Forem: Ömer Gülen

URL shortener to voice for Clubhouse 👋

Ömer Gülen — Wed, 10 Feb 2021 08:21:06 +0000

URL shortener to voice for Clubhouse - shorten.club

Yes, yes, yes, and yes, I know it doesn't convert links into the voice.

shorten.club converts links or note buckets into a readable 6-digit number, but why?

Background - What is Clubhouse?

In short: Clubhouse is an audio-based social media app. The company describes itself as "a new type of social product based on a voice that allows people everywhere to talk, tell stories, develop ideas, deepen friendships, and meet interesting new people around the world."
Basically, you can jump in and out of different chats, on different subjects, in something akin to a live, free-flowing podcast. You can simply listen or choose to throw in your thoughts. Imagine a cocktail party or, clubhouse.

Description is taken from Mashable

What can't we do in Clubhouse?

Now, you all know what you can do with Clubhouse, it's time to focus on what can't you do in Clubhouse. Clubhouse is a great platform to discuss ideas, contact your friends, organize brainstorming sessions or ask questions to known-people, yet it is very limited on other things such as resource sharing such as links, notes, buckets of ideas, movies, songs, etc.

Why can't I use other URL shorteners?

Whether you are on desktop or mobile, URL shorteners works pretty well, but when it comes to sharing the shortened URL by voice/by spelling it, it can easily become a nightmare. Since Clubhouse is a voice-driven platform, you would want your shortened URL to be pronounced easily which is not possible with a mix of uppercase/lowercase letters and digits. On top of it, Clubhouse is an international platform, there are people from everywhere which means different languages and accents, etc. Even if people speak the same language, it is hard to spell a shortened URL one by one.

How did you solve it?

As you may know every bank on the planet sends 6-8 digit numbers as SMS when it needs your authentication or 2FA works the same way, or quiz platform Kahoot.it gives 8-digit numbers to join a session. These all inspired us to make a URL shortener that converts buckets/lists into 6-digit numbers. In that way, you can easily gather the PIN and access the resources with it.

Currently, you can create buckets of links or notes. You can add, update, and delete those resources even after you share them with your audience.

Examples

A bucket of movies to watch: https://shorten.club/665292
A bucket of resources for Saas Startups: https://shorten.club/191626

We would love to hear your feedback. Let us know what you think in the comments below!

How to reset Jenkins Admin user credentials

Ömer Gülen — Sat, 11 Jul 2020 13:00:34 +0000

Several months ago I've created a development environment for my side project iytemenu. Yet, after a little development interval, I became too busy to develop the project and took a break. Then, this week my finals are done and I kinda have a free time to spend for my side projects, but when I came back I faced that I've lost my Jenkins credentials. Resetting the whole Jenkins logs/DB was too costly so I've researched and tried to find a better way. After some digging, I've found a way and I wanted to share with the community while I have the time to do.

Let's get started

NOTE: I think some versions have the Jenkins name as first letter capital. So use Jenkins instead of jenkins if you can't find the lower case one. Apply this in the following steps of the tutorial.

The first step is making an SSH connection into your host.

ssh username@host

Then, we need to edit the Jenkins config file. It is placed under this path:

/var/lib/jenkins/config.xml

To edit this config file I prefer using vi because other editors may not be installed into the host.

vi /var/lib/jenkins/config.xml

Once you've successfully opened the file in your favorite editor, we need to find <useSecurity> tag and set it's value to false.

After setting the value, you can use save and exit from the editor.

Now, we need to restart the Jenkins service.

systemctl restart jenkins

PS: If you are not root you may need sudo for this operation.

After the restart, you can check the status of the service by:

systemctl status jenkins

Now, you can go to the Jenkins dashboard. You will not need any credentials, and the logged in user section will be blank since you are now an Anonymous user.

Then, navigate to People on the left panel and remove old users by following the steps in the screenshots.

PS: Some versions don't remove the old users totally, instead it shows an anonymous user (you) with other users. This should be enough to continue to the next steps if it doesn't start from the beginning and remove all users.

After removing the old users, navigate to Manage Jenkins page from the left panel, and go to the Configure Global Security section by clicking.

