Virtualization: Testing Proxmox and ESXi

After evaluating my home environment I thought it was time to repurpose the home server into a dedicated host for virtual machines. Ideally one of these will run my Minecraft server and as time goes on I might find other uses that I can launch additional VMs for. 

However instead of running a Type 2 Hypervisor, such as a Oracle Virtualbox on top of a Window 11 machine, I wanted to see if I could get a Type 1 Hypervisor to run bare metal. I have some experience working with VMware workstation and vSphere, but not installing one. So this is a documentation of my journey, more than a  guide. Note that VMware ESXi did not work on my Dell Optiplex, because it did not find the ethernet adapter. So then I tried Proxmox instead.

Downloading the .iso for ESXi

1. I started downloading the software which required me to register on Broadcoms website. In the top corner I clicked on register and followed a simple registration process.

2. After login in I went to the free downloads portal and looked for "VMware vSphere Hypervisor".

3. There I clicked my way to version 8.0U3e. I had to accept an agreement before I could download the .iso with the cloud icon. The download itself was just about 600mb and it is already licenced.

Installation issues with the .iso for ESXi

According to guides online it is as simple as booting from the .iso.

So I was going to put that to the test, but first - backup the current installation of course!

I wrote the .iso to a USB and booted from it, so far so good, but in the middle of the installation I ran into a common problem, the ethernet adapter was not supported by the OS. So here the journey ended.

Going with Proxmox instead.

I started by getting the .iso and wrote it to a USB with Rufus.

Then I booted my server from the .ISO and let it install. One learning I took away from this was that I had to put AHCI mode instead of the RAID, this is because the Proxmox installation didn't find the harddrive.

After a while I was faced with a login screen on the computer and an IP. This was my cue to get on my usual computer and browse into the IP.

The username is defaulted to "root", password is the password you set up in the installation.

You get into a view that is segmented into "datacenter", "username", "localnetwork", "local (username)" and "local-lvm (username)".

Creating my first test VM

My first takeaway here was to upload the .iso that you want into the segment called "local (username)" and ISO-images. It is a simple upload.

Next step is to press the segment just called "username" and do the create VM button.

You fill in name, ID, select cores and an amount of RAM that will be sufficient.

I had some easily fixed issues in my first attempts, too many cores and too much ram.

After fixing that I could start it with no issues. Surprisingly enough even the network worked.

I loaded in a Win11 x64 .iso with 8gb om ram and a couple of cores to do my first testing.

After that it was a usual Windows setup.

Updating the Proxmox

You can either login on the computer directly, or use SSH root@ipaddress with your password.

Then do the usual apt-get upgrade -y && apt-get update -y and you will update the OS.

Fiction: The Descendant, a short story

The Descendant

A short story, by Erik Engström, 2025-09-17

A couple of pigeons were strutting around on the ground as a cool breeze swept across the plaza. It was early spring when I decided to have my coffee outdoors for the first time this year.

One of the birds stood out against the others, it was white and slightly spotted. They were living in the moment, scouring for crumbles that had been left on the ground from earlier guests.

After a while they took off flying towards another part of the city, and I was once again alone with myself and my thoughts.

Glancing at my wristwatch, I knew it was time, they would be here any minute so I took my cup and went inside and found a secluded corner in the café.

As expected, and as they usually are, they were on time. Temporal agents were coming through the door and gracefully coming over to my corner where they sat down, face to face with me. Apart from their wildly different clothes, they all shared the same face. My face. Or maybe I carried their face.

When you work for The Time Council you are sometimes provided a new identity. Instead of looking like a nobody, they make you look like everybody. Should one agent be caught, then the opposition would simply think that they have captured the only agent. Perhaps they would even dismiss the person as a regular time traveler as there would be photos of the same person in multiple points in time.

One of them handed me a letter addressed to Ramel Loveworth, which read:

“Mr. Loveworth, you have been permanently dismissed from the Council for interfering with the timeline. As an agent you are there to observe and not be observed. Your actions have costed us greatly. The rules are clear and despite this you intervened. Hand in your device to agent Sidney Bardlet and continue your civilian days in time alone from this point onwards.”

