unix

[wet folio]

They actually used this cmd to run the fuzzer for infinite time period, I was trying doing something that it can read multiple image file which run 1 hour each in their tool and then save the report

[shy yoke]

Hey @wet folio!

You either uploaded a .txt file or entered a message that was too long. Please use our paste bin instead.

[wet folio]

@trail sapphire https://paste.pythondiscord.com/yakizuyota Here you can see the whole code. i just making some changes some lines from their original implementation. They used cmd line to run their tool. While I create 5 image file, and wanna run each image file for 1 hour in their tool. Do you have any advices of the modification. I even add -s sleep argparse for that in the code

Just considering f' -e "samples/oracle/{args.fstype}-{i}.image" -y seed' on this path, I have those images located, does this for loop will work to perform its task:)

[proud nova]

I brought my dual-boot Windows/NixOS machine over to my brother's place but forgot to install wpa-supplicant beforehand... so now have to figure out how to bootstrap that haha. I found an SD card and SD card reader so I think I'm going to flash something bootable with Wifi onto that, then use that to chroot into my Linux machine and install the wifi stuff

[trail sapphire]

can't you just download what you need using windows or another computer, put it on the sd card or other flash drive and boot into linux and install it?

[fallow tusk]

Or just mount the Win drive in Linux after dual booting

[proud nova]

I don't know exactly how to download all the Nix packages I need from Windows

I thought about looking at the packages that failed to download to the nix cache, and thought about just downloading those on Windows and adding them to the nix cache manually once I rebooted, but I don't feel confident I can do that safely without breaking some nix internals

I'm also not sure if the list of packages from the error message was exhaustive -- it might have failed after some # of failed package installs and not listed all the URLs it tried to grab

[thin robin]

Hello

Im installing fedora and now i get an error when installing

this program has encourted an unknown error

https://zerobin.net/?6a8578f272ad876b#E9ZPXgzEH8QGIPmHIv4uMvKJayAr2R1YbP3kdfw+cug=

this is the full debug message

it has python 3.10 for some reason???

in the fedora installer debug message

[formal schooner]

a lot of fedora system tools are written in python

this seems like either a problem with your setup or with the fedora installer

[glad stirrup]

infact, most distros use python

[formal schooner]

as far as i can tell, it's one of the more common system scripting languages in distros

not many ruby or perl scripts

a typical linux distro consists largely of c, bash, and python

[sleek swan]

What linux distro should I start with?

[fickle granite]

ubuntu

next question, please 🙂

[sleek swan]

Why is ubuntu preferred?

[fickle granite]

up to date, works with lots of hardware, reasonably easy to use

also: it's already popular. The more popular a distro is, the better it is, oddly enough

(more people contribute bug reports &c)

[sleek swan]

Ah makes sense

Ty

[fickle granite]

one warning: if you install python, you need to separately install pip and venv. It's weird, I don't know why

sudo apt install python python3-pip python3-venv should get you started

[trail sapphire]

Fedora is also good

[fickle granite]

they still make Fedora?

[sleek swan]

Alright ill keep it in mind. First I gotta figure out how to start the vm

[fickle granite]

what (base) OS are you using?

if it'a Mac, I recommend multipass -- it's Ubuntu-only, but it's so convenient that I think it's worth the restriction

(I think it only works on windows if you're using a fancy-enough version of Windows and have virtualbox installed)

[sleek swan]

Im on windows using virtualbox

Thst should be fine?

[trail sapphire]

yes, they release a new version about every 6 months, it's also the preferd distro of Linus Torvalds him self

[fickle granite]

that counts for something 🙂 Had I known it was still current, it would have been my second choice

[trail sapphire]

the downside is that you have to upgrade at least every other release as there is no long term support versions and they release about every 6 months and only keep support for the current the the last version with an extra month overlap with one more version back after a release of a new version, so you can mostly only count on about 13 mounts of updates
so you are more or less forced to keep up, rhe flip side of that is that they are very up to date with the software they ship with every release

[fickle granite]

lotsa tradeoffs

[vital bay]

that's what turned me away from fedora. since i use some software that isn't open, the update from one version to the next isn't always clean or possible (neither with ubuntu nor fedora), so i just go with whatever has the longest support

[fickle granite]

I'd hope that nonfree software has a recommended distro

[trail sapphire]

if you like fedora but want lomg support you turn to redhat enterprise linux if you want a support contract or rocky linux (what centos used to be) if you want it free but without a support contract, those generally come with 7 years of support from the time a new major release is released and can be good for running nonfree software on

[shrewd stratus]

there was a Fedora derivative called Korora that had nonfree repositories included by default

but it died

[vital bay]

sure, but then you have to update the kernel by hand and make a handful of changes to the available repos to even have drivers for all your hardware

i wouldn't try to set up a server distro for laptop use

[fickle granite]

I meant: if I were using a specific bit of nonfree software that required Linux, I'd hope that that nonfree software would have some docs that tell me which distro it works best on

[vital bay]

most of the ones i use offer rpms and deb, so they kinda expect you're going with something debian or redhat-based. no comment beyond that though

[trail sapphire]

agreed, those are mostly for servers or maybe even workstations but not very suitable for laptops

[trail sapphire]

totally agree with @vital bay, it's a bit self documenting if there is offical packages for distributions, but if there is only binaries or tar archives it's not always so clear

[sleek swan]

Uh, does anyone know why the ubuntu desktop doesnt scale with the virtualbox window and why (-_) and (?+) have switched places with eachother on my keyboard?

[vital bay]

check that the keyboard lang in the vm is the same is the host's. i think you also have to change the VM video drivers if you want it to rescale on the fly

[sleek swan]

I fixed the language issue but have no clue as to how to solve the other one

It even cuts itself off by the sides

[vital bay]

what if you full screen? at any rate, there should be a vm option to automatically rescale. i don't recall where it is, though

[sleek swan]

Still cuts itself off

[shrewd stratus]

in the View menu IIRC

[sleek swan]

I can get it to fullscreen now but its a bit blurry and still cuts itself off like in the picture above

[serene rain]

is your screen resolution 1920x1080 ? Then try this: https://computernewb.com/wiki/How_to_enable_the_1920x1080_resolution_in_Ubuntu_on_VMWare (its the same in virtual box)

[sleek swan]

Ive got something thats half working rn, im afraid of screwing it up so ima keep it as is I think

[west umbra]

dont worry it is just a vm

if u wont experiment u will never learn

[mortal knoll]

Im currently compiling KDE rn but my power might go out. If I were to save my machine state and turn off my PC, will it remain even after I turn it back on? As in, will it continue to finish compiling where I left?

Too late

[main olive]

Unable to write file '/home/k1/project/example.py' (Unknown (FileSystemError): Error: EROFS: read-only file system, open '/home/k1/project/example.py')

i'm using debian 10

this happens when i try to modify files and try to install packages using pip, how can i fix this?

[digital prawn]

open terminal type

echo "$USER"

@main olive

[main olive]

i figured out a solution, thanks though!

[gleaming wadi]

When assigning a boolean value to a variable in bash/zsh, is it more uniform to write it as VAR=true or VAR=1?

i'm very familiar with scripting in both, but i've never come to a conclusion on this

[proud nova]

In bash, I believe VAR=true actually puts the string value "true" in VAR

I think that the true and false commands actually just set their respective exit code (true = 0, false = nonzero):

$ true
$ echo $?
0
$ false
$ echo $?
1

[fickle granite]

depends on who is reading that variable, and what they expect

[gleaming wadi]

it will just be me, or whoever else looks at my config file repo

[fickle granite]

so you're asking about convention ... that, I don't really know. My hunch is: if it's undefined, or the empty string, then it's false-y; otherwise it's true-ish

[proud nova]

If it were me, I would probably use the same conventions as the shell itself. Idk if this is idiomatic but the idea might be something like:

$ export VAR=$(true; echo $?)
$ echo $VAR

[gleaming wadi]

i wonder where the POSIX gives an answer to this

[gleaming wadi]

perhaps, i'll just stick to 1's and 0's for clarity

[proud nova]

Yeah but just keep in mind that 0 actually means true in bash

And maybe zsh too, idk

[fickle granite]

I wouldn't worry about it.

[fickle granite]

as an exit status, yes; but not as a string value with no context

[proud nova]

right

[gleaming wadi]

yeah i was gonna say this too

i'm more familiar with zsh than bash since i originally used Macs, and they used zsh

and in zsh you can use 1/0 or true/false and there is (virtually) no difference

thanks for the pointers btw

[fickle granite]

I suspect zsh and bash are more similar than different

I only recently started using zsh (interactively, not for scripting) after decades of using bash; and I'm surprised at how few adjustments I've had to make

[gleaming wadi]

zsh is essentially bash, just with superior configuration capabilities and loads more packages/resources made for it

[gleaming wadi]

there is little to no difference in bash scripting than zsh scripting

[fickle granite]

shame, since bash scripting is an affront to all that is holy

[gleaming wadi]

if anything i rather use bash for scripting just because it's no better or worse

i can't remember a single shebang in a shell script that wasn't #!/bin/bash

[proud nova]

I always use #!/usr/bin/env bash

I think it is more portable?

[gleaming wadi]

i was about to say that as well

yes that is correct

[fickle granite]

in theory, yes

in practice I imagine 98% of bash scripts these days are running on Ubuntu

[gleaming wadi]

sounds about right

[proud nova]

This SO post has more details for people like me who can't leave enough alone https://stackoverflow.com/a/53116875

[fickle granite]

worrying about POSIX is ... probably a waste of time

[gleaming wadi]

zsh does't comply with POSIX, and i can't imagine there are many things nowadays that are fully in-line with POSIX

[trail sapphire]

really?!?
i can say for a fact that those numbers are grossly wrong unless you mean to say that they might be able to run on ubuntu

[fickle granite]

🤷

I have no facts, just hunches.

[trail sapphire]

there are many widely used distributions out there other then just ubuntu, just saying

[fickle granite]

well sure

betcha ubuntu is the most popular though. No way to objectively measure, so it's a safe bet 🙂

[trail sapphire]

probably not as common as you might think in the enterprise world

[fickle granite]

well ... I'd be mildly surprised to see Linux in "the enterprise world", so ...

[trail sapphire]

i'm talking outside of the desktop, what do you think runs the majority of the internet?

[fickle granite]

I'm guessing Red Hat is the winner there

Yeah I'm a back-end guy, current job (granted, a small shop) uses Ubuntu docker images

AWS uses Amazon Linux of course

[fickle granite]

The majority of the Internet is AWS 🙂

desktop is indeed a rounding error

[trail sapphire]

yeah, those are some of them, and then there is so many others that are red hat or debian derivatives and others such as Alpine that does not fit in to that category

[fickle granite]

you realize I'm not terribly serious.

[trail sapphire]

hahaha, true, so very true 😆
i laugh even though it's kind of sad

i've been running linux on my main computers for a lot more then half my life and have had the fortune to be able to choose to run linux on my work computers for all but my first two years of my work life

[fickle granite]

my world is: MacOS on laptop, Linux for backend.

[trail sapphire]

not a bad setup either, still unix like 👍

[fickle granite]

for all I know Ubuntu is actually usable on laptops now, but it wasn't when I last tried (admittedly a long time ago -- 2013)

[trail sapphire]

it defensively is and Fedora and Arch as well and a quite a few more, even gentoo, but some of course takes more effort then others
it's a bit harder if you want to go the bsd route with a laptop (i'm not counting macos even if it's probably more related to bsd than it is to linux) 😬

[fickle granite]

macos is technically related to BSD, but that's utterly irrelevant.

what's relevant is that Apple spends time and effort getting their device drivers working

and doesn't bother open-sourcing them.