Tick enable security
In the Security Realm section, you will;
-- Select Jenkins' own user database
-- Untick the Allow users to sign up
In the Authorization section, you will;
-- Select Logged-in users can do anything
-- Untick the Allow anonymous read access

Your settings should be looking like this:

And do not forget to save the changes.

On successful save, it will redirect you to the Create First Admin User page.

You will set the credentials of your new admin user here.

Once, you've completed the fields you can click to the Create First Admin User button.

Jenkins will redirect you to the Success page if everything goes fine. You can see your new admin user's username or name on the right top corner. And when you try to login with your credentials it won't be a problem anymore.

That's all...

Thanks for reading,

Feel free to reach out to me at omrglen@gmail.com.

I’m open to suggestions & requests for future articles, cya 😃

Node, Express, SSL Certificate: Run HTTPS Server from scratch in 5 steps

Ömer Gülen — Thu, 02 Jan 2020 21:17:06 +0000

Node, Express, SSL Certificate: Run HTTPS Server from scratch in 5 steps

I've decided to write about this tutorial after I struggled while I was coding one of my web apps for a customer. It was a simple chart for the web but it was collecting data on a Telegram Bot. Held the data in MongoDB and prepared a quick API for fetching the data but got many problems on the way and SSL Certificate was one of them.

So in this tutorial, I will go through my mistakes & problems and my solutions to them, if you want to skip straight to the short version, you can continue from here.

In this article I will not mention MongoDB related code or problems.

1. Creating My Basic API Server with Express

In my projects, I prefer creating an npm or yarn environment after creating the project folder.

So, I've done it with the following commands:



mkdir my-project && cd my-project
yarn init

Just spammed Enter after yarn init and created the project environment with default settings.

(I prefer yarn over npm if there are no obstacles to use it.)

Then, I installed express to my project, locally with:



yarn add express

You can also use:



npm install express

Then, I created my single source file index.js and inserted these lines below:



// import express
const express = require('express');

// create new express app and assign it to `app` constant
const app = express();

// server port configuration
const PORT = 8080;

// create a route for the app
app.get('/', (req, res) => {
  res.send('Hello dev.to!');
});

// server starts listening the `PORT`
app.listen(PORT, () => {
  console.log(`Server running at: http://localhost:${PORT}/`);
});

So far, I imported the express package, created an instance of it and assigned it to the app. Set my PORT variable, and created a route for endpoint handling in my API Server and called app.list(PORT, callback()) method to start my server listening on the specified port.

Went back to my terminal and executed the command below in my project directory:



node index.js

which starts my server and logs to the console as below:



Server running at http://localhost:8080/

Then, I switched to my browser and browsed to http://localhost:8080/ and the following page appeared:

So far so good. My app is correctly listening to my port.
Afterwards, I've tested my initial trial works and wanted to test if I can handle more endpoints. So I've just added another route to my code.



app.get('/omergulen', (req, res) => {
  res.send('Hello Omer! Welcome to dev.to!');
});

I expect this to work only when I entered /omergulen endpoint in my browser.

So, I've stopped my running server with Control+C and re-started again, since hot-reloading is not inherent with how I run my app. Switched to my browser and visited the http://localhost:8080/omergulen and it was working, to be sure I re-visited the http://localhost:8080/ and it was also working as expected.

2. Why and how to use middleware with Express?

After my first API server deploys, I switched to my web app project and sent a fetch request to my API endpoint.



fetch('MY_API_URL')
  .then(function (response) {
    console.log(response);
    return response.json();
  })
  .then(...);

Nothing was happening in my DOM, but the console message was frustrating.



Access to fetch at 'MY_API_URL' from origin 'http://localhost:3000' has been blocked by CORS policy: No 'Access-Control-Allow-Origin' header is present on the requested resource. If an opaque response serves your needs, set the request's mode to 'no-cors' to fetch the resource with CORS disabled.

App.js:34 Cross-Origin Read Blocking (CORB) blocked cross-origin response MY_API_URL with MIME type application/json. See https://www.chromestatus.com/feature/5629709824032768 for more details.

After doing some quick research, I've realized I needed to configure my API Server according to the CORS Policy.

