#cybersecurity
7 messages · Page 18 of 1
i mean i dont know what im looking for lol
im looking for a way to pass binary numbers to the hash function
i have no idea if this is right
What do you mean by binary numbers?
you know how i generate random strings for my project now
Yep I see that
the prof told me to also generate hashes for all binary numbers up to 2^16 and analyze hashes for that too
so if i make a for loop and just pass the loop variable which is type string
i dont know if thats the same
im assuming no
Okay then sounds like you just need to pass in the bin(n)[2:]
im not sure the hash function will take bin(n)[2:] as argument but i havent tried it
Bin returns a string
yes but what happens when you encode that string
Hence the idea of the binary numbers being a little weird, either that means some encoding of strings of ones and zeros, or it just means all the numbers.
What interesting behaviour are we expecting?
lol yeah im lost
which way is "correct" lol
as in passing the actual bytes representation of a binary number
The objective isn't well defined so it's impossible to say, i don't think you want the bytes method as that's treating the ones and zeros as decimal digits, the truncated bin output would seem to maintain the idea of dealing with 'binary numbers'.
Or rather 'stringified binary representations of numbers'
so the second approach where i first convert to binary then to_bytes yeah?
Nope just the bin(n)[2:]
But since you're hashing them I don't think you're going to see any interesting biases at all.
Is there a way to encrypt a string in a way that allows it to be decrypted afterwards using a key?
Can you define key?
@compact sail Yes there is
Because there exists Asymmetric encryption and Symmetric encryption. Asymmetric is like public key/private key, like SSH or PGP.
Whereas Symmetric is like AES, and can be decrypted using a passphrase.
I'm looking for a reliable way to store a user's password in my software
use a database
oh wait hol up
no but you see the password has to be stored
https://passlib.readthedocs.io/en/stable/ - Hash the passwords with a master password
if so, the password is correct
because the program is supposed to auto-login into a website
that's kind of the point of the program
for the user to not have to constantly type in their credentials
like you can do requests to a webserver
I see, is this some kind of password manager
nope
@hexed quest I don't think you understand
so essentially my school has this really bad learning management system
and I'm writing an app that makes it usable
just a gui that allows you to see your lessons and tasks quickly
it webscrapes info from the bad website and makes it readable, however the website requires a user and a pass
and whatever you do you have to enter it in, and that's annoying
so I made a system that saves the user's username and password
but I need a way to make it secure
@thorn obsidian explain in pm
@hexed quest Please don't PM me
@compact sail What would prevent you from using KeePass/Bitwarden/etc then?
that still doesn't solve the issue of the website having extremely bad design
it'd be ideal if I could kind of "lock" the string
with a key
the string?
i was thinking of something like this as a system... but I don't know if it'll be secure:
user's password --> encrypted with user's password as a key --> encrypted piece of text
encrypted piece of text --> attempted decryption using user's password as a key --> either gibberish or user's password
the password
I don't see why this would be better than a password database that has already been audited.
What website(s) are extremely bad design?
Daymap, my school's learning management system.
(not a security expert) I use gpg in symmetric mode to encrypt some of my files, there seems to be a library called python-gnupg that lets you interface with gpg. That way you can have a single passphrase to decrypt the user data.
let me go check that out, seems like what I need for the most part 👍
Yep iirc that's the command line flag
I want to start cryptopals.com training but first I would like to get some advise about the topics that I am going to learn.
The first challenge in the first set is something that I can do in 5 minutes, or may be I could just use some online converters to do it for me
Now what type of approach should I choose to learn crypto:
1: Read theory about Hexadecimal & base64, then learn how to convert them using some libraries made by other people
or
2: Read theory about Hexadecimal & base64, implement **your own** libraries or code to convert them
or
3: Read theory about Hexadecimal & base64, convert stuff manually on paper then create tools from scratch for automation
Which one do you think I should do to upgrade my python skills as well crypto?
@obtuse harness you should use a project with 1
hexadecimal/base64 can be done in the standard library
they are also not crypto
this exercice is really supposed to be a warm up for cryptopals, don't spend a month on it imo
thanks @versed scroll
@thorn obsidian well I am talking about libraries in general, do I need to read those libraries and see how they do the converting? is it really necessary? or just know what is Hex and how to do encode/decode?
@hoary marten thanks
Look at base64 - https://docs.python.org/3/library/base64.html and https://docs.python.org/3/library/codecs.html , as well as how to encode() and decode()
@thorn obsidian well this is pretty easy and I can finish set1 in few hours, I thought I needed to look inside the libs
thanks
@obtuse harness np
"oops"
@thorn obsidian I found AMD's RDRAND implementation's source
You sure that isn't Sony's? zing

@marble dawn What type of pen-testing is not red-teaming?
Pentesting is more about finding all the issues you can find
red-teaming is more about a focused effort to figure out the realistic risks an organisation is taking
it's hard to explain in so many words, but this seems like a good explanation https://www.redteamsecure.com/penetration-testing-vs-red-teaming-whats-difference/
Ok, so it's kinda like pen-testing is a wide-reaching security evaluation, while red-teaming is like a stress test.
First you do pen-testing and shore up your defenses, and then you do a red-team test to see if it paid off?
I think it's more that pentesting is all about approaching a company through the eyes of a bad guy
vs red-teaming being more like standardised testing and security policy review
although they more or less address the same issues I think
Are you sure it's not the other way around?
Cause it sounds like you flipped the definitions
Judging by that article at the very least
Red teamers take their time, wanting to avoid detection (just as the cybercriminal would).
One entertaining way to look at it is that the pen testers are pirates — ready to rampage and pillage wherever and whenever they can. In this analogy, red teamers would be more like ninjas, stealthily planning multi-faceted, controlled, focused attacks.
It's kind of hard to define honestly
I think deviant ollam has a video on the difference
So I want to get into ethical hacking.. .
should i get a raspberry pi for hacking...and what are the advantages
Is there any need to set up a passphrase for RSA?
I am trying to set up SSH authentication but I am not sure why the guide says that using RSA without passphrase is insecure
passphrase in rsa context is that whenever you want to use your pvt key, you need to enter a passphrase
secure in this context is that even if your hard disk / usb / whatever, somehow your pvt key is compromised, other people can't use it still, since they don't have your pass
@tepid rover it might be the thing you're SSHing to is a critical server. Normally even if someone got hold of your computer, they couldn't access the server without a login password even if they stole your laptop. But if you have passphrase-less key, then they just copy the key off your computer and can now access the server
@spiral iron https://danielmiessler.com/study/red-blue-purple-teams/
Found this, might be worth a read(?)
Ok, so according to this article, a red team is characterized by trying to simulate an attack by a particular threat actor over an extended period of time. Whereas a pentester tries to attack through any and all attack vectors for shorter periods.
But, they also note that:
There is debate on this point within the community.
@quartz terrace Even if that was the case, I'd hope you were using FDE
I believe red team is more physical, isn't it?
I'm bad at explaning - but I think that isn't part of the red/blue team approach to it, but rather just physical pen testing
Physcial pentesting could be just a guy with a clipboard social engeineering your front desk secretary to giving access to X area that is supposed to be off limit
"Hey, I'm Gregory and I'm here to test the elevator."
hi-viz jacket and a ladder are a better lockpick than any literal lockpick
No staff's going to mistake a guy in a hoodie with a lockpick. But a hi-viz jacket and a ladder... and a lockpick? Oh the dude's just a contractor.