Most linux distros were crippled by their insistence on only shipping Free® code, which meant their device drivers sucked for ages

"Oh, you wanted to use the trackpad? Sorry; not Free®"

"Wifi? Sorry, no Free® drivers available"

[trail sapphire]

yeah, but that is only because everyone that want to have access to much of the hardware manufacturers documentation has to sign NDA forms and are not allowed to release the source for what they make
however, some manufacturers are better then others in that regards
and linux distributions has gotten more pragmatic about the whole thing that "everything shipped with the distribution must remain free" to mitigate some of those problems

[fickle granite]

yep, I updated my comment to note that it's been almost 10 years since I checked Ubuntu on laptops. I like to think the situation has improved since then

[trail sapphire]

it has, but it's still advisable to research the hardware comparability online before buying if you intend to run linux on it full time

[latent wren]

How to add input rule anywhere to anywhere?

Chain INPUT (policy ACCEPT)
target     prot opt source               destination          
ACCEPT     tcp  --  anywhere             anywhere            tcp dpt:mysql  ```

[trail sapphire]

you already have that

[latent wren]

I copied from a forms, but dunno the command to do it

[trail sapphire]

you can get the commands of all the current iptables settings by running iptables-save
it will not actually save the configuration unless you redirect the output to the right file

[latent wren]

Basically trying to ssh without key but password but dunno why

(test@12.34.56.78) Password:
Connection to 12.34.56.78 closed by remote host.
Connection to 12.34.56.78 closed.


/etc/ssh/ssh_config
    PasswordAuthentication yes


cat /etc/passwd
test:x:1002:1002:,,,:/home/test:/bin/bash


cat /etc/group
test:x:1002:


ls -l /home
drwxr-x--- 2 test   test   4096 May 23 17:15 test


/var/log/auth.log
pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=87.65.43.21  user=test
Accepted keyboard-interactive/pam for test from 87.65.43.21 port 56455 ssh2
fatal: PAM: pam_setcred(): Failure setting user credentials

[trail sapphire]

i guess the user has a password set too?

[latent wren]

Never heard of password set

test have a password

[trail sapphire]

what does your /etc/pam.d/sshd look like?

[latent wren]

# Standard Un*x authentication.
@include common-auth

# Disallow non-root logins when /etc/nologin exists.
account    required     pam_nologin.so

# Uncomment and edit /etc/security/access.conf if you need to set complex
# access limits that are hard to express in sshd_config.
# account  required     pam_access.so

# Standard Un*x authorization.
@include common-account

# SELinux needs to be the first session rule.  This ensures that any
# lingering context has been cleared.  Without this it is possible that a
# module could execute code in the wrong domain.
session [success=ok ignore=ignore module_unknown=ignore default=bad]        pam_selinux.so close

# Set the loginuid process attribute.
session    required     pam_loginuid.so

# Create a new session keyring.
session    optional     pam_keyinit.so force revoke

# Standard Un*x session setup and teardown.
@include common-session

# Print the message of the day upon successful login.
# This includes a dynamically generated part from /run/motd.dynamic
# and a static (admin-editable) part from /etc/motd.
session    optional     pam_motd.so  motd=/run/motd.dynamic
session    optional     pam_motd.so noupdate

# Print the status of the user's mailbox upon successful login.
session    optional     pam_mail.so standard noenv # [1]

# Set up user limits from /etc/security/limits.conf.
session    required     pam_limits.so

# Read environment variables from /etc/environment and
# /etc/security/pam_env.conf.
session    required     pam_env.so # [1]
# In Debian 4.0 (etch), locale-related environment variables were moved to
# /etc/default/locale, so read that as well.
session    required     pam_env.so user_readenv=1 envfile=/etc/default/locale

# SELinux needs to intervene at login time to ensure that the process starts
# in the proper default security context.  Only sessions which are intended```

# to run in the user's context should be run after this.
session [success=ok ignore=ignore module_unknown=ignore default=bad]        pam_selinux.so open

# Standard Un*x password updating.
@include common-password

Tried restart between config changings
sudo systemctl restart ssh.service

Theres a ssh debug view

debug3: send packet: type 61
debug3: receive packet: type 60
debug2: input_userauth_info_req: entering
debug2: input_userauth_info_req: num_prompts 0
debug3: send packet: type 61
debug3: receive packet: type 52
debug1: Authentication succeeded (keyboard-interactive).
Authenticated to 12.34.56.78 ([12.34.56.78]:22).
debug1: channel 0: new [client-session]
debug3: ssh_session2_open: channel_new: 0
debug2: channel 0: send open
debug3: send packet: type 90
debug1: Requesting [e-mail]
debug3: send packet: type 80
debug1: Entering interactive session.
debug1: pledge: filesystem full
debug1: ENABLE_VIRTUAL_TERMINAL_INPUT is supported. Reading the VTSequence from console
debug3: This windows OS supports conpty
debug1: ENABLE_VIRTUAL_TERMINAL_PROCESSING is supported. Console supports the ansi parsing
debug3: Successfully set console output code page from:65001 to 65001
debug3: Successfully set console input code page from:65001 to 65001
debug3: send packet: type 1
debug1: channel 0: free: client-session, nchannels 1
debug3: channel 0: status: The following connections are open:
  #0 client-session (t3 nr0 i0/0 o0/0 e[write]/0 fd 4/5/6 sock -1 cc -1)

debug3: Successfully set console output code page from 65001 to 65001
debug3: Successfully set console input code page from 65001 to 65001
Connection to 12.34.56.78 closed by remote host.
Connection to 12.34.56.78 closed.
Transferred: sent 1976, received 1760 bytes, in 0.0 seconds
Bytes per second: sent 128989.2, received 114889.2
debug1: Exit status -1```

Found temporary solution, but trying to make it perm
ssh -o PreferredAuthentications=password -o PubkeyAuthentication=no test@12.34.56.78

[latent wren]

Now softlocked ubuntu out of ssh and the new user is not in the sudoers file

[trail sapphire]

i'm guessing you have another user that is when you need to do things as an administrator?

[earnest oar]

have a good day today

[latent wren]

Cannot become root, saids new user not in sudoers file

[shut nebula]

i hate that error

painful

[trail sapphire]

and you don't have a root password set or know the root password so that you can run su - and fix the situation?

[latent wren]

Nope, aws so hard to do Ubuntu advanced options so made new ec2 instead

[wanton cedar]

🙂

[sudden leaf]

hi

[frozen tendon]

hi guys, I was trying to run pyFirmata,the library for running python with arduino and when I wrote the code,it told me that I haven't downloaded the pyFirmata library even though I did.

I am on manjaro

[runic quarry]

What did you run to install it?

Did you activate a virtual env or anything?

[frozen tendon]

I didnt

I ran pip install pyFirmata and then pip3 install pyFirmata just in case

[trail sapphire]

this is probably due to how your vs code is setup and probably runs the code under another python environment
i think this question probably belongs in #editors-ides as they know quite a bit about such things
when you have your environment fixed you might want to drop by #microcontrollers if you need help with any arduino related things

[frozen tendon]

Thank you

[trail sapphire]

i saw that you got help from Asterisk and solved your problem 🙂

[shy igloo]

yes, I helped him

[frozen tendon]

yes : D, thanks once more for telling me about that channel

[shy igloo]

funnily enough, it wasnt an ide issue

[frozen tendon]

lol

[shy igloo]

also it happened right after I joined

[frozen tendon]

lmao

[shy igloo]

soo, this is where we talk about linux? hmm.. anyway, strange day.

I might like this place

[trail sapphire]

i guess it is 🙂

[shy igloo]

okaaay, what's your linux? staying on topic

[trail sapphire]

saw that too, it was a general "python" environment issue
i generally use pyenv to install all the different python environments i work with and try to stay away from using the system python for anything other then what's already bundled with the os
also, i tend to use venv or pipenv for everything to keep my python installations clean without any packages other then what comes standard with the interpreter

[shy igloo]

So you keep it clean

Im lazy, so I use the system pip. It is a bad idea, but it works for me

I've only used venvs for setting up requirements.txt

for my projects

[trail sapphire]

i like the environment to be reproducible and dependable so that i know what's there and not there

[shy igloo]

aaah. I guess my method also means that python would be a hassle to remove

[trail sapphire]

when i'm lazy (which happens a lot) i use pipenv instead of pure venv

[shy igloo]

actually I did do it

had to remove the site-packages dir first

before re-installation

but I won't need to unnstall so it works

but I wouldn't put conda or venv in my .config/fish (i use fish)

just in case

[shy igloo]

so I guess you might have been better at solving the issue?

by helping him set up a proper venv

[trail sapphire]

haha, i think you did an excellent job
it can be hard to guide someone else without knowing what environment they use and with users of unknown skill level
and setting up pyenv might be hard to do blind
i've done it at least once on this server, and that time it took hours due to some unforeseen issues when i didn't know that the user was running the commands as root since the person didn't have any other accounts beside root on the system

[shy igloo]

well, I figured I needed to get the small stuff out first

but man if he didn't show me his prompt, it would have taken a lot longer

[trail sapphire]

i find that i learn a lot, especially about edge cases, while helping and teaching others
so it's quite rewarding personally in that sense too
when i am teaching someone in person irl i usually say that it's my fault if the other person can't understand, because then i'm not explaining it in the right way or at the right level for them and need to explain it in other ways until they understand and learn, at least as long as we both but in a serious effort

[trail sapphire]

i know that feeling 💡 🙂

[shy igloo]

yeah, I feel the same way about this

it feels great

@trail sapphire but also thx for the vscode telemetry setting. I kinda stopped using vscode so I never thought to check

[shy igloo]

Media centers, they can be very useful but, if installed as an os, only work well on embedded devices.

[burnt burrow]

I would like to build just a systray bar in python someone can help me? I just found tutorial for to show icon in existing systray bar

[main olive]

?

[burnt burrow]

For me?

[main olive]

Ya

Where we can host Unix?

[burnt burrow]

I use Linux

I want a simple systray bar in X11 with python from scratch

[main olive]

So Where you hosted

I am new to python bots

[burnt burrow]

I don't understand

[main olive]

YOu have computer of Linux

Or You connect it remotely

[burnt burrow]

Yes

[main olive]

Ohk

I want to know how to host Linux VM in cloud

[burnt burrow]

I don't know

[main olive]

Ohk

No problem

[sleek swan]

A few questions:

1. Right after installing Ubuntu and I first create an account to log in it has sudo privileges by default?

2. When I create another account in the terminal, does it have sudo by default?

3. How do I log into root, since I don't know the password? Do I have to use sudo in the first account created to modify the root users password?

4. When I do ls -l on a file, how do I find out who the owner is and what group is referred to?

[untold socket]

As for 3: You input the password after you input the username

[candid dove]

1. no ide
2. it has to be in the group sudoers /etc/sudoers
3. sudo su root
4. it shows the user and group in the output

4. should look like this -rw-r--r-- 1 <user> <group> 2583047 Nov 29 2019 <filename>

[fickle granite]

1. Yes
2. No

[trail sapphire]

3. use sudo -i instead, don't login with root unless you can't access the system in any other way

[fickle granite]

I do sudo -s, that gets me a root shell

[shy igloo]

I use sudo su, that works too

[modern karma]

add -E to preserve env variables

like $HOME

[dapper musk]

whats wrong with sudo su?

[glad stirrup]

same thing as su or sudo -s

except that you use sh, not bash as your shell

[shy igloo]

Although for me on arch it uses bash

[glad stirrup]

it does sh for me

might be a distro thing

or just a debian stable moment

[shy igloo]

I think so

On Ubuntu, arch manjaro or gentoo it uses bash

But I guess because I’ve never used Debian stable

[shrewd stratus]

sudo su will just use the shell "assigned" to the root user, right?