First, I've added added mode: 'cors' to my fetch request:



fetch('MY_API_URL', {
  mode: 'cors'
})
.then(function (response) {
  console.log(response);
  return response.json();
})
.then(...);

It was alone no use to my problem. Then, I've added my cors middleware to my API server with only two lines actually.

After installing cors package with:



yarn add cors

I just added these lines to my code:



// import `cors` package
const cors = require('cors');

// use middleware
app.use(cors());

And after running with these configurations, my problem was solved, for now.

3. How to serve Express API Server as HTTPS?

To deploy, I moved my project to my VPS and redirected my my_api_url domain to this VPS. In that way I've put a small layer of abstraction to my server IP, Also, I wouldn't need to type my IP everywhere instead I could use my own domain with fancy subdomains like api.omergulen.com.

In this step, I first tried to deploy it without certification on HTTP.



[blocked] The page at 'https://my_web_app'  was loaded over HTTPS but ran insecure content from 'http://my_api_url': this content should also be loaded over HTTPS.

Yet, my webserver was being server on Firebase Hosting and it was served as https, sending a request from HTTPS to HTTP is called Mixed Content. And it is not allowed to.

So, I just put s at the beginning of the URL :))

https://my_api_url as you can guess, it didn't work either.



GET https://my_api_url net::ERR_SSL_PROTOCOL_ERROR

Then, after doing focused research I've realized that I needed to create a certificate with a Certificate Authority. Many Certificate Authorities were paid but not Let's Encrypt.

Let’s Encrypt is a free, automated, and open Certificate Authority.

If you have shell access to your server, it suggests you use certbot.

In the certbot website, I chose:

My HTTP website is running None of the above on Ubuntu 16.04 (xenial) which was fitting to my case.

Before starting they want you to be sure to have:

comfort with the command line
and an HTTP website (API Server in my case)
which is online
and serving on HTTP port (80)
which is hosted on a server
which you can access via SSH
with the ability to sudo

Then just apply the steps on the below:

1. SSH into the server

SSH into the server running your HTTP website as a user with sudo privileges.

2. Add Certbot PPA

You'll need to add the Certbot PPA to your list of repositories. To do so, run the following commands on the command line on the machine:



sudo apt-get update &&
sudo apt-get install software-properties-common &&
sudo add-apt-repository universe &&
sudo add-apt-repository ppa:certbot/certbot &&
sudo apt-get update

3. Install Certbot

Run this command on the command line on the machine to install Certbot.



sudo apt-get install certbot

4. Choose how you'd like to run Certbot

Are you ok with temporarily stopping your website?

Yes, my web server is not currently running on this machine.

Stop your web server, then run this command to get a certificate. Certbot will temporarily spin up a webserver on your machine.



sudo certbot certonly --standalone

No, I need to keep my webserver running.

If you have a web server that's already using port 80 and don't want to stop it while Certbot runs, run this command and follow the instructions in the terminal.



sudo certbot certonly --webroot

In this step, you need to insert your domain into the terminal such as dev.to. After that it will check your web server and look for specific files which it will create and in case of success it should print out like that:



Performing the following challenges:
http-01 challenge for my_api_url
Waiting for verification...
Cleaning up challenges

IMPORTANT NOTES:
 - Congratulations! Your certificate and chain have been saved at:
   /etc/letsencrypt/live/my_api_url/fullchain.pem
   Your key file has been saved at:
   /etc/letsencrypt/live/my_api_url/privkey.pem
   Your cert will expire on 2020-04-01. To obtain a new or tweaked
   version of this certificate in the future, simply run certbot
   again. To non-interactively renew *all* of your certificates, run
   "certbot renew"
 - If you like Certbot, please consider supporting our work by:

   Donating to ISRG / Let's Encrypt:   https://letsencrypt.org/donate
   Donating to EFF:                    https://eff.org/donate-le

Important Note:
To use the webroot plugin, your server must be configured to serve files from hidden directories. If /.well-known is treated specially by your webserver configuration, you might need to modify the configuration to ensure that files inside /.well-known/acme-challenge are served by the webserver.

4. Installing signed certificate to Express API Server

You'll need to install your new certificate in the configuration file for your API Server.