Any good security modules to use?
then you need to choose another language than python
a surprisingly common question recently
Yeah, oddly enough.
Don't want people to be able to steal.it
No matter what, if the client can run the code, if they are motivated enough, they can take anything from it. Some languages hide it better than others, but if the client runs the code, their machine has full knowledge of at minimum the byte code it compiles down to. Without more info about what you re actually trying to protect, there's no real reccomendations to be had.
if the code runs at some point, it can be read. it's simply about how hard you make it to extract the code. a bit like with video copy protection, if you see the video you can copy it, even if that means resorting to recording your screen with a camera (disclaimer: dont do this piracy is illegal)
I'm looking for a good key Auth system for my code anyone know any?
@lost crystal Can you describe what you mean by "key auth system"?
my friend recently gave me this audio file to try to decode, i searched far and wide but I don't know what method was used to encode it and how I can decode it, help!
¯_(ツ)_/¯
To sanitize my datas input and avoid sql injection, using psycopg2, is parsing the inputs as parameters sufficient?
Nice
Getting address unreachable error, here's the hackerone report https://hackerone.com/reports/679969
@neon jewel They were working on their home lab apparently, should be up now.
@blissful raven use prepared statements
The Labour party has faced a second cyber-attack, a day after experiencing what it called a “sophisticated and large-scale” attempt to disrupt its digital systems.
ddos
sophisticated
please allow me to laugh
The DDoS that happened to Spamhaus was sophisticated because of the sheer size of it.
renting a botnet from dnm requires no sophistication
dnm?
darknetmarkets
@thorn obsidian
sorry, ii've just realised it's not a term everyone's used to
🤔
List of Awesome Red Team / Red Teaming Resources
https://github.com/yeyintminthuhtut/Awesome-Red-Teaming/blob/master/README.md
@thorn obsidian There's a lot there.
@violet notch 
I've not gone through it all, but yeah.. There is a lot 🙂
I want to learn about security
there are many areas in security
like?
?
like lock picking? like when you put multiple doors / laser + alerts in your vault?
Computer security types can be like: malware, phishing, networking, etc @devout sage
Protecting code
Encryption
sup, I'm studying cyber security second year and I am wanting to work on some sort of project in my spare time but i'm not sure what I could do really
any ideas?
wouldn't really be projects or python programming in my experience
school probably gonna teach you some math
practically you wanna understand how asymmetric crypto applied in TLS
and OWASP if you're into web security
Already covered maths and stuff, I just want to get some more work done in my spare time so I could talk about it in my interviews for my internships
if you're into CTF, then some interesting topics that can make use of python:
- crypto/math
- RE - binary exploits (use python to construct payloads)
Yeah I guess so
for web, there is heavily uses of python requests, practical too
Last year I made subnet calculator on python where you input ip and amount of bits it will have an it calculates all available ip's etc
for various network exploits, there's python socket (over TCP)
So i am looking to do another project like that, some kind of tool
yeah more security related
something on top of my mind: make an "app" that run various nmap commands, and produce a friendly reports on a target server
don't spam nmap random servers online though you can be sued
Yeah that sounds something I could do
Reckon I could do some sort of tool that checks for vulnerabilities in websites?
they're usually kinda wack
@thorn obsidian, it looks like you tried to attach a file type we don't allow. Feel free to ask in #community-meta if you think this is a mistake.
not very useful / lots of false positive
Ok then
Ah i see
but ZAProxy has some API iirc?
@peak aspen ill send u a pdf that teaches u good stuff about hacking with python
yeah sure
hacking with python
okay
i'm speaking with python in mind
I will check it out for sure and see what is there
Its introductory stuff but cool
keep in mind that practical security usually comes in forms that don't really fit these "toy projects"
Well at least what you need to know to learn it all
No what i sent him literally just teaches u malicious things to do with python
commonly you're either on the developer's side - make your system more reasonably secure,
or tester side - a lot less on programming, but more on using tools to poke a system
Yeah
if you're wanna be a tools developer that helps testers, now that's too specific of a job
This is great stuff, we are only just started working with linux and started scripting with shell and this allows me to get head start using python
Yeah i think theres some linux stuff in there too
@thorn obsidian how old is this?
Idk
I dont even remember where i found jt
Its a book is alls i know
2017 @peak aspen
yeah just found out
might be some what outdated
but I will for sure study it a little and see whats popping
Eh still pretty decent i doubt youd find any trouble
I have never done this but how hard would be to put on web tkinter python app?
@bold grotto
on tester side - a lot less on programming, but more on using tools to poke a system
as someone "on the tester side", 90% of the tools I use I wrote myself
with some exceptions being the obvious like burp, binwalk, etc
and yes, most of them (the tools i've written) are in python
two in golang
sure
you'd be amazed how many badly written apps/ badly configured servers out there that get serviced to pentesters
in such case i don't need to spend most of my time developing tools, instead i'll start using existing tools first
just because your specific job is more demanding in developing your own tools doesnt mean typical pentesters do what you do
the more common programming would be developing exploits / PoC, specific to particular apps serviced for the pentesting project
my specific job is pentesting
I've just found most tools I've used have been lackluster or just entirely nonexistent
as for pocs, each finding we list in the report is accompanied by a poc
"no poc no proof"
usually, the poc is just a step-by-step of what you need to to do to reproduce the issue
however, a lot of findings have attack chains that are either too complex or too long to write out step-by-step, so a poc script or such is provided instead
and you're saying your pentesting role in your team applies to majority of pentesters?
that was not my point what you do, how could i possibly know? i didnt need to know either, i'm sure smart people like you, who test complicated systems exist, that no open source tool can even provide what they need.
doesn't mean thats the only security / pentesting job you should aim to get
So what is your pentesting role then @bold grotto ? Since you know so much about this topic.. 🤔
my previous role wasn't like an on going pentest / part of devops / QA for a specific project / team, which i think xx tasks would apply more
since you grow familiar with the team development habit and can identify which tools to make for reusability
@bold grotto if your job is defined by the tools you use, what's the point of the job? anyone can run a tool
and what I said applies to pretty much everyone in my company
we got a lot of widely different apps from different clients
so we'd start right away poking around their services they expose publicly, intentionally or not
at the same time, the main app in scope required for testing will also get tested (different testers on the same project)
tasks carried depend on whether the app is commercial use, or internal use
also,
my previous role wasn't like an on going pentest / part of devops / QA for a specific project / team, which i think xx tasks would apply more
I work on 3-4 different projects (from different clients) a week
you tests 3 different projects a week, and you need to map out the whole functionality of the app while finding vulns,
and your main tasks are developing tools?
okay
when did I say my main task was developing tools?
why bring that up then
the tools I mostly use I've already written before, sure I change functionality when and where I see fit but it's barely "active development"
if you read my initial comment, i said it has a lot less focus on programming
if you're not capable of developing tools to cater to your needs as a pentester, you're a shit pentester
is why i'm bringing it up
i said not capable?
less focus mean the time spent
not where your skills are distributed
okay, sure, I get that, but it's a skill which is still imho quite necessary
feel free to call me shit pentesters though, i would make the same comment regardless since op was asking about how to get into security
also this isnt an argument anymore since i don't disagree with your point
i should clarify that I didn't mean you you specifically when I said "you're a shit pentester"
it was, like, a general statement
at my team we started not too long ago, many of my development time in the team involves with workflows improvement, issues tracking, exploit repository, etc. instead of like a better nmap or automated scan, or something along those lines
someone else do some of that (not a lot)
but then the whole team benefits from their works
if code needs to be secure, it must not run on the client machine. everything else is just obfuscation and can be reversed by someone dedicated enough.
may I ask what kind of application that is? Such requirements sound weird to me.