FWIW sudo -sE preserves the environment, so if you've got prompts in PS1 and the like, it'll be preserved for the root shell

[dapper musk]

yes it will use shell listed in /etc/passwd for root

[ashen grotto]

hi, when i do this, both $og and $1 end up printing the same thing

is there something wrong with how im substituting these variables into printf?

[trail sapphire]

it depends on your input to awk
awk will auto-create any unset variable when it's first used and set it to be the full line of input, in your case this is $og unless you assign it in some way somewhere
$1 is the first field of the current line of input as awk auto-splits the input
in your case i think your input line only contains one field which would make them the same

[languid fern]

installed linux-headers-generic still gives this error:
fatal error: linux/kref.h: No such file or directory
why?

[trail sapphire]

a little more context please, what are you trying to do in the first place?

[languid fern]

making a kernel module

[trail sapphire]

and your distribution?

[languid fern]

ubuntu 22.04

for testing, for development arch

well lets stay in 22.04 cuz i got that working on arch

[trail sapphire]

can you find the file in /usr/src/linux-headers-*/include/linux/kref.h on your ubuntu installation?

[languid fern]

no? or else it wouldnt have given that error

still lemme check

[trail sapphire]

just checking if that is the problem or if it's a problem with the search path when building

[languid fern]

ah its there in the src

wait bruh

maybe because its not in /usr/lib/include hence its not getting detected?

maybe?

[trail sapphire]

depends on your build tools/chain that you are using

[languid fern]

im using gcc

and make

[trail sapphire]

and configure/automake/autoconf maybe?

[languid fern]

no just a Makefile

simple

[languid fern]

any idea?

[trail sapphire]

if you don't rely on any of those tools to find the kernel headers how do you specify the path to them?
https://www.kernel.org/doc/Documentation/kbuild/modules.txt can be a good place to start

[languid fern]

nice

[fickle granite]

well timeout is a maximum not a minimum

no idea what the fuzzer is, so I can't comment on how long it takes

[trail sapphire]

@wet folio don't forget to catch your timeout exception, other than that i have left most of your code untouched:

#!/usr/bin/env python3

import os
import subprocess
import time

input_dir = 'imgF'

for file in os.listdir(input_dir):
    if os.path.isfile(os.path.join(input_dir, file)) and file.split('.')[-1] == 'image':
        try:
            start = time.perf_counter()
            proc = subprocess.run(
                [
                    '/your/command', 'and', 'all', 'your', 'command', 'arguments'
                ],
                timeout=3600, stdout=subprocess.PIPE, stderr=subprocess.PIPE
            )
            end = time.perf_counter()
            print(f'file: {file} ran for {end - start} seconds and exited with code {proc.returncode}')
        except subprocess.TimeoutExpired as ex:
            print(f'file: {file} timed out after {ex.timeout} seconds')
```the `start` and `end` timers and `print()` was just for illustration and not really needed
and you probably just want to do a `pass` for that exception in your code and in that case you can skip ` as ex` as well
this was mostly just to show you some example code for that use case

[wet folio]

thank you so much, I will try it do that now

[shut nebula]

What are some user management/security files that are commonly on a linux distro other then /etc/passwd and /etc/shadow ?

[digital prawn]

/etc/group

@shut nebula

[shut nebula]

yeah just found that one

thanks

[proud nova]

sudoers?

[wet folio]

sudo mount -t ntfs -o loop ext4.image /mnt
NTFS signature is missing.
Failed to mount '/dev/loop7': Invalid argument
The device '/dev/loop7' doesn't seem to have a valid NTFS.
Maybe the wrong device is used? Or the whole disk instead of a
partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around?`

any suggestion why its not working, mount operation

Originally tried this alsosudo mount -o loop ext4.image /mnt

[proud nova]

You could try making a valid ntfs image like:

# make a 1GB image
dd if=/dev/zero of=image.bin bs=1M count=0 seek=1024
losetup -fP image.bin
mkfs.ntfs /dev/loopN # where N is the loopback device from the previous step
losetup -d /dev/loopN
mount -t ntfs -o loop image.bin /mnt

For me this worked, so maybe the image you are trying to mount doesn't have an NTFS file system at the top level.

[trail sapphire]

the image name has ext4 in it, which looks to indicate that it's in the linux ext4 file system format
now you are trying to mount it as ntfs which is a windows file system format, why?

[sleek swan]

A question about bash:
Why does this whileloop not work? ```sh
#!/bin/bash

num=1
while [[ $num < 10 ]]; do
echo "$num"
num=$(( $num + 1 ))
done

[shrewd stratus]

you need to compare using -lt

[sleek swan]

Why does < not work?

[shrewd stratus]

according to the manual, it compares lexicographically

so 1 < 10 but 2 > 10

[sleek swan]

What does lexicographically mean?

[sleek swan]

Wait what?

[shrewd stratus]

alphabetically

but in the general sense

[sleek swan]

im lost, how is 1 < 10 and 2 > 10?

[shrewd stratus]

that's

"1" < "10"
"2" > "10"

[sleek swan]

Does this return true and false?

Or how does it evaluate?

[shrewd stratus]

both would return true

or 0 in bash

[sleek swan]

Why?

[shrewd stratus]

because when compared as strings, 2 comes after 10

[sleek swan]

Wdym comes after? How is the order determined?

[shrewd stratus]

!e

nums = [str(x) for x in range(11)]
nums.sort()
print(nums)

[shy yoke]

@shrewd stratus :white_check_mark: Your eval job has completed with return code 0.

['0', '1', '10', '2', '3', '4', '5', '6', '7', '8', '9']

[shrewd stratus]

when you compare "10" and "2", you compare character by character

and since "1" comes before "2" as a character, "10" also comes before "2"

again, as a string

don't think of them as numbers

that's what's a lexicographical (or alphabetical, to simplify) sort

[sleek swan]

oh like: a, ab, b, c?

[shrewd stratus]

right

[sleek swan]

cool ty

[shut nebula]

Is there a way to disable the root password if the ssh connection request is coming from a specific IP or hostname

[wet folio]

actually i dont want windows format, but when i tried without ntfs, i got such error and used ntfs flags to make it work, not sure why its not working properly for me

does anyone knows how to create apfs file system for macos, For example, On linux we use mkfs.ext4 to create it, is there command for apfs too? any ideas!

[trail sapphire]

don't get your hopes up, apfs is proprietary, just like most things apple
why do you want to create a apple file system from linux?

[trail sapphire]

for ext4, have you tried:

dd if=/dev/zero bs=4096 count=1024 of=ext4.image
mkfs.ext4 ext4.image
sudo mount -o loop ext4.image /mnt
```do you get any errors when you run that?

[trail sapphire]

not easily that i know of anyways
you might be able to with some advanced pam configuration, but i'm just guessing here
but why allow root to login directly at all over ssh?
or at least disable password authentication over ssh for root and just allow authorized keys for logging in as root over ssh

[wet folio]

nono. I am trying to understand macos FS, APFS one

I was just trying to understand, how different APFS creation and mounting using diskutil compare to linux mkfs.ext4

[wet folio]

it works now, i just add extra path where I wanna mount it. tried on different server machine, preivous one is bad haha

Is there any specific discord server, where I can learn and explore more about File System and different OS

[trail sapphire]

i don't think you will have much luck with that, as far as i know you can't create apfs file systems on linux and only mount them as read-only file systems using apfs-fuse or buy paragons apfs for linux software to gain read-write abilities

[wet folio]

it has write support for linux, https://github.com/linux-apfs/linux-apfs-rw

i found this

apfs-fuse or buy paragons apfs good suggestions to explore it

I will check it further to get familiar with it a bit

Disk utilities (diskutil)- Format, Verify, Repair (local disks.)- does it mean it create apfs file on macos?

[trail sapphire]

oh, i haven't seen that project 👍

[wet folio]

i just found it when searching through many links:)

[trail sapphire]

as that code is experimental and writes data it comes with higher risk of corrupting your apfs file system and any data on it, so don't use it for anything that has files on it that you care about
apfs-fuse is probably much safer as it doesn't write any data, so it shouldn't be able to corrupt the file system or any data on it

[trail sapphire]

i would think of "formating" a file system in this context don't just mean to overwrite and destroy the data but also to create the file system in question but worded for the casual user

[wet folio]

Make sense, in other word they use this tool to create a apfs type of FS, and mount it, i saw some said they use diskutil mount or mount_apfs as well

interesting

[trail sapphire]

yeah, unless they use the gui on mac to "format" a drive or partition and create a new apfs volume

[wet folio]

thanks a lot. learned many new things

[visual laurel]

@abstract notch there's a command called tar, just search at google, I am getting answers in another language and I don't remember its usage by myself

[abstract notch]

oof

@visual laurel im still getting ffmpeg was not found even when they are now in the same dir

[visual laurel]

Show me your play func

[abstract notch]

do i need to do something to that

[abstract notch]

iv tested it on my pc it works

[visual laurel]

You might need to change the path

[abstract notch]

on my pc i had to add it to path

[visual laurel]

Could you show the contents of this dir

[abstract notch]

nah

[visual laurel]

dir your_folder

Might be lsdir idk about your os

[visual laurel]

Why

[visual laurel]

Ok do

mv ffmpeg-5.0.1-amd64-static ffmpeg```
just to rename the file for convenience

[abstract notch]

ok

@visual laurel so do you know how to get it to recognize that it has ffmpeg on the aws, because on my pc i had to add it to path, aka system environment variables

cant even do ffmpeg -version 😬

[visual laurel]

Ok in the ffmpeg foler there's ffmpeg file, move it out to your project root with cp ffmpeg/ffmpeg ffmpeg-ex and show me your play function

You don't need to add it to path, there's a better method

[abstract notch]

i already know the play function works it was working on my pc

[visual laurel]

Man just do what I say

[abstract notch]

i had these same issues on my pc but alot easier to fix

[visual laurel]

If you knew how to set it up on linux you wouldn't need help, yet you do

[abstract notch]

bruh

i never said i knew

[visual laurel]

So yeah follow the instructions and your issues will be fixed

!d discord.FFmpegAudio parse the path to your ffmpeg to executable keyword param @abstract notch

[shy yoke]

discord.FFmpegAudio

class discord.FFmpegAudio(source, *, executable='ffmpeg', args, **subprocess_kwargs)```
Represents an FFmpeg (or AVConv) based AudioSource.

User created AudioSources using FFmpeg differently from how [`FFmpegPCMAudio`](https://discordpy.readthedocs.io/en/latest/api.html#discord.FFmpegPCMAudio "discord.FFmpegPCMAudio") and [`FFmpegOpusAudio`](https://discordpy.readthedocs.io/en/latest/api.html#discord.FFmpegOpusAudio "discord.FFmpegOpusAudio") work should subclass this.

New in version 1.3.

[visual laurel]

You don't need it to respond dude

It just has to be in your project folder

And yeah if you really want it to respond do ./ffmpeg/ffmpeg --version

[abstract notch]

@visual laurel it works

[visual laurel]

Ok cool

[abstract notch]

can i pay you or something

lol

[visual laurel]

Nah I am good thanks

[abstract notch]

🤝

you did a good deed today

@visual laurel so uh

it doesn't work

i thought it was running on the thing but it was on my pc lmao

same error

[visual laurel]

Did you parse the path as I said

That's why I am asking for your play code

[abstract notch]

ok so yea im doing it wrong

[visual laurel]

Ok how do you parse source

[visual laurel]

I need to see how you make the FFmpegPCMAudio object

[visual laurel]

Remove this freak from my eyes

[abstract notch]

💀

[visual laurel]

And don't ever show it to someone

We don't help with music bots using ytdl

[abstract notch]

cuh

i didnt even write this

[visual laurel]

Then remove it and pretend you never sent that

[abstract notch]

pretty sure it needs it tho

@visual laurel how do i make it not use it then

and also ffmpeg to work lol

[visual laurel]

Your issue isn't related to ytdl

[abstract notch]

yea ik

[visual laurel]

Search for usages of FFmpegPCMAudio(

[abstract notch]

yea its here

[visual laurel]

Is that the only usage

[abstract notch]

sry

thats it

[visual laurel]

Add a kwarg executable="ffmpeg/ffmpeg"

To pcm

[abstract notch]

like so?