First, you need to install and import some modules:



yarn add https



// import packages
const https = require('https');
const fs = require('fs');

// serve the API with signed certificate on 443 (SSL/HTTPS) port
const httpsServer = https.createServer({
  key: fs.readFileSync('/etc/letsencrypt/live/my_api_url/privkey.pem'),
  cert: fs.readFileSync('/etc/letsencrypt/live/my_api_url/fullchain.pem'),
}, app);

httpsServer.listen(443, () => {
    console.log('HTTPS Server running on port 443');
});

If you also want to maintain HTTP requests among the HTTPS requests, you can add the following lines, too:



const http = require('http');

// serve the API on 80 (HTTP) port
const httpServer = http.createServer(app);

httpServer.listen(80, () => {
    console.log('HTTP Server running on port 80');
});

In the end your final API Server code will be something like that:



// import required packages
const express = require('express');
const cors = require('cors');

const https = require('https');
const http = require('http');

const fs = require('fs');


const app = express();
app.use(cors());

// create new express app and save it as "app"
const app = express();
app.use(cors());

// create a route for the app
app.get('/', (req, res) => {
  res.send('Hello dev.to!');
});

// another route
app.get('/omergulen', (req, res) => {
  res.send('Hello Omer! Welcome to dev.to!');
});

// Listen both http & https ports
const httpServer = http.createServer(app);
const httpsServer = https.createServer({
  key: fs.readFileSync('/etc/letsencrypt/live/my_api_url/privkey.pem'),
  cert: fs.readFileSync('/etc/letsencrypt/live/my_api_url/fullchain.pem'),
}, app);

httpServer.listen(80, () => {
    console.log('HTTP Server running on port 80');
});

httpsServer.listen(443, () => {
    console.log('HTTPS Server running on port 443');
});

5. Automatic Renewal and Test of the Certificate

The Certbot packages on your system come with a cron job or systemd timer that will renew your certificates automatically before they expire. You will not need to run Certbot again, unless you change your configuration. You can test automatic renewal for your certificates by running this command:



sudo certbot renew --dry-run

The command to renew certbot is installed in one of the following locations:



/etc/crontab/
/etc/cron.*/*
systemctl list-timers

If you needed to stop your webserver to run Certbot, you'll want to edit the built-in command to add the --pre-hook and --post-hook flags to stop and start your web server automatically. For example, if your webserver is HAProxy, add the following to the certbot renew command:



--pre-hook "service haproxy stop" --post-hook "service haproxy start"

More information is available in the Certbot documentation on renewing certificates.

Confirm that Certbot worked

To confirm that your site is set up properly, visit https://yourwebsite.com/ in your browser and look for the lock icon in the URL bar. If you want to check that you have the top-of-the-line installation, you can head to https://www.ssllabs.com/ssltest/.

Well done! You have come to the end of this long tutorial.

After applying these steps you can finally go to your API Server URL and you should be seeing Hello dev.to!.

Thanks for reading

I hope this tutorial was helpful enough.

You can check my last article here:

How to add system call (syscall) to the kernel, compile and test it?

Ömer Gülen ・ Dec 25 '19

#kernel #syscall #linux #unix

Feel free to reach out to me at omrglen@gmail.com.

I’m opened to suggestions & requests for future articles, cya 😃

Happy New Year! 🥳🥳🥳

How to add system call (syscall) that prints elapsed time of a process with given PID to the kernel and test it?

Ömer Gülen — Fri, 27 Dec 2019 18:18:52 +0000

In my previous tutorial 👇

How to add system call (syscall) to the kernel, compile and test it?

Ömer Gülen ・ Dec 25 '19 ・ 6 min read

#kernel #syscall #linux #unix

I described how to add syscall to the kernel, compile it and test it in detail. As promised, I now would like to describe how to create a syscall that prints out the elapsed time of a process with a given process id(PID).

This task was my Operating Systems Programming Assignment.

In this tutorial, unlike the previous tutorial I will not go over into every detail, but point out the different parts of the process.

After changing my current directory into kernel source files, I create another folder for my new function. You don't need to do it, but I prefer it this way.

mkdir petime && cd petime

I will call my function petime short for process elapsed time.

