digital forensic

investigation. In both cases, time is critically important. Most of the password cracking tools available today

can be used to crack different types of hashes and encryption media. Though these tools utilize system

components to enhance their performance, still it might not crack it in the required duration of time. This

process of cracking pass- words could be accelerated with the help of distributed password cracking technique.

This paper aims at studying and analysing some of the popular password cracking tools and to propose a new

command-line distributed password cracking tool which aims to resolve some of the issues with current

distributed password cracking tools.

1 INTRODUCTION

Password based authentication has been there for a

very long time and though the security provided by

such an authentication mechanism solely depends on

the user: the character set used by user while creating

the pass- word, the password length, the uniqueness

of the password, the guess ability of the password

through social engineering; it’s still widely used and

it unlikely to be replaced by other authentication

mechanism in upcoming few years (Hranicky`,

Zobal, et al. , 2020) (Pervan, Knezovi´, et al. , 2022),

(Pervan, Knezovi´, et al. , 2019), (HRANICKY´ ,

2022), (Jourdan and Stavrou, 2019). Therefore, it is

of utmost importance to set a strong password every

time during registration.

Passwords which are created without using any

password policy are generally considered to be weak

because a user will most likely set a password which

would be easy for him/her to remember and so the

user ends up either having a password of small length

or would select a well-known password and in both

cases, the attacker would crack those type of

passwords in few seconds with the help of some of

the popular password cracking tools which are freely

available today. Therefore, nowadays it has become

very common to use a password policy while

https://orcid.org/ 0000-0002-8494-9752

registering a user for a service and in that way users

will be forced to create a strong Password for

themselves (Siponen, Puhakainen, et al. , 2020),

(Guo, Zhang, et al. , 2019). The most basic password

policy requires the password to be of minimum length

8, the character set used for creating password should

contain at least one lowercase alphabet, at least one

uppercase alphabet, at least one digit and at least one

special character. As today’s users have a large

presence in the online world which ranges from social

media platforms to bank accounts, creating and

remembering each and every password is extremely

difficult for a user and so users may end up using the

same password for different accounts. Though this

problem can be easily resolved using a password

manager, there is a risk of losing everything once the

password manager itself is compromised. And also

with password manager, the passwords created are so

random in nature that remembering even a few of

them can be nearly impossible for some users. Like

every coin has two sides, password cracking too can

be used in a good way and in a bad way and it solely

depends on the user’s intention. While an attacker

can use a password cracking tool for gaining access

to someone’s account/system, a penetration tester can

use the same tool for auditing the password policy of

a company and its users. These password cracking

tools can prove to be very helpful during password

Shah, K. and Vasava, D.

A Novel Approach for Distributed Password Cracking Based on Command Line.

DOI: 10.5220/0013597600004664

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Futuristic Technology (INCOFT 2025) - Volume 2, pages 581-587

ISBN: 978-989-758-763-4

581

recovery of an account/system either hit by malware

or even in simple cases like when a user forgets

his/her password or during forensics.

There are many

different password cracking techniques like brute

forcing, dictionary based attacks, hybrid attack

which involves a combination of both brute forcing

and dictionary based attack, rainbow cracking

method for hash cracking and even through the use

of social engineering techniques (Rudy, Rodwald,

et al. , 2020), (Zion, 2018), (Shah, Patel, et al. ,

2020), (Shah and Jinwala, 2019) Some of the

existing popular password cracking tools

incorporate these techniques to provide flexibility

to the user while cracking a password.

Some of these

tools utilize system components like CPU and GPU

for effective password cracking. Even though with

such powerful tools sometimes it takes a very long

time to brute-force a password. Though one can

improve system configurations like allocating more

powerful CPUs, GPUs, RAM, storage etc but it then

increases the cost drastically. Some major pen-testing

organizations have a dedicated rig for password

cracking but other organizations, researchers or pen-

testers who would like to test a system against a

password cracking tool can’t simply afford such a rig.

Having a distributed password cracking tool in such

cases will aid in testing the system on the same level

as that of the rig but with minimum cost compared to

that of the rig. Utilizing processing power of other

hosts on the network will not only speed up the

process but it also means that every host will have a

lower load compared to when a single host utilizing

all its resources for the task of cracking a password

which will result in decrease of system performance

during other task done by user while the password

cracking task is running in background.