I'd make it into a Django/Flask app
Then it can be done via a website, and you don't need to give anyone any code other than an account.
do you ship this exe to multiple users, and don't want their computer's malwares to RE the exe
Heyy guys somebody knows about php security’s?
@amber hemlock Yep
Good luck, as that's not gonna happen.
Developer tools in Chome/Firefox are trivial to use to monitor network traffic
@amber hemlock maybe experiment with an obfuscator https://pypi.org/project/pyarmor/
for traffic you may want to construct your own way of e2ee, so the data leaving your application is encrypted even before entering into the network
keep in mind that even with e2ee, data from server back to your application is much less secret
also its still subject to a debugger attached to your exe, if someone is dedicated enough
also do you use pyinstaller or something? would compile into static C code with cython, and then C compile that be feasible in your case?
@bold grotto Except, you don't want to roll your own encryption because you're bound to get it wrong.
wdym wrong
also i forgot to mention cert-pinning as well (esp for native client and mobile app), in case rogue CA or user add bad cert
@bold grotto I mean.. are you skilled with cryptography?
That's why you don't roll your own.
you re not replacing tls, its for the content
in some occasional scenarios you dont trust the traffic after load balancer or something like that
construct your own way of e2ee sure sounds like you're telling them to roll their own crypto :p
i'm not suggesting abolishing TLS/HTTPS, just adding encryption of the contents, so any point that did not go through TLS get an additional protection
i mean when you're even worried about the user's computer having malwares that dump your memory stack ...
any obfuscation could further help
but i wanna say it'd be less of priority than getting the basics right, like TLS version, force HTTPS, or cert pinning
i mean when you're even worried about the user's computer having malwares that dump your memory stack ...
I mean, in that case, it's game over already
in some occasional scenarios you dont trust the traffic after load balancer or something like that
Im very sorry but if you have "in some occasional scenarios" and "or something like that" in a sentence doubting security of load balancers (which is yes indeed perfectly correct assuming the load balancer does perform TLS termination because then the info is decrypted at the load balancer and it could manipulate it without your knowledge, so its not a "in some occasional situations" but rather with some load balancer implementations) you should probably learn about how exactly TLS/HTTPS can be attacked before
a) doubting its security
b) suggesting an enhancement
and then learn much more about cryptography before you
c) ship your own addition to it
this is not meant as an insult but just what i think
Nix is right, also it's weird when you're saying
i'm not suggesting abolishing TLS/HTTPS, just adding encryption of the contents, so any point that did not go through TLS get an additional protection
when no one brought up HTTPS.
You should be using HSTS to force HTTPS - there should be no option that it can be stripped/removed in a downgrade attack.
you’re right though @orchid notch , ive never actually developed such exploit, or perform it in a practice simulation close enough to a real attack. this is what i think from testing an implementation aiming to mitigate those attacks, and some readings
i suggested those cause i have a vague idea that tls is an absolute protection, especially in some traffic that does not go through encryption.
also i might not be capable to, but op’s team might, and it would help as extra obfuscation over the contents
its also that others can give better insights into them instead of saying things like “seems like odd requirements to me”
can we do some PGP encryptions with python ?
Yes
I tried doing shutdown -i on my mac but it doesn't open the Remote shutdown dialog, does anyone know why or how to do this?
I also tried sudo shutdown -i but it only returned `
usage: shutdown [-] [-h [-u] [-n] | -r [-n] | -s | -k] time [warning-message ...]
`
so I need a some help
@thorn obsidian That's not really security related, though. #ot0-fear-of-python
the description for the channel is hacking and other things
I think it's related
shutdown 10.1.139.29 -h now when I entered this it returned:
shutdown: -h, -r, -s, or -k is required usage: shutdown [-] [-h [-u] [-n] | -r [-n] | -s | -k] time [warning-message ...]
but I already put -h
I'm really confused
How is this related to hacking - security?
cause I'm trying to shutdown another computer remotely from mine
anyways
shutdown -h now 239.255.255.250 I tried doing this
and it shutdown my computer
this isn't even my ip address
what am I doing wrong?
.......you cant shutdown another computer with shutdown(8)
shutdown(8) is meant to give all logged in users on a Unix machine a notice that the machine is going down at timestamp x and then shut it down at that time stamp
some distros will interprert shutdown a little different and just move to some other runlevel like for example openbsd is gonna go from runlevel 2 to 0 which can be useful in some situations
but youre just using shutdown completely wrong and clearly not understanding what youre doing
@thorn obsidian
How to shut down other computers on your network. how to remotely shutdown a computer remotely shutdown computer remote shutdown computer shutdown remote com...
I just used this video
why is it wrong? If so, how do I shut down another computer remotely?
jesus christ
get away from that moron hes trolling lol
and next time you watch a yt video actually learn what the command does before executing it
lol
lol?
how is the upvote ratio so high?
i have no idea
people dont understand
they probably upvote before trying
I am deeply depressed now
do man shutdown and you'll see there is no option for a host
I was like how tf is that related to hacking
I'm not a moron
I know I have to ssh into the computer first
but need help doing that
no im saying the guy who made the video is a moron
oh srry my bad
but yeah I'm getting problems with the ssh
2025-HILLMAN:~ 38167$ SSH 46349@10.1.132.115
Password:
46349@10.1.132.115's password:
Connection closed by 10.1.132.115 port 22
here is what happened
huh? the guy who made the video is trolling? he's not a moron
so why is my command not working
do you know the password for the user
it keeps saying the connection is closed
and yes
I put the password of the user
I own the two computers
uh, yes, might have to be before the other part, I dunno
also why are you using SSH and not ssh? is SSH a custom alias or script you have?
OpenSSH_7.6p1, LibreSSL 2.6.2
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: /etc/ssh/ssh_config line 48: Applying options for *
debug1: Connecting to 10.1.132.115 port 22.
debug1: Connection established.
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_rsa type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_rsa-cert type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_dsa type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_dsa-cert type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_ecdsa type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_ecdsa-cert type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_ed25519 type -1
debug1: key_load_public: No such file or directory
debug1: identity file /Users/38167/.ssh/id_ed25519-cert type -1
debug1: Local version string SSH-2.0-OpenSSH_7.6
debug1: Remote protocol version 2.0, remote software version OpenSSH_7.8
debug1: match: OpenSSH_7.8 pat OpenSSH* compat 0x04000000
debug1: Authenticating to 10.1.132.115:22 as '46349'
debug1: SSH2_MSG_KEXINIT sent
debug1: SSH2_MSG_KEXINIT received
debug1: kex: algorithm: curve25519-sha256
debug1: kex: host key algorithm: ecdsa-sha2-nistp256
debug1: kex: server->client cipher: chacha20-poly1305@openssh.com MAC: <implicit> ```
debug1: kex: client->server cipher: chacha20-poly1305@openssh.com MAC: <implicit> compression: none
debug1: expecting SSH2_MSG_KEX_ECDH_REPLY
debug1: Server host key: ecdsa-sha2-nistp256 SHA256:A3LwG5/s+pc5qY9B8+eHcOZNvVSbh3jg+hOytqEButM
debug1: Host '10.1.132.115' is known and matches the ECDSA host key.