After I've created a C file which my code goes into.

vim petime.c

In my C code, I used SYSCALL_DEFINE1 Macro from <linux/syscalls.h>. Because I needed to pass one integer argument to the function.

Also, you can use it for different argument counts. While argument count is X, the syntax will be like SYSCALL_DEFINEX.

Example:

// For 2 parameters
SYSCALL_DEFINE2(function_name, int, first_int, int, second_int){
  // code goes here
}

// For 1 parameter
SYSCALL_DEFINE1(function_name, int, first_int){
  // code goes here
}

// For no-parameters
SYSCALL_DEFINE0(function_name){
  // code goes here
}

In addition to this Macro, I used the task_struct struct to get the process' start time and time elapsed since the boot with ktime_get_ns() function.

My C file which contains my syscall:

#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/syscalls.h>

// Used this macro to be able to pass variables to the syscall
// It takes one integer, PID.
SYSCALL_DEFINE1(sys_petime, int, pid)
{
  // Using task_struct struct from sched.h, which has task's (process)
  // various data mostly about scheduling.
  struct task_struct *task;
  u64 start_time, elapsed;

  // Get process with PID.
  task = pid_task(find_vpid(pid), PIDTYPE_PID);
  // If process is not found.
  if (task == NULL)
  {
    // Print the error to the kernel buffer
    printk(KERN_INFO "Cannot find a process with that PID: %d\n", pid);
  }
  else
  {
    // Taken from "taskstats.c" source code (line 240-241).
    // Used the logic behind calculating the delta time in that source.
    // Get current nanoseconds since boot
    start_time = ktime_get_ns();
    // Difference
    elapsed = start_time - task->real_start_time;
    // Print result to the kernel buffer
    printk(KERN_INFO "PID: %d has ELAPSED %llu nanoseconds.", pid, elapsed);
  }

  // If nothing goes wrong, return 0 as successful
  return 0;
}

ktime_get_ns() is like timestamp but it doesn't count from 1 Jan 1970 instead it counts nanoseconds since the Operating System boots.

task->real_start_time is the boot based start time of the process in nanoseconds.

So as a result, the difference of total elapsed time since boot and the boot based start time of a process gives the elapsed time of a process.

Then, I've created Makefile file of my syscall:

vim Makefile

and inserted the default-ish configuration:

obj-y := petime.o

Now, I need to let my general Makefile know about my new folder and function.

Go back to the kernel source folder and edit Makefile:

cd .. && vim Makefile

Find the line with the core-y, it should be similar to this:

core-y      += kernel/ certs/ mm/ fs/ ipc/ security/ crypto/ block/ hello/

hello/ is the folder from our previous tutorial.

I've added my new folder petime to the end of this line. Result should be looking like this:

core-y      += kernel/ certs/ mm/ fs/ ipc/ security/ crypto/ block/ hello/ petime/

After saving the Makefile, we need to add our new syscall to the syscalls tables.

I will describe the tutorial according to 64-bit Operating System, if you want to do it on 32-bit OS, you can go back to my previous tutorial find the details of the following steps for 32-bit OS.

For 64-bit OS, I add my syscall to the end of my file(table):

549 64  petime  __64_sys_sys_petime

I've put sys 2 times, like _sys_sys_ because my function name was already sys_petime but you need to prepend __64_sys_ of the function name in this table, so result becomes __64_sys_sys_petime.

Then, we need to add our function signature to the syscalls header file syscalls.h.

vim include/linux/syscalls.h

Then, add the following line to the end of the document before the #endif statement:

asmlinkage long sys_petime(int);

Now, we can compile it the same as our Hello World example.

After the, compilation we can test it with the following C code:

#include <stdio.h>
#include <linux/kernel.h>
#include <stdlib.h>
#include <sys/syscall.h>
#include <unistd.h>

int main(int argc, char **argv)
{
  // Get args from terminal and call petime (get elapsed time) with that PID.
  long int helloCheck = syscall(549, atoi(argv[1]));
  printf("System call sys_petime returned %ld\n", helloCheck);
  return 0;
}

And the result is such that:

Please comment below your questions and feedback.