1.1 Our Contributions

We have reviewed some of the existing tools carried

out in the area of password cracking, which mostly

revolves around popular tools like Hashcat, John The

Ripper and THC Hydra, and distributed password

cracking, which involves tools like Fitcrack and

Hashtopolis. Based on the analysis of these tools, we

proposed a new light-weight command-line

distributed password cracking tool which covered

some of the flaws of the current password cracking

tools.

2 LITERATURE REVIEW

A review of work done in the area of password

cracking was carried out. It involves going through

some of the existing password cracking tools like

hashcat, JohnTheRipper and Hydra to learn about

their working, some of the novel techniques they use

to crack passwords and other features which make

them top choices for anyone in the cyber world in the

area of password cracking. Some of the existing

distributed password crackers like Hashtopolis and

Fitcrack have also been considered as part of the

literature review. Yisa et al. (Yu and Yin, 2021)

reviewed some of the top open-source password

cracking tools. The review paper starts with

discussing different forms of password authentication

and then the different types of password cracking

techniques like dictionary attacks, brute-force attack,

hybrid attack and rainbow tables. Later, for every

password cracking tool that was taken into

consideration, a brief overview and features were also

listed out. Some of the best practices for password

protection were also mentioned which includes the

use of salting for storing passwords, use of strong

passwords which are formed by combination of

capital letters, small letters, numbers and special

characters and the use of multi-factor authentication.

Pahuja and Sidana(Pahuja and Sidana, 2021)

compared different password cracking tools from an

implementation point of view. The authors

considered three tools for their study. Besides

considering pre-installed tools like Hashcat and John

The Ripper, they have also considered another tool

called Fcrackzip, which is different from other two,

as this one only focuses on zip password cracking.

John The Ripper and Hashcat mainly focuses on hash

cracking but the former could be used to crack zip

files with the help of additional tools like zip2john. A

comparative analysis of above mentioned tools was

also provided in a tabular form, which showed the

functionality/working of the tools as well as the

complexity associated with each of those tools.

Kakarla et al. (Kakarla, Mairaj, et al. , 2018)have

discussed a few password cracking tools with an

emphasis on THC Hydra. They started with first

discussing different pass- word cracking techniques

then proceeded to explain the technicalities of THC

Hybra by demonstrating attacks on FTP, SSH and

SMTP servers. Lastly, few protective measures that

are taken against password cracking were also

discussed. Hranicky` et al. (Hranicky`, Zobal, et al.

, 2019), in their paper, have first discussed the

need for dis- tributed password cracking. They used

the self-proclaimed fastest cracking tool, Hashcat, to

INCOFT 2025 - International Conference on Futuristic Technology

582

develop a distributed Hashcat tool using the BOINC

framework called Fitcrack. In their literature survey,

they considered many works that were done in the

area of distributed cracking but only considered

Hash- topolis for comparing their work as at that time

Hashtopolis was the only one, which was a popular,

well-maintained and open-source solution for dis-

tributed computing, as mentioned in their paper.

Then, the architecture of their proposed work was

discussed with a brief discussion on each of their

components. The tool incorporated different

distribution strategies for different types of attack

modes. Their analytical results and comparison,

shows that Fitcrack performs better than Hashtopolis.

Password cracking tools like Hashcat (Jens

Steube (Steube, 2015)), John The Ripper (openwall

(openwall, 2011)) and THC Hydra (vanhauser thc

(Yisa, Baba, et al. , 2016), each have their own set of

features that make them popular even today. Hashcat

supports over 300 types of hashes and it is the world’s

fastest password cracker, as mentioned in their

official website. John The Ripper can easily identify

the hash type of the hash to crack, whereas for hashcat

we need to have the hash type of the hash and then we

can proceed to cracking that hash. Though, there are

other separate tools like hash-identifier (Zion3R )

which can help identify the hash but that feature is

currently not available by hashcat. John The Ripper

can also crack zip files and ssh keys with the help of

other separate tools zip2john (openwall, 2011c) and

ssh2john (openwall, 2011). THC Hydra supports

many different protocols that one can use to conduct

an online password cracking attack. There isn’t a

single tool which incorporates all of these capabilities

and therefore our work somewhat focuses on doing

so. Though existing distributed password tools like

Fitcrack and Hashtopolis can significantly improve

the password cracking time, their installation and

setup process is quite lengthy and could be somewhat

complex for newbies. Also, as a node is entirely

dedicated to the server, we need to have at least two

nodes to work with it. We can definitely use Hashcat

for a single node, but that means that we need to have

more than one tool in our toolbox. In this paper, we

try to eliminate this issue, by developing a command-

line lightweight distributed password cracking tool in

python3, which could work both as a standalone

application as well as a distributed password cracking

application.