debug1: Found key in /Users/38167/.ssh/known_hosts:1
debug1: rekey after 134217728 blocks
debug1: SSH2_MSG_NEWKEYS sent
debug1: expecting SSH2_MSG_NEWKEYS
debug1: SSH2_MSG_NEWKEYS received
debug1: rekey after 134217728 blocks
debug1: SSH2_MSG_EXT_INFO received
debug1: kex_input_ext_info: server-sig-algs=<ssh-ed25519,ssh-rsa,rsa-sha2-256,rsa-sha2-512,ssh-dss,ecdsa-sha2-nistp256,ecdsa-sha2-nistp384,ecdsa-sha2-nistp521>
debug1: SSH2_MSG_SERVICE_ACCEPT received
debug1: Authentications that can continue: publickey,password,keyboard-interactive
debug1: Next authentication method: publickey```
debug1: Next authentication method: publickey
debug1: Trying private key: /Users/38167/.ssh/id_rsa
debug1: Trying private key: /Users/38167/.ssh/id_dsa
debug1: Trying private key: /Users/38167/.ssh/id_ecdsa
debug1: Trying private key: /Users/38167/.ssh/id_ed25519
debug1: Next authentication method: keyboard-interactive
Password:
@velvet field
SSH 46349@10.1.132.115 -v
``` this is what I typed
and after you type the password?
yours
Password:
debug1: Authentications that can continue: publickey,password,keyboard-interactive
I also tried putting the other computer's password and got Password: Connection closed by 10.1.132.115 port 22
oh, I didn't realize what you were asking, the password you are entering is for the user on the remote host.
I think you might just be typing the password wrong
if you have access to the remote host, you can check the ssh logs to see why it's failing
how?
it depends on what linux you are on, for ubuntu and some others/var/log/auth.log will contain SSH password failures
sudo grep 'Failed to authenticate' /var/log/secure.log on SSH server
security find-generic-password -wa labnol I'm trying to get the wiki password but it said could not be found
am I missing something in the command
ok this is related to the channel, is there any way of shutting down all the users connected to a wifi network?
Unplug it
¯_(ツ)_/¯
whenever I try to ssh into another computer it keeps saying the connection was lost
does anyone know why?
Connection closed by [SERVER_IP]
please help me
Like I've said before, this is not security related, please move your discussion to #unix
it's about hacking tho?
you're debugging an issue with ssh
i fail to see how this is about hacking
@thorn obsidian Are you trying to get access to other computers you should not have access to?
Well, then it doesn't really fall under hacking, but rather, like @thorn obsidian said, it would be a topic for #unix or maybe even just off-topic, since it's probably not Python-related, but just ssh related.
I'm not on unix
Well, it's clearly not related to this channel; we have a #491524019825278977 channel if you're on Windows, but if it's not Python-related, an off-topic channel would probably be better
I'm on mac
out of curiosity of opinion, if i were to hash a master username and password and encrypt accounts in AES-256 for a password manager, who would believe that to be sufficient security?
hashing with blake2b and a salt* i must add. scrypt was not a preferred method for me as the byte response would potentially be too much for a small sqlite storage option.
that depends how you use AES to encrypt the passwords
if you actually want it to be secure and you cant answer yourself the question wether your design is secure you should probably rather use something exisitng
if its okay to be insecure build your own
any vim ninja?
any vim ninja?
so in my one file, txt is like this
127.0.0.1localhost
i want to remove only 127.0.0.1
there are 400 lines
sed 's/127\.0\.0\.1//g' file
@glad cobalt
if that looks good to you, add the -i flag to edit the file in place
but this isn't related to security... please, in the future, ask questions like this in #unix :)
@orchid notch I do believe it to be secure, I was just curious of other's opinions. Currently I just use system user with the master password (asking for it with a prompt) instead of its hexdigest. That, at least in my eyes prevents an attacker from gaining access to any of the accounts without knowing the actual master password and passing that victim's system user name as well or just using brute force
That as I already said depends on how you used AES
There are ways to use AES that would make it trivial to attack even with a 32 character password
do you have an example of a different way to use it then? i just create a new AES object with the previously mentioned items as the key ensuring the key is 32 bytes, and then use that to encrypt and decrypt.
There are many ways to encrypt using a key and the AES algorithm
AES is a cryptographic primitive meant to be used in a certain block cipher operation mode
Plain AES aka AES in ECB mode is easy to crack
While things like AES CBC or if you want an authenticated cipher even AES GCM are way harder
@safe lark You'd use KeePass or the like, rather than rolling your own.
i use the suggested EAX mode. i was aware of the operation modes yes, and certainly wasn't going to use the other modes
and philip i already use something else. it was a project i felt like doing and i did it
i am fully aware how pointless it would be to use mine. it was simply something that occupied me for a day
hence why when i came here i asked out of curiosity for other's opinions on the methods i was using, rather than asking if i should
i must say though, please just discuss with people if their method would be secure and if not how to go about making it better. I'm sure KeePass and the like wasn't made when someone told the developers just to use LastPass or the like
You'd be correct, since KeePass ( according to https://keepass.info at least ) has been around since 2003.
KeePass is an open source password manager. Passwords can be stored in highly-encrypted databases, which can be unlocked with one master password or key file.
?
I'm sure KeePass and the like wasn't made when someone told the developers just to use LastPass or the like
yeah, my point was don't just tell somebody "don't try, you won't do any better. use what's there" i wasn't actually referring to release dates
Also, this is #cybersecurity - and it'd be more secure to not roll your own encryption/ciphers/hashes/etc, and instead go with a system that's been audited. Have you thought of timing attacks to your implementations? Do you have a Phd in math/cryptography? This isn't meant to bring you down or be seen as a personal attack. If it's for learning, by all means, but otherwise I'd suggest using KeePass/Bitwarden instead.
i'd argue that this entire server is based on learning or helping others further themselves. seriously coming at me and saying "if you don't have a PhD don't waste our time lol" is rude and degrading even though i don't care what you think
I think you've misconstrued what I've said.
As long as he is calling into library provided methods like he said he does it's not his fault if he has timing attack vulnerabilities, at least read what he says before criticizing him
thank you Nix, truly
@orchid notch I have read what he's said.
and yes, i use pycryptodome. since the original pycrypto was abandoned (at least to my knowledge) and pycryptodome was a fork that carried it along
according to them directly
PyCryptodome is not a wrapper to a separate C library like OpenSSL. To the largest possible extent, algorithms are implemented in pure Python. Only the pieces that are extremely critical to performance (e.g. block ciphers) are implemented as C extensions.
could you help me understand how essential that is to the security of it as a whole?
@orchid notch Everything from the beginning of the conversation of using sqlite, to your response of if you actually want it to be secure and you cant answer yourself the question wether your design is secure you should probably rather use something exisitng. The condescending at least read what he says before criticizing him isn't appreciated.
But, this is where I leave the conversation regardless. Good luck @safe lark
@thorn obsidian I mean the first statement is basically a mirror of your: use something existing but provided with an actual reason when he should do that and why
is there a way for people to know whether im using recaptcha v2 or v3 on my website?
is there any clue that gives it out?
How to track location using mobile number?
5. Do not provide or request help on projects that may break laws, breach terms of services, be considered malicious/inappropriate or be for graded coursework/exams.