[visual laurel]

Yes

./ffmpeg-ex better tho

Without that might not work

[abstract notch]

ok

[abstract notch]

ClientException: ./ffmpeg-ex was not found.

how is this even possible

maybe i do without the /

[visual laurel]

Is this in the same folder with the entry file

[abstract notch]

what is the entry file

like the thing im running?

[visual laurel]

Yeah

The concrete python file

[abstract notch]

oh bruh i deleted it

bruuuuuuuuuuuuuuuuuuuuuuuuuu

it deleted

[visual laurel]

Just copy it from ffmpeg again

[abstract notch]

because git pull deletes it

[visual laurel]

And add it to gitignore

[abstract notch]

that was also deleted

yeesh 1 sec ill get it back

@visual laurel ok so i installed ffmpeg-5.0.1-amd64-static in the root directory so it dont get deleted, and i did "ffmpeg-5.0.1-amd64-static/ffmpeg-ex" but it says it cant find it, how do i tell it to start in the root dir

[visual laurel]

Could you share the structure of your project

[abstract notch]

its just all in one folder

so there is root, then mg, then kh, and thats where the .py file is

[visual laurel]

Or try this

$ cp ffmpeg-5.0.1-amd64-static/ffmpeg ffmpeg``` and set the executable to `./ffmpeg`

[abstract notch]

and the ffmpeg-5.0.1-amd64-static is in root

[visual laurel]

Don't forget to gitignore all those

[abstract notch]

yea well thats what i apparently forgot to do

last time

[abstract notch]

i did that but same error

how do i tell it that its in the root folder

not the same folder

like ~ folder

[abstract notch]

thats fair

you helped enough

[digital prawn]

what is a reasonable swap disk size (ssd) to go with 8gb of ram. are there any downsides to a large swap size?

[fickle granite]

well obviously: it uses up a lot of your disk

other than that, I don't know of any downside

allocate a bunch for swap; check in once in a while (with the "free" command) to see how much is being used; if only a bit is ever used, and you need the disk space for files, then shrink it some.

[digital prawn]

thanks

i need more swap to process videos in my py app

runs out of ram fast

[fickle granite]

you can add swap without rebooting iirc. "mkswap" and "swapon"

I know exactly nothing about processing videos ... I assume your app needs to hold more than one frame in memory at a time?

[plush ether]

Can someone please please help me fix this error
pygame.error: File is not a Windows BMP file

On linux

while using pygame.load("bla.png")

[digital prawn]

yea, so basically every frame must be converted to another format and stored in memory to 'pre-render' the video (which isn't supposed to be written to disk as it is rendered in the app), rendering in chucks would not be quite appropriate in this case due to the nature of the app and how it processes and outputs the content so I think I will have to find a way to compress the data before it's store in memory, or to further reduce the amount needed to render but i've already made optimizations I could think of

[trail sapphire]

why do you need to store all the frames in memory for this app instead of processing them in chunks as they are needed with a buffer of frames that are expected to be needed soon?

[digital prawn]

that would be ideal but it would require a large rework of a lot of existing structure

[trail sapphire]

have you seen there is a new channel on the topic now: #media-processing

[digital prawn]

yes, i was speaking here

[trail sapphire]

would the same be true for using a temp file or scratch file, memory mapped perhaps?

[digital prawn]

considered temp file - but either way i think it would still have to be read in chunks and then output accordingly

same as rendering in chunks

maybe it's the only solution if i can't optimize or compress the data any further

[fickle granite]

You're experiencing a traditional engineering tradeoff: either you buy more RAM, or you do a bunch of development work.

No easy answer. You're the best person to judge which is the best choice.

[digital prawn]

i understand that, my weighing in on this is that this will be a public program eventually so i'd like to that the optimization route. currently looking in to compression with zlib

[trail sapphire]

consider the memory mapped file as a temporary scratch file, then you can treat the data in your program as it's all in ram memory even if it isn't, that will save on ram for the users of your program

[trail sapphire]

you'll also find that general lossless compression will not do a very good job with image, video or audio data streams, that's why lossy encoders using techniques such as inter-frame, motion estimation and temporally compression is used for such media in codecs today

[digital prawn]

hmm thanks. i just came across one major problem

rendering video in chunk require async yes? 1 to render and 1 to display?

[trail sapphire]

if you still want to make a try with general lossless compression I would recommend something like zstd compression which os generally better and gives you sort of random access to the compressed data in chunks instead of having to seek from the beginning all the time

[digital prawn]

no way around that

i wrote a function to render in chunks, but now i am stuck because i can not display while rendering the next chunk

rendering in between chunks is too slow

[trail sapphire]

asyncio are mainly for io bound loads and not really suited for cpu bound loads, so probably won't help you as much as you were probably hoping for

[trail sapphire]

i'm guessing you are writing this program in python as you are talking about it here on the python server, right?

[digital prawn]

yes i am

i can not figure out how to render and display at the same time

which is what i think is ideal other than memory mapping

but thats a big rabbit hole

worth it you think??

[trail sapphire]

then threads might not even help you that much as cpython is using a GIL implementation unless you use some other python implementation that has no GIL like anaconda among others

[vital bay]

you can use dask or multiprocessing to get around it

[trail sapphire]

yeah, multiprocessing and IPC is probably what i would suggest, but this is venturing in to #async-and-concurrency territory now 😉

[vital bay]

certainly

[digital prawn]

i don't understand mmap

for my use sorry let me explain why

[trail sapphire]

sure, if the video stream is processed using a c lib which releases the GIL for much of the processing time, threads may help somewhat but would still need to hold the GIL for exchanging data with the python parts of the program

[digital prawn]

i have a list of strings that i generate. this list gets really big, and runs out of ram. so mmap i must create a tmp file, write my list of strings to it, and retrieve later to output? I don't understand how it differs from plain tmp file. I'm trying to store list of strings but this does not make sense to me

my video processing is all python

it processes a video in to text strings output

and you can imagine, the strings are large, and there are lots of them

[trail sapphire]

are you converting the video to ascii art to render it in a terminal or what?
otherwise I don't understand how or why you would represent binary image data as strings

[digital prawn]

spot on (ansi/ascii/unicode art)

not classic style ascii art

@trail sapphire

[untold socket]

This has surprisingly good quality

[terse pier]

FAA is the supervillian in this history

[wet folio]

I have one question, like I used DD command to create .image file and then mke2fs to make a ext4.image file. Now question is, i used bs=something and count=something, is there any specific command to use and check all details of a .image file like what node/bs/count they used to create it? any suggestions

[torpid cipher]

someone correct me if I'm wrong, but I don't think the block size you used to copy matters after the fact, it's more for optimizing the speed of the image/copy/etc... I don't think you can get that information later, unless you looked at the command history

[trail sapphire]

just like @torpid cipher notes, the parameters to dd is inconsequential after the fact and information about it will be lost (other then in the command history of the system where the command was run)
bs will just set how big the each chunk that is written will be, and setting this too low will hurt write performance while creating the image file with dd
and a higher bs number will help performance while creating the image file with dd, but only up to a point
bs * count = total image file size, which will be the end result
you can check the health and extract some information about the file system within the image file by running fsck -nf "$IMG" where $IMG is the name of the image file
much more information about the file system can be accessed using tune2fs -l "$IMG", which will include any parameters that you have set during creation of the file system with mke2fs or mkfs.* or set after it's creation with tune2fs
my personal preference is to run mkfs.ext4 "$IMG" instead of mke2fs -f ext4 "$IMG", but they are really the same thing

[wet folio]

thanks for the detailed explanation. i appreciate that

I have one inquiry that I come through, I found a tar.gz file, when I tar -xvzf tar.gz it actually generate a .img or .image file. May I know how did they created such image file first and then make tar.gz

how did they actually make it, any specific rules they followed to make such extension

[trail sapphire]

tar will only extract the archive file, it won't generate it

[torpid cipher]

you need to ask the person, there's more than one thing an *.img file could be

I mean, it's probably a filesystem image, given the question about dd, but I'm just saying 🙂

[wet folio]

so, after i done with dd command, and mke2fs, i need to tarball that .img file right

[torpid cipher]

tar czf whatever.tar.gz whatever.img

[trail sapphire]

if it's only one file tar is unnecessary, just compress the image file

[wet folio]

thats fine. make sense. actually its on code online. that person wont reply haha

[wet folio]

got it. redundant operation then

does one tar.gz file can contain multiple .image file?

[torpid cipher]

yeah

[trail sapphire]

a tar archive can contain multiple files of [almost] any kind

[wet folio]

thanks. its clear now

[torpid cipher]

👍

[wet folio]

I have one issue with my code that you helped me previously, actually after i tarball that file, the file name looks so weird, and i try to add that file name on it but seems i am doing something wrong. can I share it?

[trail sapphire]

just use compression by it self if you have just one file and use tar if you have multiple files you want to put into one archive

[wet folio]

got it. will do it for to test it

[trail sapphire]

share away 🙂

[wet folio]

cmd = (
    f'AFL_SKIP_BIN_CHECK=1 "./combined/afl-image-syscall/afl-fuzz"'
    f' -S "{instance_name}" -b "{shm_name}"'
    f' -s "fs/{args.fstype}/{args.fstype}_wrapper.so"'
    f' -e "samples/oracle/{args.fstype}-{i}.image" -y seed'
    f' -i "{in_dir}" -o "{out_dir}" -u {args.cpu_id}'
    f' -- "lkl/tools/lkl/{args.fstype}-combined-consistency"'
    f' -t "{args.fstype}" -i "samples/oracle/{args.fstype}-{i}.image"'
    f' -e "emulator/emulator.py" -l "/tmp/mosbench/tmpfs-separate/{args.log_id}/log"'
    f' -d "/tmp/mosbench/tmpfs-separate/{args.log_id}/" -r -p @@'
)

[trail sapphire]

there are better compression algorithms now then the classic gz (gzip) to choose from, bz2 (bzip2) is one, xz is even better and then there is the new kid on the block zstd (zstandard)

[wet folio]

last time you simply the code like this, so what happened the image file on that dir was like ext4-10.image , now when i extract that tar.gz file, it gives a image file name like test_inline_file.img, bit different

[wet folio]

noted down.

[trail sapphire]

just edited my above message slightly
also note that you need other options to the tar command to use another compression algorithm, not just the file extension

[wet folio]

like replacing {i} with test_inline_file this right

[trail sapphire]

it shouldn't, tar should preserve the original filenames of the files you put into the archive

[wet folio]

does renaming affect on that file?

[trail sapphire]

if you rename the file before you put it into the archive or after you extract it from the archive, yes

[wet folio]

after i archive, it will affect it, right

make sense

this linux fs so complicated to understand haha

[trail sapphire]

no, it shouldn't, archiving makes a copy of the file, so any changes to the original files after that will not affect what has been archived
but if you extract the archive and then make changes to the extracted files there will of course be changes to them, but the files in the archive will still be unaffected by any such changes

[wet folio]

clear now

[trail sapphire]

you can also try to run the command file "$IMG" on any file to learn the type of file as long as the file command knows about the format and can distinguished it from other formats, try it on your file system image files for example

[wet folio]

Will do it now, thanks 😊

[cerulean bolt]

$ sudo apt-get update
...
$ sudo apt-get install wine32
Reading package lists... Done
Building dependency tree
Reading state information... Done
Package wine32 is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
However the following packages replace it:
libwine

E: Package 'wine32' has no installation candidate

How can i fix this ?*

[ember quiver]

Read the instructions for the relevant distro: https://wiki.winehq.org/Download

[trail sapphire]

and you might want to say which distribution and version you are running

[cerulean bolt]

kali linux

[trail sapphire]

version?