I stared at the paper, deep down I knew that I had broken the rules but even now I was also convinced that I did the right thing. Countless lives had been saved from a terrible fate no one knew about now. A brief moment followed and I removed the wristwatch and put it in the outstretched hand of agent Bardlet.

Without a word they got up and left as proudly as they had entered, leaving me alone in the corner again. My eyes drifted to my wrist, realizing what it all had costed me.

As we had been to every point in the past and the future, we also developed a near perfect understanding of cause and effect. What people sometimes attributed to random occurrences or accidents we always knew how to trace back to the source.

What is difficult with having access to any point in the timeline is not that you don't know what to change, because you will not run out of such things.

It is knowing that you are able to but shouldn't change it, and now I know the price for doing so.

 

Fiction: The Returned, a short story

The Returned ❄️

A short story, by Erik Engström, 2025-10-29

What is scarier than entering cryo sleep? Well, that would have to be waking up from it. That's at least what the scientist told herself when she was looking down at the cryo chamber laying before her in the subterranean vault. A plaque with the name Rose Monime was the only thing giving the metal box a purpose to the unknowing eye. 

To the scientist this was a purpose greater than herself, the box had been there before her, and it had originally been intended to be there after her. In fact, no one at the Ark facility alive today had personally seen Rose. She had been frozen in time since a previous era, when cryo sleep, cloning and factory made humans were illegal, and at best instantly deadly. A horrible fate awaited most brave souls participating in the early, secret, trials. Rose had been one of the first to successfully enter cryo sleep.

Every day the scientists painstakingly ran checks and documented the vital signs provided by the now dated technology. Every day also meant a world that became more different from the world that Rose had known, should she ever return. 

And the day eventually came when the union between cutting edge science and luck were put to the test. With specially crafted microwave instruments and state of the art quantum particle analysis the scientists brought her back from this undefined state of being, neither alive or dead.

The pain of thawing after decades or centuries was indescribable. Every cell waking up from deep slumber was making it clear that this was against the wishes of any creator and the laws of nature.

The scientist looked at Rose as she became conscious after an almost 24 hour long recovery process. Her old memories, being electric impulses, were gone and new impressions flooded an almost factory reset brain. No one really knew what to expect either. In the same room, cryo chambers stood empty like open caskets gathering dust, indicating that there once had been others. She was the first one to not only make the journey but to also make it back, but some journeys make the traveler a different person. 

Rose's bewildered gaze met the scientist's and for a moment it looked like the two women shared a silent understanding. She spoke to Rose, but the woman in the cryo chamber said nothing. Her eyes had lost their human quality and she was just blankly looking at the scientist as if there was no one inside. As if the human spirit had left the body, leaving only a primitive animal behind.

Fiction: The Exception, a short story

A chaotic scene unfolded, the phones had been ringing nonstop for the past hours.

There was some kind of breach in the firewall, programs that weren't supposed to interact with one another had suddenly been interconnected.

Milvago Tintie was one of the latest workers that had been hired to upgrade the systems in the company some week ago. While waiting for the interview he had met Ella Rove, a girl that had been at the company for a year or so.

There was something about Ella that sparked a thought in Milvago, something he hadn't encountered before. A door deep within had opened and simple logic was not enough to explain this strange algorithm.

A few weeks passed by at the office and he recognized how different he was from the other colleagues working there. They never asked him to have coffee together and others thought that it was strange that he never ate lunch while sitting alone at the table in the lunch area.

It was not the first time he had experienced this kind of exclusion because of who he was. So it came as a great surprise when Ella one day sat down at this table. They had a lively conversation about everything and nothing, getting so caught up that they lost track of the allotted three quarter of an hour long lunch break. 

Getting back late from the lunch that day he was still focused on her and the regular work tasks soon received less attention. He looked forward to the next moment they would speak, and the next moment after that. 

And so the weeks kept churning, they would meet often and spend time talking and one day he simply came to the conclusion that he was in love with her. Caught up in these thoughts he decided he would visit her one day, bring a gift and ask her if she felt the same way.