Then you need to contact law enforcement, not us
It may be legal if you track your own phone, but it’s a bit silly
On second though it’s not about who, but about how
Many ways to if they visit places they shouldn't or give it to people they can't trust
I think this is the place to go.
I am trying to import an SSL self signed cert with Python 3.8 to the cert store in order to connect to and pass requests to my device.
I have spent a lot of time and sorted out many issues, just stuck getting cert_human working
If anyone knows how to import, add and verify a cert with python, that would be very helpful
I had a heck of a time getting to the point I am, so any help would be appreciated.
This is for a non profit in Uganda ^_^
@thorn obsidian You'll need to present some code, as we have no idea what cert_human is
I think I made a mistake @thorn obsidian. I'm going to try again and then post something.
Let's try a help channel?
Sure
!free
Doesn't look like any channels are available right now. You're welcome to check for yourself to be sure. If all channels are truly busy, please be patient as one will likely be available soon.
Ouch..
Well, in a bit, then.
Not urgent. Neet to eat etc
Sure, no problem 👍
Thank you very much. First python day was fun
Looks like they wanna go unsecured, so no more issues with that in python. Moving on to the payload and log in bits. I'll be back if still struggling, (read as, "I'll be back.").
...
usrnm = input()
passwd = getpass()
usrnm_hash = sha512(usrnm)
passwd_hash = sha512(passwd)
hs = usrnm_hash + passwd_hash
f = Fernet(base64.urlsafe_b64encode(hashlib.blake2s(hs.encode()).digest()))
...
Is this good encryption key to encrypt personal files ?
so you're getting the sha512 of your username and your password, concatenating the hashes, hashing those with BLAKE, encoding as base64, and finally using that as a Fernet key...?
dunno dude, seems like a lot of steps to achieve.. not a lot
and the definition of "good encryption" depends entirely on your threat model
@thorn obsidian If you're encrypting personal files, VeraCrypt or LUKS are the go-tos.
These patterns are used to secure banknotes ~ money
I have also seen them in exam booklets
@thorn obsidian How is this related to the channel?
@thorn obsidian @thorn obsidian thank you guys
Up to 10 percent of the Internet is encrypted by the groovy, random motion of this wall of lava lamps. Read more about these lava lamps here: https://www.atl...
Last time i've read an article about this wall, they were saying they didn't use it really for production purposes
Ah seems it's really used https://www.cloudflare.com/learning/ssl/lava-lamp-encryption/
@thorn obsidian what kind of programs
like open source projects?
then go to #303934982764625920
note that it has to be mostly Python
and open source
thank you
anybody knows about MyLockbox software?
in a client/server system, where encryption is done by a key exchange handshake, which party should be expected to initiate the handshake? the client or the server?
the client
depends on the implementation, but 90% of the times it's a Client Hello first
Can I get some help creating rainbow tables?
Im following this guide http://kestas.kuliukas.com/RainbowTables/
Problem is I keep having false positives on identifying wether a hash in a given chain
As you could use those go be a very very bad person and crack other people's passwords
I don't think you're gonna be help
@orchid notch You can do evil things with a lot of things, but you can only prevent them if you know how they work
As a Sysadmin, I regularly attack our network maliciously
Yeah and it's cool that you do that
However the server rules state that we shall not help with such behaviour because we can never know the intentions behind the person asking the question
....nice, posted a question here and instantly forgot about it lol
late, but thanks nix and xx \\🐱/
is there a way of shutting down computers around me without sshing?
5. Do not provide or request help on projects that may break laws, breach terms of services, be considered malicious/inappropriate or be for graded coursework/exams.
Aside from obfuscation, since Python is generally interpreted, is there other ways secure your source code? Like non-"security through obscurity" ways? Or is it just "if you're that worried pick a different language" situation.
How would I protect my .exe file running on compiled .py code from being cracked and reverse-engineered?
You wouldn't
U can't hah
It's a bad practice to carry a master password as private method initialized in a class?
I don't understand the question at all, and yet I'm pretty sure the answer is "yes"
My situation is this
The user gives me a master psw
with that he unlocks all his passwords and can save new ones
But I don't want him to write it for every action
So I thought that after he writes it one time, and it's checked, I should temporarily store it in a variable of the class
my guess is: if you're writing code that accepts a password from a user, you're already doing it wrong
security/crypto stuff is notoriously hard to get right; the wise engineer uses ready-made stuff as much as possible
can I show you what I did?
sure
Many things are in spanish
but let me explain it to you
ah! well a password manager is the rare case where it does make sense to ask for a password from the user 🙂
I assume you're doing this mostly for fun
To learn as much as I can, and to use it so I don't have mines in a txt
for heaven's sake, use a real password manager, not one you wrote yourself
I don't expect it to be really secure, but to be better than a txt
if you use a Mac, you have one already, called "Keychain access"
I don't use Mac
ok, if your expectations are low, then I guess you're fine
yes they are low, I wan't it more like to organize passwords for me base on categories and pages
I do not intend to read your code carefully; that'd take too much effort. But if you can post maybe 10 lines of code that demonstrates your question, I'll look at it
yes I was going to do that
To save a password, I ask for a master password, wich I use for creating a key
it's odd, but I can't think of a good open source password manager
using this:
It the same if it is not open source, people who whant to hack it can open files and see what they like
def generar_key(contraseña):
contraseña = contraseña.encode() # Convert to type bytes
salt = b"\xb8M\xce\x94\xbd\xfa\x0f\x1e\x04\r\x83\xea\xa1\xbbD'" # CHANGE THIS - recommend using a key from os.urandom(16), must be of type bytes
kdf = PBKDF2HMAC(
algorithm=hashes.SHA256(),
length=32,
salt=salt,
iterations=100000,
backend=default_backend()
)
key = base64.urlsafe_b64encode(kdf.derive(contraseña)) # Can only use kdf once
return key```
Then with that key, I encrypt a little message and save it to a db
how can I do to not ask the user to enter the master password every time he want's to do something in the same session
ah
sure, save it in memory somewhere
there are reasons that's insecure, but I doubt you care about that
an attribute on an instance is as good as any place, I guess
but if I store it as a variable of the class, then anyone creates a new file, imports it and then he can send an email with it
I want to do it as much secure as my knowledge admits
I didn't understand your first sentence there
I mean, that if in __init__ I have self.master = "" and later I assing the real master, anyone could get it doing that
?
certainly you're not going to enter the master password into your source code
I wasn't suggesting that
no no no me neither
I mean, I create a blank variable in initialize, and later when the user enters the master password, I store the value in that variable if it is correct, and it will get ereased when the program stops
yes, of course it will get erased when the program stops
if you want to preserve the password after the program stops, then you're screwed.
There's no way to do that that's any more secure than storing the list of other passwords
I don't wan't to store it after the program is stopped
well yes
then I don't understand your problem
I mean, that the problem if doing self.master_password = master_password is that anyone could create a new file, import MyClass, and do print(MyClass.master_password)
it sounds like you're not in control of the computer you're using
what kind of computer is this, that has random strangers on it, able to access your running program's memory?
I strongly suspect you're worried about something that isn't a real problem
Butttttttttt 😦
I also suspect you cannot demonstrate the problem you're actually worried about
write 10 lines that demonstrates the problem. Then we'll talk.
heh
in that new file i will do another instance of that class
well it's still insecure if you execute it from that new file
or if you modify the main
but
that would be
a) I don't understand what you mean; b) I don't believe you anyway 🙂
I won't discourage you from exploring security programming; but I will suggest you be very humble
i know I don't expect of this to be secure as lastPass or any real program like that, but I want to do the best I can, that inclues to fix any possible problem
sure
but you won't be able to fix every problem you think of.
you'll think of 10 problems. Five won't be worth worrying about. Three you'll fix. The other two you won't know how to fix.