3 PROPOSED WORK

A light-weight tool written in python3 which could

serve as both a standalone application (like hashcat,

hydra etc) as well as a distributed application (like

Hashtopolis, Fitcrack, etc) for the purpose of

password cracking. By default, the tool will work as

a standalone application but it can be used as a

distributed application by simply providing the

appropriate flag/argument. The node/host could

either be used as a server machine or as a client

machine when using the tool as a distributed

application. Currently, it incorporates the following

attack techniques to crack the password : dictionary

based at- tack, bruteforce based attack and online

dictionary attack. The distributed password cracking

process starts with the hash type identification phase

as shown in Fig 1, in which the hash type of the

provided hash is categorized into the following hash

types : MD4, MD5 and the SHA-1, SHA-2, SHA-3

families, based on the hash length of the provided

hash(es). A character set check ensures that the

entered input is indeed a hexadecimal string and a

valid hash. The type of identification can result in

having multiple hash types for a provided hash. For

example, a hash of 512 bits, could either be a SHA512

hash or SHA3512 hash. In such cases, the tool takes

both of the hash types into consideration and so for

Figure 1: Flowchart of the proposed scheme

A Novel Approach for Distributed Password Cracking Based on Command Line

583

each guessed password, it calculates hash digest for

both the hash types and compares both of them with

the provided hash. After this hash type identification

phase, depending on the attack technique, wordlist

pre-processing phase in case of dictionary attack or

brute force setup phase in case of brute-force attack

starts. The entire process of the proposed password

cracking work is depicted in Figure 1.

3.1 Dictionary Attack

The process of wordlist-based attack starts with the

phase of pre-processing of wordlist(s). Support has

been provided to process and use multiple wordlists

in order to crack the hash. The preprocessing phase

includes reading the wordlist(s) and merging them to

generate a single wordlist (with no duplications of

words). That wordlist is then divided into smaller

chunks. The reason for having a pre-processing

wordlist phase is because, some of the popular

wordlists like ‘rockyou’ and ‘10-million-password-

top-list-1000000.txt’ by ‘seclist’, contain millions of

words and it is very likely that they have a high

percentage of overlapping words and this could result

in overall increase in the cracking time of the hash.

Also, there is no point in repeating the words in a

password cracking process. The wordlist pre-

processing process usually starts in a different thread

and leaves the main thread for the server to listen for

incoming connections from the clients. This way, the

overall performance of the program is improved

especially when there are a number of wordlists and

a number of clients to connect.

After the wordlists pre-processing phase and after

all the clients are connected, the server sends the

chunk of data to clients, which includes the hashes

that need to be cracked, the chunk that needs to be

used and the attack technique that needs to be used

for cracking the hash and some other configurations.

Each client including the server itself will participate

in the cracking phase and will utilize the chunk they

received to crack the hash. If the hash is cracked by

any of the clients, then it will send a message to the

server informing it about the same, along with the

cracked password. The server will then update its list

of hashes left to crack and will send that updated

hashes list to others in the network, when they send a

message re- questing the next chunk of data. This

process will continue until there are no hashes left to

crack or there are no chunks left with the server. At

the end of this whole process, the server will close all

the connections and list all the passwords cracked

during the complete process.