[cerulean bolt]

idk tbh

[opaque ginkgo]

isn't Kali rolling

anyways I think you can just do sudo dpkg --add-architecture 386

then sudo apt update

then rerun install command

[dire junco]

play windows games in kali, why just don't use windows

[wet folio]

1st: $ sudo mount -t ext4 ext4.image /tmp/mount $ tree /tmp/mount /tmp/mount └── foo └── bar ├── acl ├── baz ├── fifo ├── hln ├── sln -> mnt/foo/bar/baz └── xattr

Second : `$ sudo mount -t ext4 new_image.img /tmp/mount
$ tree /tmp/mount/
/tmp/mount/

0 directories, 0 files`

Any suggestion, why this difference, or does not have information about inode, dir path so on

[wet folio]

found --bind command to mount, does it help tho

[ionic goblet]

@teal estuary Also, many of the linux errors originate from using WSL (assuming thats what you are using)

You can use a vm (not reccomended as ml will need plenty of your resources)
or use it bare metal

the process is very straight foward and you require a usb (you can use both windows and mint)
(I reccomend installing linux mint which is based on ubuntu. this means all resources present on your wsl system will be present on linux mint)
https://www.youtube.com/watch?v=IzVnTSklWa0
or
https://itsfoss.com/guide-install-linux-mint-16-dual-boot-windows/

[teal estuary]

what is bare metal?

[ionic goblet]

Installed on a non-volataile storage device, generally your hdd or ssd

@teal estuary

For the gui part, my advice is to choose cinnamon. if you want a more customizable and overall one of the best desktop managers/gui you can do that later

||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​||||​|| _ _ _ _ _ _ @teal estuary

[trail sapphire]

that doesn't looks like a mounted ext4 file system, there should be a lost+found directory in the file system root, like /tmp/mount/lost+found

[trail sapphire]

it's called "bare metal" when you run something directly on the hardware without any form or virtualization or hypervisor (like VMWare, Hyper-V, VirtualBox, WSL, Parallels, KVM, Xen, QEMU or such) in between, like when you run your main OS directly on your computer

[wise forge]

Huh, and I started to call, running on bare metal VPS, when it is run in Linux OS(lightsale/EC2), and not in some sort of application/container management systems like AWS ECS/EKS

[trail sapphire]

lightsale/EC2 is a VM, running things in that os environment isn't running it on bare metal, but at least it's not yet one more level as you get if you run docker or k8s on top of a VM, but you can also run k8s on bare metal without any VM in-between

[wise forge]

Yeah, we can

[wise forge]

But everything is cloud

And it is just does not make sense to do any other way xD

At maximum, people rent bare metal stuff/ or just with guaranteed resources

[trail sapphire]

just that little detail that "cloud" and "serverless" is just another way of saying "server hardware and hypervisor i don't have to manage myself" and "now i don't even have to manage the os or the application stack myself" but it all still runs on hardware but in someone else's data centers

[teal estuary]

okay I'm finally ready to get around to this. But I'm wondering as to what the difference is between WSL and running dual boot? Cause I do like the ability to just be able to boot up WSL rather than having to restart my entire system to switch OS since I have ADHD and in the amount of time it takes to commit to switching over it's highly likely that I get distracted and end up doing something else or otherwise lose the motivation to do "productive" things so the less time between switching the better for me

[proud nova]

(Jumping in here in the middle of the thread, pardon) WSL may be suitable for a lot of what you want to do. I use it a lot but also dual boot. In my experience here are some differences:

1. WSL is of course slower than running Linux right on your CPU. If this becomes a bottleneck you could consider dual booting.
2. Dual boot will let you run Linux services using systemd, OpenRC, or some other Linux init system (responsible for setting the system up at boot time essentially). I don't think WSL has any support for running services, so you'd probably still manage them using Windows services.
3. You can run graphical applications using WSL, but the configuration is tricky. (You have to make the WSL system talk to an X server like vcxsrv https://sourceforge.net/projects/vcxsrv/). But some desktop environments don't work well like this because they rely on services (see point 2)

I tried running KDE on WSL for instance, and gave up once I realized I didn't know how to run dbus, an essential dependency which runs as a service

Looks like this person figured it out https://scottspence.com/posts/gui-with-wsl

[proud nova]

If none of that makes sense then I guess just start simple. WSL is really easy to get going if you have Windows, compared to dual booting.

[fickle granite]

I found it fairly fiddly, but that might be because I was running Windows under Parallels.

Frankly I'm impressed it worked at all.

[proud nova]

heh same, I'm also impressed

On windows x86_64 it works pretty smoothly in my experience.

[teal estuary]

When you are talking about "slower" are we talking like the difference between a carolla and civic or like a lambo and me on a unicycle?

[fickle granite]

I will bet you won't notice any slowness.

[proud nova]

One area where I have noticed slowness is accessing Windows files from WSL (or vice versa I suppose).

If you keep your WSL files local to WSL it should be plenty fast

[ionic goblet]

For your own usecase though, WSL will have a strong impact on Machine Learning and slow it significantly (from my own experience). I had also faced many errors originating from WSL itself, I could not reproduce them in a seperate Linux boot when following the exact steps. If you have AMD cards, linux also has a better support for that, but it depends on what exactly you are doing with ML

[teal estuary]

ugh that's annoying cause I can't go fulltime linux with how much it frustrates me but also I'm a gamer so that may cause issues. But yeah ML is my main interest in programming so I definitely gotta figure out something for this situation

[ionic goblet]

Most games actually work on linux, except those who have an anticheat dedicated to Windows like Valorant

[shy igloo]

Although they may take longer to launch and you can’t have them on ntfs partitions

But you can check protondb for info regarding specific games

[dapper musk]

you can? since 5.15 NTFS r/w driver has been added. but why would you want to have them on NTFS partition?

[shy igloo]

proton cant handle symbolic links

but I have no other option because my linux partition is 64 gb and my windows partition is bigger and i cant get rid of windows

[trail sapphire]

can't you use bind mounts in place of symlinks?

[wet folio]

What could be the reason for such failed? any suggestions

[trail sapphire]

no idea, don't know what that fuzzer program is or does or why some unspecified compression would fail

[wet folio]

thats fine.

how can I change drw-r--r-- to -rw-r--r-- ? not able to change d part, also how can I replace it d with l? any suggestion

[ember quiver]

chmod 644 based on https://ss64.com/bash/chmod.html

[wet folio]

thanks, helpful

It seems 644 does not work for me, its still keeping d file type

[shrewd stratus]

l would be a link

do you want to create a symlink to a directory?

[trail sapphire]

d just means it's a directory and l that it's a link, you can't change the type of file without destroying the old and creating a new

[wet folio]

lrwxrwxrwx 1 sln -> mnt/foo/bar/txt

something like this

[shrewd stratus]

yeah. that would be a link

you need the ln command for that

[wet folio]

does those file have any extension? i cant see any extennsion name of file either txt or esle

[wet folio]

will do, thanks

[shrewd stratus]

in unix, files don't need an extension to be usable

[wet folio]

-rw-r--r--+ one file like this, how did it added + at the end

[wet folio]

so i can simply use touch command to create file?

[shrewd stratus]

probably a sticky perm bit (separate from the rwx perms)

[wet folio]

i see

[shrewd stratus]

https://en.wikipedia.org/wiki/File-system_permissions#Symbolic_notation

In some permission systems additional symbols in the ls -l display represent additional permission features:

+ (plus) suffix indicates an access control list that can control additional permissions.

[trail sapphire]

@wet folio check it with lsattr -vpl and then the filename

[wet folio]

ok will try it

i have one file acl but i can not see either its .txt or else. even type file acl it shows nothing. how can i create similar file which does not have any extension. I tried like this sudo echo "created a file" > acl1 for example, it shows permission denied

[trail sapphire]

i fixed my above commands
and with root privileges you can change them with chattr, +i which marks the fule immutable is a bit more common than many others and you can undo it with chattr -i

[trail sapphire]

sudo has no effect on files that you redirect to or from referenced on the command line, those are accessed with your normal user, for that use an interactive shell or wrap it in a script that you run with sudo
a common trick is to use <some command> | sudo tee <filename> or <some command> | tee -a <filename> for append mode

[wet folio]

so i dont have to use sudo always

[teal estuary]

So I'm extremely new to linux, but with me wanting to do ML it seems like I'm going to have to get used to it but I'm trying to figure out what the best way of going through the transitions but I need more info before I really make any decisions. Dual booting is the closest I'm going to get to fully converting. At least unless I get hooked onto it. The first thing I'm curious about is how multiple drives works on Linux, like in the video that @ionic goblet linked me it goes through how to dual boot off of the same drive that has mutiple partitions, obviously since I have only one SSD it makes sense for me to install it into there. But my SSD is 1TB, so I probably don't need to split that in half so idk how much I should allocate to Linux. Also will I be able to access other files on my windows side of the partition from within Linux or will I have to boot up the windows partition to access it? I also have 2 HDD installed in my machine, I assume that Linux will be able to access those just fine?

aarav also mentioned that most games actually work on linux except tthose who have anti-cheat dedicated to windows like valorant. Is there a way I can check before hand? Does EAC work on Linux? cause it seems like a lot of games I play seem to have this garbage anti-cheat installed

Also less important to my switching over and more out of curiousity, I looked up what the difference is between Unix and Linux and it seems that Linux is a family of operating systems and Unix refers to how those operating systems are created? So all Linux is Unix but not vise versa? The topic says unix or unix-like systems such as Linux or MacOS which is grammatically ambiguous, is MacOS unix-like or is it unix? Also is Linux unix or unix-like?

[warped nimbus]

Unix was a specific operating system that had several versions. Then it became standardised, so various vendors could make their own implementations of the Unix standard.

Unix was popular so other operating systems spun off it. These spin-offs are known as "Unix-like" operating systems because they behave like Unix, but aren't exactly Unix (they don't conform precisely to the standards nor is it their goal to).

[teal estuary]

Oh okay so Linux systems are just basically Unix+DLC?

[warped nimbus]

I suppose one could say that.

Linux is technically just a kernel.

[teal estuary]

Cool, do you know the answer to any of my other questions?

I assume discord works just fine on Linux?

[warped nimbus]

Yes, you can access Windows partitions from Linux and vice versa provided you have the correct file system drivers (though I've read that the Windows drivers for ext4 shouldn't be trusted for writes).

[trail sapphire]

there are no pure unix systems left today in the wild, all that is left is unix-like dialects like linux, macos, bsd and so on

[warped nimbus]

I don't use Discord on Linux but I suppose it works. In any case, you could just use it in your web browser.

[warped nimbus]

Well, what would you classify something like AIX as then?

[teal estuary]

You guys use discord through a web browser? That sounds awful

[vital bay]

discord works mostly ok on linux. the supported version is a snap

[trail sapphire]

that's probably a more close relative to the original unix systems then gnu/linux

[vital bay]

streaming properly with audio and video working right is still a hassle

[trail sapphire]

i use discord on the web regardless of platform other then on mobile devices where I use the apps

[teal estuary]

Lol well that's not good as a daily driver since I watch tv with friends all the time xD

Anyone know the answers to my other questions in #unix message though?