How to add system call (syscall) to the kernel, compile and test it?

Ömer Gülen — Wed, 25 Dec 2019 17:13:46 +0000

One of my example syscall methods will print Hello world to the kernel buffer.

The second one gets PID of a process and prints the elapsed time of it to the kernel buffer in my other tutorial:

How to add system call (syscall) that prints elapsed time of a process with given PID to the kernel and test it?

Ömer Gülen ・ Dec 27 '19

#kernel #unix #linux #tutorial

1. Setup environment

I prefer to get root and apply the steps as root to prevent getting permission failures on the way. If you don't want to apply the steps as a root, you can just use sudo for root required commands.



su -

will switch to root user after entering the password of the root. Then, first you can check the active kernel version of your OS with:



uname -r

This prints out 4.19.0-6-amd64 in my case. Then, we need to get the source of a kernel. I will use slightly newer version (4.20.1) from my version with following wget command.



wget https://mirrors.edge.kernel.org/pub/linux/kernel/v4.x/linux-4.20.1.tar.xz

This will download the source of 4.20.1. It may take a while, like 2-3 minutes depending on your internet speed.

Extract the compressed kernel code with



tar -xvf linux-4.20.1.tar.xz

It will create a folder named linux-4.20.1.tar.xz and extract the compressed code into that folder.

Now we will change our directory to new kernel code.



cd linux-4.20.1

2. Add "Hello world" syscall to kernel

I prefer creating new folder for my own stuff while adding a new syscall.



mkdir hello && cd hello

After that, I will create a C file for my syscall implementation. I prefer vim hello.c to create and edit the file and insert following C code.



#include <linux/kernel.h>

asmlinkage long sys_hello(void) 
{
        //printk prints to the kernel’s log file.
        printk("Hello world\n");
        return 0;
}

We need to create a Makefile in the hello directory.



vim Makefile

and then insert this:


 := hello.o

Then, go to the parent directory (kernel source main directory):



cd ..

We need to add our new syscall directory to Makefile, in this way it will compile our syscall, too. To achieve this, search for core-y in the Makefile then, find the

In vim you can do search with /core-y after pressing ESC.


 += kernel/ mm/ fs/ ipc/ security/ crypto/ block/

``` line and add `hello/` to the end of this line.

As a result, the line should be looking like this:

```core-y

 += kernel/ mm/ fs/ ipc/ security/ crypto/ block/ hello/

Next step is adding the new system call to the system call table.
If you are on a 32-bit system you’ll need to change syscall_32.tbl file.



vim arch/x86/entry/syscalls/syscall_32.tbl

For 64-bit, change syscall_64.tbl.



vim arch/x86/entry/syscalls/syscall_64.tbl

I will continue with my 64-bit OS and my steps will be accordingly.

We need to keep table's structure while adding our syscall. So that we will add our line to the end of the line. My last syscall has number of 547, so that I will use 548, you also should use N+1.

My syscall data:


       64        hello          sys_hello

Example file:



546     ...     ...     ...
547     ...     ...     ...
548     64      hello   sys_hello

Now, we need to add our syscall method signature to syscalls header file which is syscalls.h.



vim include/linux/syscalls.h

Then, add the following line to the end of the document before the #endif statement:



asmlinkage long sys_hello(void);

3. Compiling our kernel

Before starting to compile you need to install a few packages. Type the following commands in your terminal to install required packages:



apt-get install gcc &&
apt-get install libncurses5-dev &&
apt-get install bison &&
apt-get install flex &&
apt-get install libssl-dev &&
apt-get install libelf-dev &&
apt-get update &&
apt-get upgrade &&
apt-get make

Now, you can configure your kernel by using the config menu by executing:



make menuconfig

IMPORTANT NOTE: While entering this command, you might want to maximize your terminal screen. Otherwise, you might get and error and a pop-up screen will not appear.

Once the above command is used to configure the Linux kernel, you will get a pop up window with the list of
menus and you can select the items for the new configuration. If your unfamiliar with the configuration just check for the file systems menu and check whether “ext4” is chosen or not, if not select it and save the configuration.