Then there will be 20 problems you didn't even think of.
I know, but I won't stop till those two are fixed or that's what I want to think
take a look at https://docs.python-guide.org/scenarios/crypto/ ... it recommends a couple of libraries that will do most of the heavy lifting for you
thank you very much
I personally would avoid the GPG library, since GPG terrifies me
sorry if this isnt the place, but
[6:50 PM] offby1: it's odd, but I can't think of a good open source password manager
by this do you mean "good" as in "secure"? or just in terms of a nice interface?
both
terrible interface == insecure, as far as I'm concerned
since I'll probably use it wrong
the Keychain thing built into Macs isn't a really good interface, but it's good enough
hm i guess thats pretty fair lol
when i was choosing i couldnt find anything bad about kpxc (what i went with) other than being a bit clunk
in practice I use LastPass, which I kind of hate, but it does work, and I'm in too deep to switch 😐
never heard of KPXC, but I'd tried keypass ages ago and found it too clunky to use
huh, I can install it via homebrew, which is very encouraging
i like it a lot, especially that you can set a global keybind, which then reads the title of the active window and opens a popup to pick a password to autotype
slightly creepy but probably handy 🙂
with a browser plugin it can also read current web address but thats the line for me personally lol
if you're serious, you have to study the source code to kpxc and the browser plugin
and hell, the browser itself 😦
oh theres no way im ever touching that plugin
didnt get anywhere trying to read kpxc but as far as i can tell from behavior, its only smart enough for window titles; if you press the bind on a terminal window with a title like a website, it offers those passwords
that UI really looks like amateur hour 😐
I admit I'm a bigot, and that I judge apps by their UI polish.
but if they can't afford to make a good initial impression, I fear they also can't afford to get the code right
pay no attention to the default icons with white feathering around the edges in dark mode 
couldn't help but 😐
@olive lark do you want to see my UI?
but if you don't like keepassxc ui I dont think you will like mine
no thanks; I probably won't like it, but won't be able to offer any helpful suggestions, either
I just complain a lot
@olive lark keepassxc is widely acknowledged as the best password manager, i'm sorry the UI has turned you off but as time goes on, you'll come to realise that the better a programmer is, the worse their graphics sense is, and we're big endorsers of "simplicity is key"
LastPass I would stay far away from, personally
I do think programmers have a tendency to overcomplicate GUIs so depending on your interpretation, "simplicity is key" doesn't apply.
Or maybe it's just a lack of thought that naturally makes it look complicated with clutter
I dunno
cluttered, yeah, maybe
any rainbow table experts out there
im trying to find a good reduction function for my tables
Hello!
Hey there
wat
hello
if you want to get the passwords for all stored wifi's, then use this
from subprocess import check_output
from re import split
output1 = [split(r' +: +', i.strip())[1] for i in check_output("netsh wlan show profiles", shell=False).decode("utf-8").split("\n") if ": " in i]
print("Wifi Passwords:", end="\n\n")
for wifiNames in output1:
output2 = [split(r' +: +', i.strip())[1] for i in check_output("netsh wlan show profiles \"" + wifiNames + "\" key=clear", shell=False).decode("utf-8").split("\n") if (": " in i and split(r' +: +', i.strip())[0] in ("Key Index", "Security key", "Key Content"))]
print(f"{wifiNames:<40}:", end="")
if(output2[0] == "Present"):
print(f"{output2[1]}")
else:
print()
__import__("os").system("pause")
For windows
uhm...
Aloha
vamos
out of interest, would you be able to plug in a camera and a mic to a raspberry pi and from there be able to use it in a python module, or is there any config first beforehand?
I don't think this is related to #cybersecurity
well i want a secure server with a facial recognition and voice recognition
think its more #algos-and-data-structs or #microcontrollers maybe
np 😄
does anyone know how android encrypts disks and sdcards and how i would decrypt it if my device breaks?
not me
@tepid rover https://source.android.com/security/encryption/
Note: Full-disk encryption is not allowed on new devices running Android 10 and higher. For new devices, use file-based encryption.
i just sucsesfully printed hello world and now i want to hack nasa
This is a little off topic but could I potentially infect my machine by webscraping with free proxies from the internet? I’m only using requests and no browser, but I haven’t been able to find an answer regarding my specific usecase. Thanks
As long as whatever you're using the scrape the websites is not evaluating JavaScript + as long as whatever tool you are using to evaluate your results is not vulnerable to some parsing flaws (both highly unlikely) you're gonna be good
Does anyone know of any good sources to help protect the scripts source code etc from 'sneaky eyes' / thiefs.. E.G I plan to compile all my script into a .EXE file e.g py2exe but was hoping to protect it as best as possible from reverse engineers?
@median heath You wouldn't
Why so?
You'd put in considerably more effort than is worth it, and it's trivial to decompile.
@median heath use Nuitka, compiles Python code into C's executable, which is much harder to crack
Also, anyone knows a good way to create a whitelist/authentication system for the program running on Python?
I am using Postgresql database on the server to store all user credentials
but I am not sure where to start and how to exactly organize the communication between the database on my server with all the credentials, and the user running the program.
@plucky geode I'll take a look in a second.. 🙂
How many end-users are you going to have?
I'm using an external company which uses a licensing system upon opening to validate if the users license is valid and then it can be limited to one PC/multiple etc.
It's only a few lines on code
of*
Take a peek, very user friendly. There is some cost, dependant on user-amount. I think it's reasonable
Yeah, looks sleek, but I was thinking about building the system myself, so that I could customize the code if needed
Same!, I've been looking also, would much rather manage myself. But for now it'll do
Plus don't really want to get into monthly subscription thingy
I didn't have much luck when trying to find something that;ll work for me and that I could manage of my own DB etc
But i'm a rookie, so eh
Hmm, I was thinking about creating separate users for the database. Basically, 1 user = 1 license.
Therefore, people will connect to the database automatically via their DB username, and program will authenticate them.
It works with the Postgresql, but I don't think such thing exists for the SQLite or MySQL?
In theory, sounds easy, I guess the security aspect is the issue
Yeah, maybe someone with more experience will share his thoughts
@plucky geode I hope you're not letting users operate on the db directly... that's a recipe for disaster
I recommend setting up an API. Users are handed API keys which can be used to start sessions
I'd make it so you can't have more than one session active per API key
Then, the session identifier/whatever can be used to talk to the API, which operates on the database
This way you can easily restrict who has access to what parts of the database without having to mess around with DB users
@thorn obsidian Thanks for the help, I think that Postgres has native users features, and I can restrict access to what parts of DB users have, as well as what they can only read or write to. Natually, users won't know the credentials to access the DB on their own. I will make program communicate with DB instead of actual users.