[vital bay]

i haven't played around with this on pipewire, but pulseaudio is terrible for it

[warped nimbus]

As for games, I don't game on Linux but I recall there are websites that can tell you how compatible they are with Linux. There are also communities around gaming for Linux which you can use to find discussions about particular games.

[trail sapphire]

that's very unlike my use case 😆

[vital bay]

you can probably just use WSL2 instead of dual booting

[teal estuary]

well that was my initial idea but then people said there will be speed issues

[warped nimbus]

I think I answered all your questions.

[vital bay]

you can easily access stuff in other partitions and drives from linux, but not from windows

you will never want to run real ML stuff on your own hardware anyway. speed shouldn't be an issue

toy scenarios work fine on WSL2

but anyway the performance is quite good, almost comparable to bare metal

[digital prawn]

anyone use i3-gaps?

[teal estuary]

Oh? At higher end you run it on cloud? Doesn't that end up causing speed issues?

[trail sapphire]

puleeaudio is terrible all around, i have had a much better audio experience ever since my distro switched to pipeline, even during the early days of that software

[vital bay]

wdym speed issues? running on cloud means you can parallelize over tensor cores and gpus

[teal estuary]

or does it not send info back and forth? You just kind of run it on cloud and take the result elsewhere?

[vital bay]

it doesn't send info back and forth, everything is on the cloud and you have remote access to a UI

[teal estuary]

ahhh okay

[vital bay]

well, you CAN send info back and forth. that can indeed be slow depending on how you do it

[teal estuary]

Well then we'll use WSL2 until it's a problem then I guess move onto that.

[vital bay]

but for instance, i do small tests on my local install, then i run the big stuff on colab or the university cluster

[teal estuary]

now I just gotta figure out how to properly set up everything in WSL2 cause the way I have it now is apparently not ideal xD

Cause right now I have WSL using Ubuntu with brew and pyenv which I've been told is bad

But who actually knows? I swear everytime I ask anything about Linux I end up getting sent in a completely different direction

[trail sapphire]

i prefer pyenv and venvs for my installs to keep it clean

[main olive]

Wait how come i never knew this channel existed

[vital bay]

i can't recommend anything on that end. i personally just use apt and also use anaconda instead of installing python directly

and indeed, every person will recommend something different. part of why people like linux in the first place, but also python itself has tons of potential for personalization and jank

[main olive]

Has anyone tried void linux and had audio problems?

[trail sapphire]

only heard of it but never tried it

[main olive]

@trail sapphire its pretty fast ngl

[teal estuary]

Which version of Linux would you recommend on WSL? How much does it actually matter if I'm just using tthe command line?

[main olive]

And smol

[vital bay]

my recommendation if you are new to it is ubuntu, simply because most of the errors you google will have solutions from people using ubuntu

[main olive]

Fedora or ubuntu

I like fedora more then ubuntu

[vital bay]

ubuntu 22.04 to get python 3.9. if you need the most recent stuff, fedora would do. fedora is annoying in that you have to upgrade the os or clean reinstall every 12 months max, and it's not as beginner friendly. the packages are way more up to date though

[main olive]

For some reason flatpak packages dont work on fedora

For now

[trail sapphire]

Fedora is very beginner friendly today and has been for a while

[teal estuary]

well Fedora doesn't show up on wsl --list --online so that's already harder 😛

[vital bay]

comparatively, sure. cuz it used to be a real pain in the ass 😛

but people still go wrong right at the installer

[shy igloo]

It’s steam

[vital bay]

also yum has to be fiddled with right from the beginning to get access to some repos

[shy igloo]

But yeah never thought of that

[vital bay]

i'm not saying it's bad btw, fedora is fantastic. i just would recommend ubuntu over it (personally) to a newbie

[shy igloo]

There’s official instructions from red hat if you want fedora wsl

[trail sapphire]

i think steam will not be able to tell the difference with a mount --bind

[teal estuary]

I already struggle enough with Linux I probably should avoid unofficial instructions for now 😛

[shy igloo]

Sorry they’re from fedora

Made a mistake

[shy igloo]

Also rpm fusion needed for certain apps

[vital bay]

that's part of what i mean

[shy igloo]

I thought about getting fedora but the package manager is slow and I hate vanilla gnome (did try some spins)

[trail sapphire]

for wsl2 ubuntu is probably a safe choice even if my own preferences leans toward arch or fedora, but that's not on wsl

[teal estuary]

well this has already gone poorly xD

seafu@DESKTOP-0GUEKUR:~$ pyenv install 3.7.13
Downloading Python-3.7.13.tar.xz...
-> https://www.python.org/ftp/python/3.7.13/Python-3.7.13.tar.xz
Installing Python-3.7.13...
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Doc/library/ctypes.rst
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Lib/test/test_unicode.py
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Modules/_ctypes/_ctypes.c
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Modules/_ctypes/callproc.c
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Modules/_ctypes/ctypes.h
(Stripping trailing CRs from patch; use --binary to disable.)
patching file setup.py
(Stripping trailing CRs from patch; use --binary to disable.)
patching file 'Misc/NEWS.d/next/Core and Builtins/2020-06-30-04-44-29.bpo-41100.PJwA6F.rst'
(Stripping trailing CRs from patch; use --binary to disable.)
patching file Modules/_decimal/libmpdec/mpdecimal.h
(Stripping trailing CRs from patch; use --binary to disable.)
patching file setup.py

BUILD FAILED (Ubuntu 20.04 using python-build 20180424)

Inspect or clean up the working tree at /tmp/python-build.20220609162127.4294
Results logged to /tmp/python-build.20220609162127.4294.log

Last 10 log lines:
checking for --enable-universalsdk... no
checking for --with-universal-archs... no
checking MACHDEP... checking for --without-gcc... no
checking for --with-icc... no
checking for gcc... no
checking for cc... no
checking for cl.exe... no
configure: error: in `/tmp/python-build.20220609162127.4294/Python-3.7.13':
configure: error: no acceptable C compiler found in $PATH
See `config.log' for more details

apparently I'm missing a C compiler xD

[warped nimbus]

You likely haven't installed the build-essential package

[teal estuary]

I indeed did not

[warped nimbus]

Even then you may end up needing some additional packages to build Python from source

[vital bay]

get build essential and try again, install what is missing, rinse and repeat

[teal estuary]

well we'll burn that bridge when we get there. It seems to be installing fine now

[warped nimbus]

This answer suggests additional packages https://stackoverflow.com/a/55331977

Well you can get away with less packages, but some things will end up being disabled in Python. For example, you may not end up having openssl.

[teal estuary]

BUILD FAILED (Ubuntu 20.04 using python-build 20180424)

Inspect or clean up the working tree at /tmp/python-build.20220609162417.6519
Results logged to /tmp/python-build.20220609162417.6519.log

Last 10 log lines:
  File "/tmp/python-build.20220609162417.6519/Python-3.7.13/Lib/ensurepip/__init__.py", line 29, in _run_pip
    runpy.run_module("pip", run_name="__main__", alter_sys=True)
  File "/tmp/python-build.20220609162417.6519/Python-3.7.13/Lib/runpy.py", line 201, in run_module
    mod_name, mod_spec, code = _get_module_details(mod_name)
  File "/tmp/python-build.20220609162417.6519/Python-3.7.13/Lib/runpy.py", line 142, in _get_module_details
    return _get_module_details(pkg_main_name, error)
  File "/tmp/python-build.20220609162417.6519/Python-3.7.13/Lib/runpy.py", line 109, in _get_module_details
    __import__(pkg_name)
zipimport.ZipImportError: can't decompress data; zlib not available
make: *** [Makefile:1141: install] Error 1

well that bridge appeared faster than expected 😛

[trail sapphire]

https://github.com/pyenv/pyenv/wiki#suggested-build-environment

[teal estuary]

I'm going through that link Mark sent

[trail sapphire]

that's all the packages that should be installed for builds to be successful

[trail sapphire]

i would suggest this link as it's pyenv:s official documentation

[teal estuary]

Can I copy and paste multiple lines at once or is it gonna cause problems?

[trail sapphire]

it's one command that can be written on one line if you omit the \ at the end of the lines, that only means "ignore new line" so that you can split the command on multiple lines, so it's made to run all lines at once

[teal estuary]

seafu@DESKTOP-0GUEKUR:~/python_projects/ra_tutorial$ /home/seafu/.pyenv/shims/python /home/seafu/python_projects/ra_tutorial/ra_tutorial.py
2022-06-09 21:46:22.968848: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-06-09 21:46:22.968900: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
  File "/home/seafu/python_projects/ra_tutorial/ra_tutorial.py", line 12, in <module>
    import reverb
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/__init__.py", line 27, in <module>
    from reverb import item_selectors as selectors
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/item_selectors.py", line 19, in <module>
    from reverb import pybind
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/pybind.py", line 6, in <module>
    _load_op_library.reraise_wrapped_error(e)
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/platform/default/load_op_library.py", line 47, in reraise_wrapped_error
    raise error
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/pybind.py", line 4, in <module>
    from .libpybind import *
ImportError: libpython3.7m.so.1.0: cannot open shared object file: No such file or directory

I am getting this error but when I try sudo apt-get install libpython3.7 I get this:

seafu@DESKTOP-0GUEKUR:~/python_projects/ra_tutorial$ sudo apt-get install libpython3.7
Reading package lists... Done
Building dependency tree       
Reading state information... Done
Note, selecting 'libpython3.7-stdlib' for regex 'libpython3.7'
The following package was automatically installed and is no longer required:
  libfwupdplugin1
Use 'sudo apt autoremove' to remove it.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

No idea what this all means. Should I sudo apt autoremove that plugin thing? lol I swear Linux is always messing with me.

[fickle granite]

you can but it won't hurt you if it doesn't

it's just a notice that you've got some cruft

[teal estuary]

Okay and what about the other part? Cause that one I can't just ignore xD

[fickle granite]

what other part?

[teal estuary]

the missing libpython3.7 that I can't get

[fickle granite]

oh, installing that package didn't do it?

[teal estuary]

that's what happens when I tried to

it didn't upgrade, install, remove or upgrade anytthing

I also don't know if this is something that I will have to install differently since I used pyenv. I'm also exactly sure if running it in VSC on the system version or the pyenv versions

[fickle granite]

what executable are you running, and (more importantly) where'd you get it?

[teal estuary]

what executable for what?

I opened VSC by typing code . in my WSL terminal

[fickle granite]

looks like you're trying to run /home/seafu/.pyenv/shims/python -- what is that? Where'd you get it?

[teal estuary]

Uhhh I'd assume through pyenv

[fickle granite]

I really don't know what you've done

[summer trail]

I've never used pyenv before, but - what do you see if you run this?

find /home/seafu/.pyenv/versions/ra_tutorial-env -name 'libpython*'

[teal estuary]

I see nothing

[summer trail]

OK, how about from:

/home/seafu/.pyenv/shims/python -m sysconfig | grep -i Py_ENABLE_SHARED 

[teal estuary]

Py_ENABLE_SHARED = "0"

[summer trail]

ok, so the version of Python that pyenv has doesn't have a shared library, and so it seems like it doesn't meet the requirements of libpybind

[teal estuary]

what does that mean in english? xD

[summer trail]

the module you're trying to use doesn't seem like it can be used with the Python interpreter that you installed through pyenv.

how did you install the module? Just a regular pip install?

[teal estuary]

so I can't use pyenv for it? or I need to get it for pyenv? Lol I'm confused xD

[summer trail]

how did you install the reverb module?

[teal estuary]

with pip install tf-agents[reverb]

[summer trail]

I don't know that it would help, but you could try

pip install --ignore-installed --no-binary reverb reverb

[teal estuary]

nope did not help

[summer trail]

what happened when you ran it?