It was a Thursday before a public holiday when he dressed in his best shirt and suitable clothes.
There was a bit of rain, so he brought an umbrella and got into the taxi.
It arrived just outside of the building and he stepped out, looking at the top floor.
"This is it" he said as he opened the door to the lobby and headed for the elevator.

He rang the bell at her apartment. He smiled wide and waited with anticipation.
This was the most important occasion in a long time, perhaps in his whole existence.

She opened the door carefully, smiled at first when she saw him. Her eyes moved towards the gift he carried and the smile faded somewhat as if she already knew his intentions. She looked him in the eye and a brief silence followed.

He started speaking but she cut him off gently but decisively.

"We can't be together Milvago" she said while the rain trickled down his face, "you know they would never allow it". 

A tear traveled down her cheek and she now smiled a broken smile as tears started forming.

This was an emotion he could never show. It was only reserved for humans. 

Fiction: The Castling, a short story

The Castling ♟️

A short story, by Erik Engström, 2025-09-17

Change was in the air. But there was a storm brewing far off in the distance as well. The days were shorter and the first falling leaves were a sign of the change of seasons. 

A man was walking along the seashore as the wind swept in over dried seagrass and crushed seashells.  

He stopped by the edge of the water, where the ocean met the grayish sand. His boots sank into the sand as he looked at a puddle to see his own reflection. He had not yet grown used to the face looking back at him. 

Somewhere out there he knew a stranger was in the same predicament. Someone looking into the reflection, perhaps as confused as himself. Maybe someone was trying to find out the truth at this very moment. Even if it was possible, would he want to transfer back to an aging body, stuck in a monotonous life? Whoever had taken his place surely wasn’t the winner.  

Judging by the name driver license in the wallet, he had somehow taken over the body of Anton Fossa, a young banker living in an expensive apartment in the city center. He himself was registered as Cody Poe Belmont, a reserved immigrant living a simple life as a priest without really leaving much of a paper trail. 

A distant sound at sea pulled Cody back from deep thought. Ever since the castling of souls had occurred, one thought had kept him awake at night, keeping him in fear. 

The life that he had ended up in had everything most people could ever dream of. He was young again; the gaze of people wandered toward him with attraction and his job paid well. Yet the question lingered echoing in the back of his head at all times.  

If the other person could find him, after all Anton was probably quite established, why did Anton never contact him? Had the stranger found hidden potential in his life that he himself had missed all these years? 

Later the same evening, as Cody was looking through the library in his new apartment, he pulled out a book that looked different from the other business, sociology and psychology books. A leather-bound book that gave him chills merely by touching it with his fingers. 

He realized that this was a book of great evil, and the illusion that he desperately had held onto that the whole situation was an accident was dispelled, as the title taunted him with jagged letters. 

“Satanic rituals, sacrifice and the transference of souls” 

Hardware: Replacing thermal paste on Framework Laptop 13

Hello,

This will be a shorter post covering how I dealt with the overheating Framework Laptop 13.

Essentially I had bought the laptop pre-owned, and thus it is difficult to get a complete picture of how worn the computer was at the time of the purchase.

However, you can run simple diagnostic tools for your CPU, memory, SSD and battery to get an estimate.

What I noticed, essentially from the beginning, was how was the PC would become even from a medium CPU load. Trying a lot of things from a software point of view I decided to look over the hardware.

After a fruitful dialogue with the Framework support I decided to give it a go and replace the thermal paste.

This is how I went about it:

1. Unscrew the bottom screws, they are captive so they remain in the bottom part.

2. Carefully lift up the keyboard cover from the lower right corner, don't pull too much or you will hurt the touchpad cable.

3. Gently lift the connector for the touchpad cable from the motherboard (there should be a tab, you pull straight up)

4. With the cover off, you unscrew the fan using the T5 screw still, also remove the connector for the fan. One screw I couldn't get fully loose but it worked like a captive one.