OR: You can simply execute the following command and use the default configuration.



make defconfig

Now, finally we can compile the kernel with the following command:



sudo make

The compilation took 40 mins to 1 hour on my uni-core VM. As an alternative solution you can increase the core count of your VM in the VM settings and use them to compilation with the -jn parameter. n is the number of cores dedicated to your VM.

In my case, I've used the following ocmmand for 8-core VM which reduced the compilation time to 20 seconds to 1 minute:



make -j8

4. Installing our kernel

After the successful compilation, to install/update kernel use the following command:



make modules_install install

The command will create some files under /boot/ directory and it will automatically make a entry in your grub.cfg.
To check whether it made correct entry, check the files under /boot/ directory . If you have followed the steps without any error you will find the following files in it in addition to others.

5. Testing our syscall in new kernel

Now all we need to do is restart the system:



shutdown -r now

After computer restarts, in grub's advanced options, you can see there are 2 options (without counting the recovery mode options).

Once your computer is up again, you can run the following command the check your kernel version:



uname -r

which prints 4.20.1 in my OS after installing the kernel.

After checking the version of the kernel, we will test our syscall with tiny C program.



vim hello_test.c

Insert the following C code into the hello_test.c file:



#include <stdio.h>
#include <linux/kernel.h>
#include <sys/syscall.h>
#include <unistd.h>

int main()
{
        long int helloCheck = syscall(548);
        printf("System call sys_hello returned %ld\n", helloCheck);
        return 0;
}

Then compile and run it:



gcc hello_test.c -o hello.o && ./hello.o

it prints System call sys_hello returned 0 if there is no problem on the execution and now we can check the kernel log buffer to see if our Hello world is there with the command below:



dmesg

It will print tons of line and in the end we should be seeing our Hello world.



[    ........] ...
[    ........] ...
[    ........] ...
[    2.858022] IPv6: ADDRCONF(NETDEV_UP): enp0s3: link is not ready
[    2.858144] ip (1443) used greatest stack depth: 12424 bytes left
[    2.860434] Adding 10483708k swap on /dev/sda5.  Priority:-2 extents:1 across:10483708k 
[    3.391290] hrtimer: interrupt took 4499049 ns
[    4.887969] e1000: enp0s3 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[    4.888216] IPv6: ADDRCONF(NETDEV_CHANGE): enp0s3: link becomes ready
[  306.497742] Hello world

We can observe that our Hello world is printed to kernel log buffer. So it worked!

Error - Fix

Debian Certificate Problem

On my first trial I have faced with an error such below:



make[1]: *** No rule to make target 'debian/certs/debian-uefi-certs.pem', needed by 'certs/x509_certificate_list'.  Stop.
make: *** [Makefile:1055: certs] Error 2

- SOLUTION -

I've found the solution of this problem in the debian.org's bug report section. The issue was reported by Heinrich Schuchardt as the following:

The Debian kernels are currently built with CONFIG_SYSTEM_TRUSTED_KEYS="debian/certs benh@debian.org.cert.pem"
This was introduced with this https://alioth-lists-archive.debian.net/pipermail/kernel-svn-changes/2016-April/022904.html.

We are now two years after Ben's contribution and we are still not using kernel module signing (CONFIG_MODULE_SIG is not set in config).

As there is no need for the kernel trusting Ben's certificate, please, remove the setting.

Best regards

Heinrich Schuchardt

As a result, I've cleared the CONFIG_SYSTEM_TRUSTED_KEYS config in the configuration file.

In linux-4.20.1 (which is my kernel source directory), there is a file named .config which stores the configuration of the kernel. With the following command I opened it:



vim .config

Then searched for CONFIG_SYSTEM_TRUSTED_KEYS and changed



CONFIG_SYSTEM_TRUSTED_KEYS="debian/certs benh@debian.org.cert.pem"

into this:



CONFIG_SYSTEM_TRUSTED_KEYS=""

then, saved and closed the file, and tried to compile again and there was no issue.

I will publish the second example in another article since this one is already so long.

Please comment below your questions and feedback.

How to add system call (syscall) that prints elapsed time of a process with given PID to the kernel and test it?

Ömer Gülen ・ Dec 27 '19

#kernel #unix #linux #tutorial