[teal estuary]

seafu@DESKTOP-0GUEKUR:~/python_projects/ra_tutorial$ pip install --ignore-installed --no-binary reverb reverb
Collecting reverb
  Downloading reverb-2.0.1.zip (3.4 kB)
  Preparing metadata (setup.py) ... done
Skipping wheel build for reverb, due to binaries being disabled for it.
Installing collected packages: reverb
  Running setup.py install for reverb ... done
Successfully installed reverb-2.0.1
seafu@DESKTOP-0GUEKUR:~/python_projects/ra_tutorial$ /home/seafu/.pyenv/shims/python /home/seafu/python_projects/ra_tutorial/ra_tutorial.py
2022-06-09 22:09:12.692434: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-06-09 22:09:12.692493: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
  File "/home/seafu/python_projects/ra_tutorial/ra_tutorial.py", line 12, in <module>
    import reverb
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/__init__.py", line 27, in <module>
    from reverb import item_selectors as selectors
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/item_selectors.py", line 19, in <module>
    from reverb import pybind
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/pybind.py", line 6, in <module>
    _load_op_library.reraise_wrapped_error(e)
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/platform/default/load_op_library.py", line 47, in reraise_wrapped_error
    raise error
  File "/home/seafu/.pyenv/versions/ra_tutorial-env/lib/python3.7/site-packages/reverb/pybind.py", line 4, in <module>
    from .libpybind import *
ImportError: libpython3.7m.so.1.0: cannot open shared object file: No such file or directory
seafu@DESKTOP-0GUEKUR:~/python_projects/ra_tutorial$ 

[summer trail]

aha - https://github.com/deepmind/reverb/issues/15#issuecomment-697691409

seems like you need to export PYTHON_CONFIGURE_OPTS="--enable-shared" when installing a Python interpreter with pyenv

[teal estuary]

so I will need to re-install the versions?

pyenv --help

lol wrong window

[summer trail]

yes, I think you need to run export PYTHON_CONFIGURE_OPTS="--enable-shared" and then reinstall the Python version with pyenv install

[ionic goblet]

Its best to boot of your SSD since its the fastest persistent data storages. Think of paritions like virtual drives, and any operating system needing one of them to be stored on. It dosen't matter what the partition is on, your HDD or SSD. You can allocate as little or as much as you want. Probably depends on your ML models/ and other things. Linux can access windows partitions (ntfs) but not vice versa (windows cant access linux parititions). You can check using https://www.protondb.com/ for your games.

And almost all software that works for Windows can work on linux with an emulation layer, like wine which comes pre installed

Also at minimum I reccomend 8gb to linux, but if you want you can do way lower than that

Discord runs natively on linux, so you need not worry about that

[teal estuary]

hmmmm lol I think I just ran into an issue. There are visual elements to my project which now don't show because I'm using WSL2 with it xD

But on the bright side for once I don't have any actual error codes xD

[main olive]

say invalid syntax help me ?

[fickle granite]

can't help unless you show us the actual code

all I know from what you've showed me is that you've got a syntax error, and that you use Windows

[trail sapphire]

you need to provide more context for us to help you, you must show us what you are trying to do as well

[untold socket]

Are you trying to run a CLI command in Python REPL?

[teal estuary]

Can someone look over this https://seanthegeek.net/234/graphical-linux-applications-bash-ubuntu-windows/

and tell me if it's sound before I try it?

[shy igloo]

I have used this in the past

It stopped working eventually but it works well enough

although I’ve only ever used windows xserver apps for graphical ssh connections

@teal estuary

Although I’d recommend using gwsl instead of vcxsrv

[teal estuary]

what is gwsl?

[shy igloo]

An x server software

On the Microsoft store and I think itch.io

@teal estuary

It has menus to make launching wsl apps easier

[teal estuary]

idk what an x server actually is xD

[shy igloo]

Oh sorry

It’s the thing that lets you run graphical Linux apps

The display backend

It’s used on Linux but if installed on windows it can be used to launch wsl apps

Many providers but I use gwsl

[teal estuary]

hmmmm

[shy igloo]

But yeah it is fine to install a wsl but I recommend doing research on each Avalible distro

I had no issues

[teal estuary]

a wsl? I thought there was only the one?

[shy igloo]

Okay, let me explain

Wsl is the backend

There are many front ends

On the Microsoft store under the wsl section

Ex. Ubuntu, openSUSE, etc.

[teal estuary]

yeah explaining it to me like I'm 5 is a good idea, I am completely clueless when it comes to Linux

[shy igloo]

It’s okay

The server was meant for ppl like you

But you can also do your own research on wsl, Microsoft has an official guide as well

[teal estuary]

now I just need to figure out how to use the microsoft shop.... this thing is awful.

[shy igloo]

yeah

it is.

but ubuntu would be a good choice for beginners

you just need to know how to use the terminal

[teal estuary]

well ubuntu is already what I'm using, my issue atm is that my program that I'm running through WSL has visual elements that I can't see since it's currently terminal only

[shy igloo]

oh i see

then u just need wsl and gwsl

just start gwsl before your ubuntu and type the name of the program in terminal

also the wsl distro appears in your start menu

easy to launch

also need to enable wsl in windows features like in the article

@teal estuary

[teal estuary]

hmmmm well it doesn't appear to work

[frosty hazel]

why run linux gui's on windows?

[teal estuary]

when I get to the point where it gets me to test using xeyes I just get Error: Can't open display: localhost:0.0

[shy igloo]

did you open the gwsl software?

[shy igloo]

It's for linux APPS

and yeah, I would ask that but I guess he doesn't want linux on his main windows machine

so thats the second best option besides a vm

[frosty hazel]

i get that but why not just run them natively on windows?

[shy igloo]

I guess the app is linux-only

he said it was xeyes

[frosty hazel]

ah

imo wsl takes up enough memory as it is, graphical linux apps must take up even more

[shy igloo]

true: I don't use wsl, I use arch (linux) but I guess it IS his only option

[green brook]

Should I check if a directory exists before adding it to PATH or na

[teal estuary]

yes

[trail sapphire]

i would check that it's not already in the PATH

[green brook]

The paths I'm adding aren't already in PATH

[teal estuary]

any ideas for this error?

well idk what I did to fix itt but xeyes works now xD

[trail sapphire]

it looked like you didn't have three xserver running in windows

[teal estuary]

Now just to figure out why it's not working in VSC

xeyes works when I type it into the terminal in VSC but the code I'm running does not display anything.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import base64
import imageio
import IPython
import matplotlib.pyplot as plt
import numpy as np
import PIL.Image
import pyvirtualdisplay
import reverb

import tensorflow as tf

from tf_agents.agents.reinforce import reinforce_agent
from tf_agents.drivers import py_driver
from tf_agents.environments import suite_gym
from tf_agents.environments import tf_py_environment
from tf_agents.networks import actor_distribution_network
from tf_agents.policies import py_tf_eager_policy
from tf_agents.replay_buffers import reverb_replay_buffer
from tf_agents.replay_buffers import reverb_utils
from tf_agents.specs import tensor_spec
from tf_agents.trajectories import trajectory
from tf_agents.utils import common

# Set up a virtual display for rendering OpenAI gym environments.
display = pyvirtualdisplay.Display(visible=0, size=(1400, 900)).start()

env_name = "CartPole-v0" # @param {type:"string"}
num_iterations = 250 # @param {type:"integer"}
collect_episodes_per_iteration = 2 # @param {type:"integer"}
replay_buffer_capacity = 2000 # @param {type:"integer"}

fc_layer_params = (100,)

learning_rate = 1e-3 # @param {type:"number"}
log_interval = 25 # @param {type:"integer"}
num_eval_episodes = 10 # @param {type:"integer"}
eval_interval = 50 # @param {type:"integer"}

env = suite_gym.load(env_name)

env.reset()

[trail sapphire]

wow, more imports than there is other code? 😅

[teal estuary]

lol i'm just starting the tensorflow refinforcement agent tutorial, I'm assuming there is gonna be much more that gets added as I go on 😛

But I believe that this should be showing some sort of graphical images which it isn't so before I move on I should probably solve that xD

[teal estuary]

Any idea why it's not showing?

I also tried executing it from outside of VSC but still same result

[trail sapphire]

maybe you could try something really simple like this just to see if the gui work from python:

#!/usr/bin/env python3

import tkinter as tk

root = tk.Tk()
window.minsize(300, 200)
message = tk.Label(root, text="Hello, World!")
message.pack()
root.mainloop()

[teal estuary]

hmmm I can't seem to pip install tkinter?

ohhh apparently it's tk

or not? Lol idk wtf is going on xD

Traceback (most recent call last):
  File "/home/seafu/python_projects/ra_tutorial/ra_tutorial.py", line 3, in <module>
    import tkinter as tk
  File "/home/seafu/.pyenv/versions/3.7.13/lib/python3.7/tkinter/__init__.py", line 36, in <module>
    import _tkinter # If this fails your Python may not be configured for Tk
ModuleNotFoundError: No module named '_tkinter'

hmmm but it's saying that python should come with tkinter. except on Linux?

Lol my google searches are making it seem like it's as easy as just sudo apt-get install python3-tk but is it as simple as that? nay nay because Linux always has to try to infuriate me

[trail sapphire]

it's part of python, no need to install it

oh, well, if you use the python that comes with the distribution it's another story

[teal estuary]

I'm not using the distro one I'm using 3.7.13

[trail sapphire]

they chop up python into several packages

if you use apt-get to install things you are using the python installation that comes with your distro

[teal estuary]

I'm putting so much effort into getting this to work it probably woulda been faster to just use a dual boot with a GUI

I also tried python3 -m pip install tkinter

[trail sapphire]

tkinter is part of the python standard library, so with a normal installation of python it should already be present

[teal estuary]

you'd think so but yet I still get that error xD

[trail sapphire]

and you have done sudo apt update;sudo apt install python-tk ?

[teal estuary]

just did it and 0 upgrade, installed, removed and not upgraded

[trail sapphire]

what do you get if you run which python3?

[teal estuary]

/home/seafu/.pyenv/shims/python3

[trail sapphire]

oh, then your right, your not running the distribution version of python 👍

but i wonder if there was some missing dependencies on your system when you built python (while installing it with pyenv) that made it skip tkinter

on a clean base install of debian where i've built python with pyenv i got the same problem, when there i had only installed build-essential, no other packages before i ran pyenv install <version>

[teal estuary]

lol well I just installed Linux Mint somewhere but Ièm not sure where....

I think I might have accidentally just installed it onto my flash drive -.-

[teal estuary]

lol I can't figure out how to get into Linux Mint. When I reboot I don't get the option to boot Linux

I swear Linux and I do not get along xD

Well I've found my way onto Linux but now Idk how to do anything. Lol like my monitors are not ordered correctly gonna be a lot of googling

[ornate sundial]

I have no knowledge of tooling, so I was wondering if there's a method that allows us to easily reinstall packages to new versions of python. Having to upgrade python and then reinstall the same packages seems unnecessarily repetitive.

[fickle granite]

/usr/bin/python3 -m pip freeze | grep -v @ | xargs ./venv/bin/pip install seems to be working

disclaimer: just made that up

[teal estuary]

Lol so my main game atm I can't play on Linux, feels bad man

In fact almost all of my games don't work on Linux according to steam xD

[teal estuary]

What are Desklets?

[ornate sundial]

Are there any resources to begin learning about tooling?

[light gorge]

Maybe this https://missing.csail.mit.edu/

[tepid drift]

hey everyone, can I ask how to hide the menu bar while playing mpv in kali linux?

[shy igloo]

go fullscreen?