5. Unscrew the heatsink, carefully remove it together with the fan. The heatsink has captive screws.

6. You will notice a lot of thermal paste probably, in my case it has solidified to such a degree that alcohol didn't remove it, I had to carefully scrape with away with the spudger and the cotton swab, together with the alcohol.

7. When it was dried I applied generously with thermal paste. Screwed everything back, plugged back the touchpad cable, put the keyboard cover back and put the screws back in the bottom part of the PC.

The computer showed a lower temperature, but most importantly I could no longer feel the burning heat.

Kali Linux: Using the Framework fingerprint reader

As a Windows user I am also using facial recognition and fingerprint scanning as means of unlocking the operative system. In the same spirit I wanted to connect it to my Kali Linux dual boot.

A quick look at Bing gave me the following information. 

Start by opening the terminal, use Ctrl + Alt + T for example, then run each of these commands.

sudo apt install fprintd libpam-fprintd

sudo pam-auth-update

This will open a dialogue box, where you can select the "fingerprint authentication" option.

fprintd-enroll

Now you will touch the fingerprint sensor repeatedly until it says complete.

fprintd-verify

This let's you confirm functionality.

sudo fwupdmgr refresh

sudo fwupdmgr get-updates

sudo fwupdmgr update

Finally, lock your screen with Windows-key + L. It should allow you to test the fingerprint reader.

Hardware: Upgrading Framework Laptop 13

At this stage I have had my Framework Laptop 13 for almost a year I think, and by now I have accepted it for its strengths and shortcomings.

A brief overview of what I think about the machine to this day is as follows. It is a great machine, it looks and feels really good and the keyboard is amazing. The main issue I have is that the i5 1340p processor gets very hot. After talking to the support there are a few things I can do about it to start with. New thermal paste, cleaning it with compressed air and of course replacing the fan module (which I consider the last step). 

Running a battery report I found out that I am running the 55wh-battery, and it is down by approximately 20% battery health. Usually I run the laptop plugged in and capped at a maximum charge of 90%, so it is rare that I notice this. But my next purchase would probably be the slightly bigger battery.

So what did I upgrade in my first upgrade session? I got a new bezel that is green and translucent, an additional USB-C expansion card (as I never use HDMI, which I got when I bought it on Ebay), and a keyboard in the second generation, with my own native language. Which was sorely missed, due to the lack of pipe symbols. I am writing PowerShell scripts every so often, and a pipe symbol is essential for that.

The upgrades were easy, when you replace the Bezel, make sure to flip down the screen towards the table complete, start removing the bezel from the top right corner and be careful of the cable in the bottom left corner. Follow the manual that they provide on the website.

The expansion card replacement is the easiest part, just unlock the slot, remove the old one and insert the new one. Plug and play.

The keyboard was the trickiest, for this you should absolutely follow the guide and make sure to unscrew the right screws and also place them back where they belong. It takes a lot of time and is tedious, but not difficult.

Windows: Process Monitor

This post will cover some basics of Process Monitor, a program that is part of the Sysinternals Suite, which is a set of tools that was created by Mark Russinovich.

To download just the tool click here or if you want the entire Sysinternals Suite (170 mb) press here.

Writing the draft for this post I felt the need to research the difference between Process Monitor and Process Explorer, which can be generalized as the following:

Process Monitor (Procmon64)

Captures live information about the processes on your system, how processes interact with the file system, registry and network. Much like Wireshark you can open and save capture logs, which you then can filter.

Process Explorer 

Displays other aspects of the processes, such as handles, DLL:s, memory usage and resources. It shows how processes are related to each other as parent or child process in a tree view. Process Explorer can also be used as a "task manager on steroids" and replace task manager as the default app.

Using Procmon64

This section will explain how to create a basic capture and how to save it with an applied filter. Process Monitor starts gather information as soon as you start it so should you want to stop it press Ctrl +E. To clear the view in the window press Ctrl + X.

When you are ready to start the program that you are troubleshooting or analyzing, start "capture event", with Ctrl + E.

When you have reach the point where you want to stop, stop the capture.

Next you might want to apply filters, use Ctrl + L to get into the filter view.

As an example I took powershell.exe.