3.2 Bruteforce Attack

This process starts with the brute-force setup phase

on the server side, in which, based on the given

configuration options/arguments such as pass- word

pattern/charset, minimum and maximum password

length, all possible characters at the

first/start/leftmost position of the word is guessed and

then a list of tuples is generated, where the first item

of the tuple would be one of all the possible characters

that were guessed for the first position, whereas the

second argument/element in the tuple will indicate the

length of strings that are to be generated starting with

the character at the first item. For example, (a,8)

indicates that the node having this tuple will have to

generate all the strings starting with character ‘a’ and

of length ‘8’. After this setup phase, the cracking

phase begins where each node in the network will

take one tuple from the list of tuples and will perform

the brute force attack. As the process of handling

client and bruteforcing are on separate threads

(Python-Software-Foundation, 2012b), the server

will also participate in the cracking phase. Unlike

dictionary attack, which is a I/O bound task,

bruteforce attack is a CPU bound task and so we have

used the concept of multiprocessing (Python-

Software-Foundation, 2012a) for increasing the

performance of the system. By default, it uses 10

processes to do the work. As with the dictionary

attack, this process will continue until there are no

hashes left to crack or there are no tuples left with the

server. In case a hash has been cracked, the client and

the server will work the same way as it was in the

dictionary attack. At the end, all the connections are

closed and all cracked passwords are listed by the

server.

3.3 Online Attack

For the online attack mode, the flow will be similar to

that of a dictionary- based attack except that instead

of cracking the hash, this time it will be working on

cracking the password of the login page of a web app.

For such an attack, it uses selenium under the hood.

Selenium not only takes care of the cookies, but it also

takes care of the other security elements like the

hidden fields of a web app, if any. And this was the

reason selenium was preferred ahead of the ‘request’

module, which is generally used when dealing with

requests in Python. The user only needs to provide the

target url, the username and the wordlist to work with.

The server will then send the chunks that were

generated during the pre-processing wordlist phase to

clients along with the target url and the username.

INCOFT 2025 - International Conference on Futuristic Technology

584

Each of the nodes, including server, will then

initialize the cracking process by running the browser

through selenium in headless mode and it will then

find the appropriate location in the web page (target)

for entering the username and the guessed password

from the wordlist. As selenium requires appropriate

drivers to run a browser, another module ‘webdriver-

manager’ was used to download the necessary drivers

for selenium to work with. The process in case of any

successful login, will be the same as it was for the

successful hash crack process in brute force and

dictionary attack except that the server will not

inform others on the network as it did in other two

attacks. The process ends when there are no chunks

left to work with.

4 RESULTS AND ANALYSIS

For comparative analysis of the tool, two distinct

physical machines were used, one acting as a server

and the other as the client machine. As shown in the

table 1, two completely different nodes with different

architecture and OS were used to carry out the test.

Both nodes also differ in memory space allotted to

them to see if there is any performance issue because

of it. The nodes were communicating with each other

on a home network, rather than having a completely

isolated environment as in most of the real-world

scenarios we would have regular traffic going through

the network. And so it’s important to test the

performance of these tools in such an environment.

Table 1: System Configuration of server and client machine

Server Machine

Client Machine

Ubuntu 20.04.5

LTS

x86 64

Kali GNU/Linux

Rolling x86 64

Kernel

5.15.0-52-generic 5.16.0-kali7-amd64

CPU

AMD Ryzen 5

4600H

with Radeon G

Intel i5-2300 @

2.793GHz

GPU

VMware SVGA

ada

VMware SVGA

ada

Memory

3924 MiB

2938 MiB

For dictionary attack, some of the current popular

wordlists were used, like ‘rockyou.txt’ (which is

already available in Kali Linux under the

‘/usr/share/wordlists/’ directory) and all the wordlists

by seclists (Miessler, 2012) under the passwords

category. The results of this test are shown in table 2.

The cracking time of Hashtopolis and DPC (proposed

tool) were noted on using different wordlists. And the

hash of the last word in the list was used in each

of the experiments. Note that the time noted was in

the format : ‘hh:mm:ss’. The category ‘all’ in table 1

indicates that all the wordlists from SecLists (it

includes the following wordlists :‘10-million-

password-list-top-100.txt’, ‘10- million-password-

list-top-1000.txt’,‘10-million-password-list-top-

10000.txt’, ‘10-million-password-list-top-

100000.txt’, ‘10-million-password-list-top-

1000000.txt’. ‘10-million-password-list-top-500.txt’,

‘100k-most-used-passwords-NCSC.txt’, ‘10k-most-

common.txt’, ‘1900-2020.txt’, ‘500-worst-

passwords.txt’, ‘SplashData- 2014.txt’, ‘SplashData-

2015-1.txt’, ‘SplashData-2015-2.txt’, ‘best1050.txt’,

‘best110.txt’, ‘best15.txt’, ‘common-passwords-

win.txt’, ‘medical-devices.txt’, ‘top-20-common-

SSH-passwords.txt’, ‘top-passwords-shortlist.txt’,

‘worst-passwords- 2017-top100-slashdata.txt’) +

‘rockyou.txt’ were used.