[sleek swan]

#!/bin/bash

num=0

while (( $num != 1 )); do
echo "Enter first string"
read st1

echo "Enter second string"
read st2

if [[ $st1 == $st2 ]];  then
  echo "The strings are the same"
else
  echo "The strings are not the same"
fi

again=""
while [[ $again != "n" || $again != "y" ]]; do
echo "Compare again? (y/n)"
read again
done

if [[ $again == y ]]; then
  echo "Comparing again"
elif [[ $again == n ]]; then
  echo "Quitting..."
  num=1
fi
done
``` Why does this get stuck on the while loop checking for `n` or `y`? When I enter y or n it doesn't break

[shrewd stratus]

use &&

[sleek swan]

Why should I do && rather than ||?

[summer coral]

[ $again != "n" && $again != "y" ]

[sleek swan]

Bruh nvm now I see

[sleek swan]

Does aynone know how to do float math in bash?

[shrewd stratus]

Using bc

[sleek swan]

when I do "2.5 + 4" | bc i'm basically passing the string on the left to the bc program?

[shrewd stratus]

usually you'd do echo '2.5 + 4' | bc

a string by itself can't be piped

[sleek swan]

Wait wdym?

I meant to put a echo infront of it but forgot if that's what you meant

[shrewd stratus]

yea. that's what I meant

[grizzled bay]

So I was poking around askubuntu.com on a way to upgrade/update from ubuntu 20.10 to 22.04 and found this https://askubuntu.com/q/1404281/1603295 post

In the comments of that post a user called Hannu said:

(For other readers, FWIW: 20.04 to 22.04 will be possible as they both are "LTS", expect to be notified; it will happen as 22.04.1 appears)

How would I upgrade/update from 20.10 to 22.04? I'm running this on a Vultr server which I SSH into and I'm a bit of a Linux noob so I'm not sure how |I would go about doing a clean install using SSH command line

Thanks for any answers.

[fickle granite]

you don't do a clean install; you do an upgrade

like that post says: at some point your message-of-the-day (that's the spew you see when you first log in) should announce the upgrade, and tell you what to do

until then, chill

[grizzled bay]

I haven't seen one of those (or just didn't notice it), but I issue is that the version I am using 20.10 is no longer supported and all the articles I've seen have said that the subsequent updates I need are no longer in the repo

[fickle granite]

I don't know what "Vultr" is. It's possible that you simply don't have enough control to upgrade.

ok so it's a hosted service

presumably they are in control of your distro

[shy igloo]

Vultr is a cloud server platform, you might need to backup and re-set up the server

[fickle granite]

https://www.vultr.com/docs/how-to-upgrade-ubuntu-18-04-lts-to-20-04-lts/

that's one rev old but perhaps the general ideas still work

[grizzled bay]

okay thanks all

[grizzled bay]

Okay, on a different note: I'm trying to setup a virtual environment on my Ubuntu 20.10 server and I've having some issues. I apologies for how long this is but these are the steps I have tried

root@vultr:/home/dpy-2.0-test# python3 -m venv env
The virtual environment was not created successfully because ensurepip is not
available.  On Debian/Ubuntu systems, you need to install the python3-venv
package using the following command.

    apt install python3.8-venv

You may need to use sudo with that command.  After installing the python3-venv
package, recreate your virtual environment.

Failing command: ['/home/dpy-2.0-test/env/bin/python3', '-Im', 'ensurepip', '--u                                                                                                                                                             pgrade', '--default-pip']

root@vultr:/home/dpy-2.0-test# apt install python3.8-venv
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
  python3.8-venv
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 5,456 B of archives.
After this operation, 27.6 kB of additional disk space will be used.
Ign:1 http://us.archive.ubuntu.com/ubuntu groovy-updates/universe amd64 python3.8-venv amd64 3.8.10-0ubuntu1~20.10.1
Err:1 http://us.archive.ubuntu.com/ubuntu groovy-updates/universe amd64 python3.8-venv amd64 3.8.10-0ubuntu1~20.10.1
  404  Not Found [IP: 91.189.91.39 80]
E: Failed to fetch http://us.archive.ubuntu.com/ubuntu/pool/universe/p/python3.8/python3.8-venv_3.8.10-0ubuntu1~20.10.1_amd64.deb  404  Not Found [IP: 91.189.91.39 80]
E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?
root@vultr:/home/dpy-2.0-test#

[trail sapphire]

the text you quoted said that it will work to upgrade from 20.04 LTS -> 22.04 LTS as they are both LTS
just note that your Vultr instance is on 20.10, which is not a LTS release, so i don't know if that will be possible in that case (but i think it should be)

[trail sapphire]

try a apt update first

[fickle granite]

try "apt update" and then re-do the "apt install"

[fickle granite]

d'oh

[grizzled bay]

I'm getting this error when I try thatbash root@vultr:~# apt update Ign:1 http://us.archive.ubuntu.com/ubuntu groovy InRelease Ign:2 http://us.archive.ubuntu.com/ubuntu groovy-updates InRelease Ign:3 http://us.archive.ubuntu.com/ubuntu groovy-backports InRelease Ign:4 http://us.archive.ubuntu.com/ubuntu groovy-security InRelease Err:5 http://us.archive.ubuntu.com/ubuntu groovy Release 404 Not Found [IP: 91.189.91.39 80] Err:6 http://us.archive.ubuntu.com/ubuntu groovy-updates Release 404 Not Found [IP: 91.189.91.39 80] Err:7 http://us.archive.ubuntu.com/ubuntu groovy-backports Release 404 Not Found [IP: 91.189.91.39 80] Err:8 http://us.archive.ubuntu.com/ubuntu groovy-security Release 404 Not Found [IP: 91.189.91.39 80] Reading package lists... Done E: The repository 'http://us.archive.ubuntu.com/ubuntu groovy Release' no longer has a Release file. N: Updating from such a repository can't be done securely, and is therefore disa bled by default. N: See apt-secure(8) manpage for repository creation and user configuration deta ils. E: The repository 'http://us.archive.ubuntu.com/ubuntu groovy-updates Release' n o longer has a Release file. N: Updating from such a repository can't be done securely, and is therefore disa bled by default. N: See apt-secure(8) manpage for repository creation and user configuration deta ils. E: The repository 'http://us.archive.ubuntu.com/ubuntu groovy-backports Release' no longer has a Release file. N: Updating from such a repository can't be done securely, and is therefore disa bled by default. N: See apt-secure(8) manpage for repository creation and user configuration deta ils. E: The repository 'http://us.archive.ubuntu.com/ubuntu groovy-security Release' no longer has a Release file. N: Updating from such a repository can't be done securely, and is therefore disa bled by default. N: See apt-secure(8) manpage for repository creation and user configuration deta ils. root@vultr:~# could this be because 20.10 and 21.04 are EoL. I think the next version I can move to is 21.10

[fickle granite]

wouldn't think so

that's pretty odd

https://fridge.ubuntu.com/2021/07/25/ubuntu-20-10-groovy-gorilla-end-of-life-reached-on-july-22-2021/

it says

it will be archived to old-releases.ubuntu.com in the coming weeks.
which might explain what you're seeing

[grizzled bay]

is there a way I can update/upgrade to one of the old releases as a stop gap to move to a supported release

[fickle granite]

I guess, but what'd be the point?

less likely to fail?

if you're going to suffer through an upgrade -- and you will suffer 🤣 -- you might as well upgrade to the newest release you can, to maximize the delay until the next round of pain

can you just back up your data, and buy a fresh machine?

[grizzled bay]

yeah that might be a good idea

[shy igloo]

So the moral of the story: don’t use Ubuntu.

[fickle granite]

that seems harsh

I've been using ubuntu for ages with no problems

I think the morals are more likely to be:

• don't delay upgrades for years; or maybe
• don't use a cloud service that doesn't make upgrades easy

[main olive]

What is unixm

?

[summer trail]

Unix is a family of operating systems that includes pretty much every modern server, desktop, and mobile OS except for Windows.

[dapper widget]

i have ubuntu 20.04 now but i want to use windows too
can i make it without format ubuntu

[fickle granite]

what exactly do you mean by "format"? I assume you're worried about losing all the data that's on your disk

I expect you will have to run a program called "format" (or some fancy GUI will do that for you) but it might be able to preserve all your existing data.

[dapper widget]

that's about my data, programs accounts etc.

[fickle granite]

of course

[dapper widget]

i used wrong word

format means reinstall

erase all data

i guess i have to backup my all data and reinstall

[fickle granite]

not necessarily

that's what I've been trying to tell you

it's going to be tricky but https://www.google.com/search?q=ubuntu+resize+partition should point you in the right direction (as well as pointing you in six different wrong directions)

[dapper widget]

thank you so much

i got my answer

[summer trail]

If you have another computer or hard drive available, you should definitely back up your data before you start, in case something goes wrong

[forest adder]

ъ

[main olive]

Ubuntu mate > ubuntu, debian and linux mint

[dapper widget]

i want to play game on windows that's why i want

i guess i cant play on virtual machine

[trail sapphire]

you probably can but i wouldn't recommend it, it's usually slow unless you have a special setup that allows the vm to get direct hardware access to the graphics card

[broken forum]

Steam works pretty well in GNU/Linux with many games. Is there a specific game that wouldn't work?

[dapper widget]

EAC does not support for many online games

especially rust

[dapper widget]

then i have to dual boot

ty

[broken forum]

Ah fair enough.
(although, given how spyware-like EAC sounds, it's not necessarily a bad thing for me :p )

[dapper widget]

I think you're right too

fuck that anticheats

[dapper widget]

dualboot will give more performance i think

ty btw

[ember quiver]

Dual booting is hell in my opinion, but probably the best solution if you don't have enough horsepower available. I prefer to run Windows as the native OS to avoid annoying hardware issues, etc. but between WSL2 and QEMU VMs I effectively live in Linux. This way I can quickly and seamlessly switch between Windows and different Linux VMs. It helps that I have a dedicated Linux server to run VMs on QEMU though.

The main problems with dual-booting are 1) if you need to use Windows daily, rebooting that often is annoying and 2) Windows is aggressive about taking over your partitions so it's sometimes a pain to get to place nice with Linux.

[fickle granite]

I haven't dual-booted in years. I'd probably just install Windows, and use some cloud service for my Linux jones

[shy igloo]

I've been able to dual boot arch and windows

10

so far things have been fine

although i made a whole other partition for linux

instead of splitting one

[narrow elk]

guys

im a beginner

could you help me with the scripts

[trail sapphire]

we are not automated bots to complete tasks for you 🤖

[trail sapphire]

!rule 8 because this looks like education and other people shouldn't do the work for you, then you wouldn't learn

[shy yoke]

Rules

8. Do not help with ongoing exams. When helping with homework, help people learn how to do the assignment without doing it for them.

[spice gale]

can someone help me open an image in terminal?

[paper moon]

I do think you should attempt to define what you would like to do a bit more. As the current answer would be xdg-open image-file.png

Yeah that said "no such file or directory". maybe double check that you had the path right to the image.

[spice gale]

ok

[paper moon]

I love using the TAB button to verify that i wrote correctly as it will autocomplete if it is correct.

[spice gale]

is it better to only go through one file?

[paper moon]

My guess would be that the files part of your directory structure /home/user/file/Pictures/Wallpapers is wrong.

So you could try to write ~/Pic<TAB>/Wall<TAB>/untit<TAB>.

Something like that should be helpful to get where you are going.

[spice gale]

does caps matter?

[paper moon]

Yes.

As you can have both a pictures folder and a Pictures folder.

So you need to define which it is with capitalization.

[spice gale]

what about .jpg , .png , .gif ,or .jpeg?

[paper moon]

Depends on the file. But again you can have both a picture.jpg and a picture.jpeg file in the same folder.

[spice gale]

ok

I FIGURED IT OUT

ITS THE file:///

YYYYYEEEEEEESSSSSSSSS

I DID IT