API and Sessions sounds like a good idea, and I might implement it in the future, but I need a rough and fast way for now XP
if you plan to use postgres users for permissions then this is also an interesting topic: https://www.postgresql.org/docs/12/ddl-rowsecurity.html
I've been working on an encryption/cipher technology for a while now and I've just merged a PR to implement AES encryption. Could somebody take a look maybe and let me know if they see any glaring issues in it? It's not a lot of code and just 1 file. https://github.com/M4cs/Pixcryption
the first issue is: you're rolling your own encryption. If you're just doing it for fun and learning, great; but please don't use this on anyone else's data
@olive lark why is rolling his own encryption a problem? i'm just curious.
typically people who know a lot about crypto warn newbies to be careful, because it's much harder than it looks
https://security.stackexchange.com/questions/18197/why-shouldnt-we-roll-our-own <-- second google hit, e.g.
I'm not rolling my own encryption. It's using AES for encryption. I'm rolling my own cipher layer on top of AES
shrug
@olive lark has a point. I've been around encryption/ciphers for 15~ years and know AES a little bit, and I'd still recommend not rolling your own.
@chilly elk there is using AES and there is using AES properly
just because you use AES does not make your system secure
in fact you can create a false sense of security for your users if you say "we're using military grade encryption" just because youre doing AES
AES has to be used with a proper implementation thats invulnerable to side channel and timing attacks
in a proper mode
with proper parameters
and exactly at that point it becomes secure
Ok I'm under 18 rn
so I want to do cyber security
I'm confused with CEH and OSCP
Which should I do
Is CEH good for getting jobs
And others like CompTIA are required or no?
OSCP holds far more value than CEH
ik that but is CEH good enough to get a decent job in Cyber Security?
meh, it's less about your certifications and more your portfolio
like things you've done, projects etc
I'm employed in infosec atm and I have 0 certs :) (apart from CCNA, but it's not really related to my field)
friend of mine got a low level job at a security company. started by just doing secure erase or following scripts to compile diagnostics for smarter people.
worked his way up the chain
yea that's the way to do it
if you prove you know what you're doing, you'll have no issues with that
yeah, same here
company's paying for me to get my certs
which is.. really cool, I guess
yeah
con watching is probably how i got into reading security stuff in the first place
Hi guys, just a completely out of the blue question here, not relating anything to do with python (maybe some to do with security, not sure where to ask this). Has anyone purchased an e-gift card before? And if you have, do you know if it contains like a Card Number or a Card PIN?
if you mean a store specific card like a steam card, it will have a code on the back which you enter in usually
https://eng.getwisdom.io/hacking-github-with-unicode-dotless-i/
As it turns out, .lower(), .upper() or .casefold() on unicode in regards to e-mails is bad. 😄
o.O

ouch about sending to the provided email, because it clashes, that's very easy to overlook when coding
Hi, question. Does anyone here have any experience in digital forensics? If so I could really use some help
I'm trying to mount a .img file that is supposed to be a forensic copy of a hard drive. But neither Kali nor Windows will let me mount it. Says it is corrupt
You don't mount direct images of disks, you use forensics tools to look through them
can use some tools like volatility
hello
hello
hello
well we're about to have mac and cheese, and I'm sorry to say we didn't invite you over
guys want inputs as in material, things to learn.... on cyber security threat analysis-- like where to start.....or any suggestions
was looking into materials from https://medium.com/cyberdefenders/python-for-cybersecurity-lesson-4-network-traffic-analysis-with-python-6321f4c9d3f7
but this does not share any insights on threat analysis.....
do share your inputs
🙂
My one suggestion is be carful with things you download: noobs in this realm are often targeted for trojans, etc. Good luck o7
hey, how best to start writing a security system in Python?
"security system"?
my question exactly
@thorn obsidian What's a "security system"?
@thorn obsidian it has resolved itself
Why my recommendation in IG is too relevant for me...but I didn’t even search anything related to it
Does google collect all cookies and distribute it to other companies?
yes
or ads on instagram are powered by google ads
beacuse it's literally what i searched few minutes before
instagram is owned by facebook
does facebook purchase google data? who knows.
but they have enough of their own ad data gathering techniques
facebook are insanely good at targeting ads
doesn't matter if you have a facebook account or not.
any site with a "share to facebook" button is probably supplying data to facebook about your browsing habits
i mean they know your IP, your browser, your browser version, likely operating system, etc, etc
it's enough data to build a semi-unique fingerprint
so i got a voicemail transcript this morning of what sounded like two kids using a kind of dialer and putting in numbers
i have the whole thing recorded
they put in like 35 different numbers in a specific sequence
i have google call screening so if i had been awake they wouild have been sent to the bot
now i have the whole transcript and the entire convo recorded
what do i do with it
i feel like they were trying some shenanigans to get past my 2fa that i have on my gmail
i doubt they did it
people can control your bot by placing a phone call?
no i think they were trying to bypass 2fa
well I wouldn't worry about that
Vocaroo is a quick and easy way to share voice messages over the interwebs.
if you wanna hear it
google should know what they're doing as far as security goes
theres one idiot using the phone and the other sounds like hes in a vc telling him what to do
shrug
like i legit have no idea what they were trying to do
and it looks like they used a real phone to do it?? cause im able to call and text it
and it shows up as being a real number
so theyre getting call bombed rn
so he responded to my text
i spent like an hour trying to get him to tell me what he was doing
i called his local police and like a minute later the kid actually called ME and i TALKED to him. he just said the same bullshit he was saying in the sms
whats a cyber project
Implement TOTP from scratch?
What's TOTP @daring sedge ?
The Time-based One-Time Password algorithm (TOTP) is an extension of the HMAC-based One-time Password algorithm (HOTP) generating a one-time password by instead taking uniqueness from the current time. It has been adopted as Internet Engineering Task Force standard RFC 6238, ...
Oh
Well that sounds cool
thanks 🙂
also does anybody know a library that messes with linux's login?
there is no such thing as "the linux api"
if you wanna login as a user just use the login command line tool imo
@icy saffron Are you talking about hooking into Linux authentication via PAM?
no.
You can just ask your question. The server has over 30k people. If someone reads your question and knows a thing, they might respond.
😩 😩
any of u ever used JonDo proxy?? How is it any comments?
does it like change server/ip after some interval of time ??
it looks like a mostly abandoned alternative to tor that has 15 downloads a week and is written in java
It looks like you define the hops from a list of publically accessible nodes. If your aim is to layer proxies, you are probably looking for proxychains
@rocky horizon
if your aim is 'anonymity, good enough unless your adversary is a nation state', you're looking for tor
I've heard of JonDo proxy, but never used it. I didn't know it was still a thing, and wouldn't use it now.
@tribal stag is right. You'd want Tor unless you've pissed off the NSA, which in that case gg.
5. Do not provide or request help on projects that may break laws, breach terms of services, be considered malicious/inappropriate or be for graded coursework/exams.
<@&267629731250176001>
!warn 311919809895858177 "That discussion is not appropriate for this server"
:incoming_envelope: :ok_hand: applied warning to @rocky horizon.
Umm its my local server so🙄🙄🙄🙄
Does not change the nature of it.
I just asked if jondoe will do my shit
Hey, between a ssh connexion, and a VNC connexion, which one is the safest?
I guess it's best to use ssh + X11
weeeeeeeeeellll
since I'm not a cmd killer, I think I would need sometimes a interface
there are some security concerns with a certain mode of X forwarding over ssh
as is mentioned in the OpenSSH client manpage
-X Enables X11 forwarding. This can also be specified on a per-host basis in a configuration file.
X11 forwarding should be enabled with caution. Users with the ability to bypass file permissions on the remote host (for the user's X authorization data‐
base) can access the local X11 display through the forwarded connection. An attacker may then be able to perform activities such as keystroke monitoring.