As seen in table 2, in some cases the proposed

DPC tool works better than Hashtopolis, while in

some cases the Hashtopolis is better than the DPC

tool. Thus, we can say that they are very close to each

other in terms of performance.

Table 2: Dictionary based attack comparison

Wordlist

No of

words

Hash

algortihm

Hashtopolis

DPC

10-million-

password-

list-

top-

1000000.txt

999,998

SHA1

09s 5s

rockyou.txt

14,344,392

MD5

29s 32s

10-million-

password- list-

top-

1000000.txt

rockyou.txt

15,344,390

SHA1

8s 6s

10-million-

password- list-

top-

1000000.txt

rock

ou.txt

15,344,390

MD5

31s 40s

All

15,613,886

SHA1

3m 45s 33s

All

15,613,886

MD5

3m 10s 37s

In cases where multiple wordlists were used (as in

‘All’ category - where a total of 22 wordlists were

used), DPC performs way better than Hashtopolis

because of its wordlist preprocessing phase (where

repeated words among wordlists are eliminated), as

seen in Figures 2 and 3.

Comparison with FitCrack was not possible,

because the errors during installation and setup of

A Novel Approach for Distributed Password Cracking Based on Command Line

585

FitCrack client were not resolved within the duration

of this experiment. Thus, we can say that between

Hashtopolis and DPC, the latter performs better when

it comes to working with multiple wordlists.

Figure 2: SHA1 hash cracking performance comparison

Figure 3: MD5 hash cracking performance comparison

5 CONCLUSIONS

Even today passwords remain one of the most

dominant authentication methods. And so security of

passwords still plays a vital role in overall security of

the organization. Mainly, weak passwords are the

ones that are cracked easily while strong passwords

take relatively longer time to crack. The time to crack

may vary depending on factors like password length,

character set used to create password, the hash type

used to encrypt, the hardware con- figuration of the

system used for cracking etc. Password cracking

process, if successful, can become a valuable factor

in deciding the outcome of a cyber investigation. And

to further speed up the process of cracking, we can

make use of distributed password cracking tools.

Tools like Hashtopolis and FitCrack which are built

on top of hashcat do pretty good work when it comes

to distributed password cracking, provided they are

already installed and configured properly to use. But

when it comes to instantaneous use, these tools take a

considerable amount of time for installation and

initial setup. As the proposed tool is written in Python

and is bound to be used as a command line tool, it is

simply a matter of time to install and it’s ready to use.

The server also participating in cracking helps in

cases where there are not more than two nodes

available. And also gives the tool the ability to work

as a standalone application. Merging wordlist features

of this tool can prove to be useful when working with

multiple wordlists at a time. And with support for

online attack (http/https protocol), it stands out

among the other tools.

For making a one-stop solution for password

cracking, an extended work to make it feature-rich

application can be done. Various features like adding

support for other hash types, support for other

protocols and support for GPU based cracking can be

added to enhance the capabilities of the tool. In order

to make sure that the server and client communicate

in a secure fashion, efforts can be made in using ssh

protocol for communication.

In cases where the security/investigation team

have weak infrastructure, distributed password

cracking tools can prove to be very helpful, as it uses

the computing power of multiple nodes for cracking

which in turn depicts the behaviour of a high-end

system. Hence, we can say that there is a need for

distributed password cracking tools considering the

challenges we face today in penetration testing and

digital forensics.

REFERENCES

Dave, D., Kava, M., Gupta, R. K., and Shah, K. (2021).

Deep learning approaches for intrusion detection

system. In 2021 IEEE International Conference on

Technology, Research, and Innovation for Betterment

of Society (TRIBES), pages 1–6. IEEE.

Guo, Y., Zhang, Z., and Guo, Y. (2019). Optiwords: A new

password pol- icy for creating memorable and strong

passwords. Computers & Security, 85:423–435.