Select filter "program name" and "contains", write the name of the program in the textbox (for my example I just wrote powershell). Take add and apply, and then it filters out the data in the main window for you.

Maybe you want to filter out only registry queries for example, then you can add the additional filter "operation", "contains" and "RegQuery".

To save your log use Ctrl + S, it natively saves it in the .pml format but you can choose .xml and .csv as well.

Cybersecurity: Living of the land

Living of the land means to use resources that are already on the machine, as opposed to bringing external or homemade tools to the target device for example. The LOTL technique uses native tools which can make intrusion detection difficult as they leave minimal footprints and often are considered trusted. 

These native binaries can be used to break out of restricted shells and here are some examples that exist for both Unix (GTFOBins) and for Windows (LOLBAS).

Recently I watched a video from John Hammond that covered this cybersecurity technique. The YouTube video was in turn largely based of the research from Grzegorz Tworek.

I had previously seen how replacing utilman.exe with cmd.exe could grant administrator level command prompt from outside Windows before and in a similar fashion the video showed another replacement action. Grzegorz example that John is covering in the video shows how the native program tpmtool spawns cmd.exe which in turn calls for logman.exe in an unsafe way. The way this is executed is similar to "binary planting" and "DLL hijacking", the computer is simply tricked to execute something else than intended.

This is how he demonstrated the technique:

With Procmon64 actively gathering information about processes the command line tool "tpmtool drivertracing stop" was then run in cmd which resulted in an error. He then filtered out results for tpmtool which he was investigating and ran the command in cmd again. By doing this he could see that the tpmtool opens another cmd.exe window in a "process create" operation. 

Another filter for cmd.exe was also applied, which showed both a process create and process start, that according to the event properties starts yet another program in cmd, named logman and it is this program that could theoretically be replaced with something else. 

The reason why this exploit works is because logman.exe has no directory specified, you are essentially telling the computer to see if there is a program with that name where it is currently looking. If another exe-file is placed in the working directory of the initial cmd ("C:\Users\username") and is renamed to logman.exe, then the next time "tpmtool drivertracing stop" runs in cmd it calls on the fake logman.exe file.

For more information on how to use Process Monitor, I have written another post covering some basics here.

PowerShell: New .ps1 in context menu

The context menu is the menu that shows up when you right-click in your explorer window. This is where you can create new files, folders and shortcuts.

So as I am a bit of a PowerShell nerd I thought it would be handy to have the .ps1 file in the context menu where you create new files as well. I found a reliable source at Winaero and a reg file created by Sergey Tkachenko (press the click here to download) to get the zip folder.

Compiling the research this is essentially the steps you need to take:

1. Open regedit as administrator and go to HKEY_CLASSES_ROOT and find the key for .ps1 files.

2. Change the value of "(Default)" to ps1legacy. Mine was set to ps1_file_auto

3. Create a new key (folder) under the .ps1 extension in the left pane. Name it ShellNew and within it create a new string value called NullFile and make sure the data is empty.

4. On the top level of HKEY_CLASSES_ROOT, create a new key for it named ps1legacy, give the "(Default)" the data Windows PowerShell Script. Create a new string value in the same place with the name FriendlyTypeName, give it the same data as the other string value.

Here is the result! 😊

Lastly, let's explore how we with some lines of PowerShell code could create the same result. This code snippet is boiled down to the essentials on purpose for simple step-through, reviewing and modification.

# Run as admin #

Set-ItemProperty -Path "Registry::HKEY_CLASSES_ROOT\.ps1" -Name "(Default)" -Value "ps1legacy"

New-Item -Path "Registry::HKEY_CLASSES_ROOT\.ps1\ShellNew" -Force

Set-ItemProperty -Path "Registry::HKEY_CLASSES_ROOT\.ps1\ShellNew" -Name "NullFile" -Value ""

New-Item -Path "Registry::HKEY_CLASSES_ROOT\ps1legacy" -Force

Set-ItemProperty -Path "Registry::HKEY_CLASSES_ROOT\ps1legacy" -Name "(Default)" -Value "Windows PowerShell Script"

Set-ItemProperty -Path "Registry::HKEY_CLASSES_ROOT\ps1legacy" -Name "FriendlyTypeName" -Value "Windows PowerShell Script"

PowerShell: Working with CSV files

I have previously tried to setup a way of working with databases with PowerShell. Recently I came across a way to use a simple CSV file to store data and then fetch it from PowerShell. This post will cover a few pointers I picked up along the way as well.