For this reason, X11 forwarding is subjected to X11 SECURITY extension restrictions by default. Please refer to the ssh -Y option and the
ForwardX11Trusted directive in ssh_config(5) for more information.
-Y Enables trusted X11 forwarding. Trusted X11 forwardings are not subjected to the X11 SECURITY extension controls.
An attacker may then be able to perform activities such as keystroke monitoring On server-side, or from the controller computer?
i think what's meant is that a user that has compromised your server could get access to your local X11 server as the X forwarding allows the remote connection to act as an X client. And as X is a protocol not caring much about security the principle is that every keystroke is sent to every x client
which could in theory allow an attacker on the server to keylog you via this connection
at least thats what i think they mean
hum ok. I'll stick with ssh command line for now then.
X over ssh is painfully slow, whenever I've tried it
i remember a colleague of mine having a nice program that helps with that
but i dunno remember the name
yeah at work some people use some super-complex thing that claims to make it go faster, but it too sucks. I just use console-mode only for my remote stuff
https://www.nomachine.com/ is what they use. I found it not worth the effort.
i mean im all for console there was justone occasion he had to use it
and it wasnt really complex at all, just some command line tool
if your network connection is snappy, your "screen" is small, and your personality is patient, plain X over SSH will be fine 🙂
what is the best way to Encrypt Android phone ?
Hey. What would be the best way to go about securing some on-disk json data with a password? From what I can tell, a key derivation algorithm like argon2 to generate the key for an algorithm like chacha20 or an AES mode should be good? I'm a little hazy on the IV component though. Is it alright to be saving that as plaintext along with the encrypted bytes to a file?
This is very small amounts of data so performance won't be an issue
How are you using the information?
i.e., are you storing it in a DB? Is it just random JSON data stored on-disk? If it's the latter, VeraCrypt/LUKS would work. We need more information.
Data that needs to be store on disk of the users of the application I'm doing
And transportable as a file
who are you protecting the data from?
Malicious entities/software
If a computer is compromised the data is not
I guess it is possible for keyboard inputs to be captured
they surely can
But that's unimportant
including when the guy is typing in his password
the only case I can think of, where on-disk encyption makes sense, is: someone steals the computer and takes the disk out, and wants to get the information off of it.
but if there's evil software running on the computer, game over; we all lose.
Not necessarily
yeah yeah if the evil software is running under some user account that doesn't have privilege to read your file, sure OK fine
From a software perspective, yes
Not from a human one
Stealing a folder from a hardcoded path is much easier than setting up a keylogger and sifting through possibly weeks of data for 7 keystrokes
knock yourself out
The way I see it, there's no reason not to encrypt this data
It decreases the attack surface and adds a time and effort barrier
This application would be password protected anyhow so the user isn't suffering
@sly sonnet is this software running on end-users machines or ones you control?
End user machines
there's a few ways you could go about this
depending on who you're protecting against
if it's other software with filesystem access to the user's data at rest (ie when your software is not running), then:
generate N random bytes to use as a key. Store them in the system keyring. Then access that key from the system keyring on startup and use a good symmetric cipher to decrypt the user's data.
That sounds quite good but wouldn't that hurt portability?
yes
I would rather the files be self contained so they could be transported
system keyring support is better on Windows and Mac than on Linux afaik
oh hm
Yeah you could do that as an export/import system
backup/restore
where it reencrypts with a random key and gives the key to the user
That does sound like a good solution but this would ultimately needs to support all three platforms as well though
I don't have experience working with OS keyrings. it's something you could look into.
keep in mind that portability often means special casing all platforms 🤷
Maybe there's a library that abstracts it
How come?
If they all do the same thing differently then you don't have much of a choice
Something potentially easier and more portable, but trickier in other ways, is to do the same thing, but have the user store the symmetric key, and using argon2 or some other key-stretching algorithm.
Already using argon2
yes
anyway the gist is the same
use argon2 to derive a symmetric key
use a symmetric cipher to encrypt the data at rest
Yep doing that
And the nonce component?
Just store that in plaintext along with the encrypted data on disk yes?
I'm a bit over my head here but I don't think you need a stream cipher here
lemme read a bit more
Oh I'm just as hopeless don't worry
from what I can tell you'd be better off with a block cipher such as AES
but I don't think AES is the right choice here either…? since it's asymmetric you need a public and private key
I don't believe AES is asymmetric
AES is absolutely not asymmetric
With everything I read I get more and more confused as to watch I should use
As it does not matter whether the same plaintext always produces the same ciphertext, AES mode SIV seems to be good for my purpose
I don't understand what you're making, @sly sonnet
I was under the impression this was on a server or some such. What's the JSON and what's it doing on an end-user's machine? Why does it need protected?
okay so, I have a key based auth for a flask app, connects to a server and verifies key and then renders the next page. ill mention Ive only done the backend (py) and we hired someone else to do the front end.theres a problem with the way it is right now tho. when youre at the login page, you can manually go to the url of one of yhe other pages, bypassing the auth. whats something I can do to avoid this? should it be backend or front end? ill also mention that the urls of the other pages are in the html
first of all authentication should obviously be handled in the backend
and
just have a cookie that shows the user is logged in or not and if he isnt dont allow the access i guess
yeah the auth is backend, which was done by me. and okay so youre saying that my problem is on the front end?
well i dont understand how your issue is even possible if the key gets verified
it gets verified, then renders home.html. but the way the html is set up, I guess it has like each page url hardcoded in, so like if you just put in a url, it jumps auth and everything is fully functioning. granted this app isnt gonna be reaching a bunch of people, and doesnt hold sensitive information, this is more or less practice with larger projects, and me working on security. im not sure what to do at this point tho, as everything ive done to try to eliminate functionality just returns internal server errors
aka breaks the entire thing
that just bullshit
that design
if you for example have a profile
you should have a profile.html template in your backend written in jinja2 for flask
that gets filled the information from the account thats associated with the key
so which part is bullshit? lol as I said, I didnt write the html/css/js so if thats it, I have to talk to the person who did. if its something else, thats on me
okay but as I said, first time doing this shit, you’re acting like I do this for a living
people gotta start somewhere g
nah if i was acting like you were doing this for a living i wouldve told you to stop doing^^
when someone asks for advice, they need help obviously. the proper response isnt to call everything bullshit lol
well im just saying the entire design is flawed
because youre asking for what part is flawed
k thanks
so
usually when designing a website that has authentication, or any user stuff in general with for example a profiles page
you have a template written in https://jinja.palletsprojects.com/en/2.10.x/ when youre using Flask. Then once you get a request you verify the API token and find the associated user in your DB, fill out the flask template with the info you got and send it back. so for example youd have the URL
which gets a request with a JSON content like
{
"token": "uuid4 thingy or so",
}
then you check in your DB if "token" is
- a valid token
- for what accoutn its valid
Then fetch all profile info from that account, fill it into the template which will generate you some html
and then return that to the web browser
okay thank u
@thorn obsidian Sorry about any confusion. This is an application run by users which has sensitive information personal to the user that needs to be stored on disk and password protected
Im trying to write a telnet honeypot to for IoT goal is to report nets but need help with login system as it needs all login to be true
I suggested that @thorn obsidian should look into telnetlib but I have no actual experience with this, so I sent him here to get help. you guys are so talented, maybe anyone have an idea?
yeah i know nothing abt telnetlib im using sockets trying diffrent things atm