HRANICKY´, R. (2022). Digital forensics: The

acceleration of password cracking.

Hranicky`, R., Zobal, L., Ryˇsavy`, O., and Kol´aˇr, D.

(2019). Distributed password cracking with boinc and

hashcat. Digital Investigation, 30:161– 172.

Hranicky`, R., Zobal, L., Ryˇsavy`, O., Kol´aˇr, D., and

Mikuˇs, D. (2020). Distributed pcfg password

cracking. In European Symposium on Research in

Computer Security, pages 701–719. Springer.

Jens Steube, G. G. (2015). Hashcat - advanced password

recovery. https://hashcat.net/hashcat/. (last accessed on

20/11/22).

Jourdan, P. and Stavrou, E. (2019). Towards designing

advanced pass- word cracking toolkits: Optimizing the

INCOFT 2025 - International Conference on Futuristic Technology

586

password cracking process. In Adjunct Publication of

the 27th Conference on User Modeling, Adaptation and

Personalization, pages 203–208.

Kakarla, T., Mairaj, A., and Javaid, A. Y. (2018). A real-

world password cracking demonstration using open

source tools for instructional use. In 2018 IEEE

International Conference on Electro/Information

Technology (EIT), pages 0387–0391. IEEE.

Miessler, D. (2012). Common-credentials. https://github.

com/danielmiessler/SecLists/tree/master/Passwords/C

ommon-Credentials. (last accessed on 20/12/22).

openwall (2011a). Johntheripper.

https://github.com/openwall/ john. (last accessed on

22/11/22).

openwall (2011b). ssh2john.

https://github.com/openwall/john/ blob/bleeding-

jumbo/run/ssh2john.py. (last accessed on 05/12/22).

openwall (2011c). zip2john.

https://github.com/openwall/john/ blob/bleeding-

jumbo/src/zip2john.c. (last accessed on 03/12/22).

Pahuja, D. and Sidana, P. (2021). Implementing and

comparing different password cracking tools.

International Research Journal of Engineering and

Technology.

Pervan, B., Knezovi´c, J., and Guberovi´c, E. (2022).

Energy-efficient dis- tributed password hash

computation on heterogeneous embedded system.

Automatika, 63(3):399–417.

Pervan, B., Knezovic, J., and Pericin, K. (2019). Distributed

password hash computation on commodity

heterogeneous programmable platforms. In 13th

USENIX Workshop on Offensive Technologies

(WOOT 19).

Python-Software-Foundation (2012a). Multiprocessing.

https://

docs.python.org/3/library/multiprocessing.html. (last

accessed on 10/12/22).

Python-Software-Foundation (2012b). Threading.

https://docs. python.org/3/library/threading.html. (last

accessed on 10/12/22).

Rudy, J. and Rodwald, P. (2020). Job scheduling with

machine speeds for password cracking using

hashtopolis. In International Conference on

Dependability and Complex Systems, pages 523–533.

Springer.

Shah, K. and Patel, D. (2020). Exploring the access control

policies of web-based social network. In ICDSMLA

2019: Proceedings of the 1st International Conference

on Data Science, Machine Learning and Appli- cations,

pages 1614–1622. Springer.

Shah, K. A. and Jinwala, D. C. (2019). Novel approach of

key predistri- bution for grid based sensor networks.

Wireless Personal Communications, 108:939–955.

Siponen, M., Puhakainen, P., and Vance, A. (2020). Can

individuals’ neutralization techniques be overcome? a

field experiment on password policy. Computers &

Security, 88:101617.

vanhauser thc (2014). Thc-hydra.

https://github.com/ vanhauser-thc/thc-hydra. (last

accessed on 22/11/22).

Yisa, V., Baba, M., and Olaniyi, E. (2016). A review of top

open source password cracking tools. International

Conference on Information and Communication

Technology and Its Applications.

Yu, F. and Yin, H. (2021). Password cracking of pdf 2.0

documents on gpu. In 2021 IEEE 6th International

Conference on Computer and Communication Systems

(ICCCS), pages 721–725. IEEE.

Zion3R (2018). hash-identifier.

https://github.com/blackploit/ hash-identifier. (last

accessed on 05/12/22).

A Novel Approach for Distributed Password Cracking Based on Command Line

587