The first thing I did was to create the CSV file, which works as a rudimentary database for the purpose. I started by creating the datasheet in Excel and converting it to a CSV file. Perhaps you even have a complete file from the beginning that you got from some statistics page for example.

How you can work with a CSV file

The code was written on one computer first, where the file got a comma delimiter. Then I rewrote the code on another computer which had semicolon as delimiter, this caused trouble for my script. Because PowerShell defaults to the comma delimiter, so you need to specify that in your script when you run the Import-CSV cmdlet. Store the data in a variable and set the delimiter, that way you can return later and troubleshooting becomes easier if it has trouble reading the data. In my case it detected the file, but I got no indication what was wrong at first all that happened was that PowerShell didn't fetch the data from the CSV, most likely because it could find the individual columns.

Here is what it can look like if you have the script and the CSV in the same root folder. Look into the CSV file to see which delimiter to use.

$Imported_CSV = Import-CSV -path .\database.csv -delimiter ","

With this simple line you now have read in the data into your variable and you can start work with it. Some inspiration for this part come from this video. To count the entries, simply run $Imported_CSV.Count and see it as representing the amount of rows. With this knowledge you can also inspect individual lines by calling on their index, remember to count from 0 which is the first row of content, not the header. Using classic PowerShell index enumeration, [-1] is the last row. You can thus cook up something like this:

($Imported_CSV[-1].Name) to get the data in the name column for the last object. This requires you to know the column name in advance. To find out the column names you could run the following code:

$Imported_CSV[0].Psobject.properties.name

With the index, I could also change a specific datapoint. When I then call on the main variable again, it lists the changed datapoint. Using the Export-CSV command I could then also save the change to the file.

($Imported_CSV[1].Name) = "CMD";
$Imported_CSV # Shows the change #

You can extract all rows from one or more columns using a standard Select-Object, and with this you could apply different grouping, sorting and formatting approaches. Here is an example of getting the data:

$Imported_CSV | Select-Object -Property Age, Name

When you have designed your data segment (using the index and the properties for example), pipe your selected data into Export-CSV and set the delimiter you want to use. If you run Get-Process or Get-Service you could extract that data into a CSV as well.

The row in your Excel file (also the CSV) with the name of the columns (such as name, age, location) is called the header, if the top row is not the header row you can use the parameter -header for that. There are a few things to keep in mind when using this functionality which you can read more about here.

To add a column we can use the following code. Simply put it goes through each row and adds an extra "cell" on the rightmost side. You can give the $row variable a different name, but I liked its simplicity. The "Header" is simply the name of the new column. The property value is what value that should be given for all rows in the new column.

foreach ($row in $Imported_CSV) {Add-Member -InputObject $row -NotePropertyName "Header" -NotePropertyValue "DefaultValue"}

To add a new row to the CSV file you create a new object that is appended. For the new row you enter the values for each column.

$newrow = [pscustomobject]@{
Name = "PowerShell"
Age = "7.5.0"
Location = "PC"
}

$Imported_CSV += $newrow

Finally you can export your file as mentioned earlier through Export-CSV:

$Imported_CSV | Export-CSV -path "C:\temp\exportfile.csv" -NoTypeInformation -Delimiter "," -Encoding Default;

The script concept idea

My script lets the user search with a GUI that connects to the "database" file to work with it. The GUI lets you select different columns in which you can look for matching data. Column names could be "name", "age" and "location" for example and are connected to radio buttons. Essentially you tick a radio button corresponding to a column, you write a search word and then it gets all the data connected to that particular data point. An easy way to search for something you know, to get the rest of the data that you might not know.

The search function is tied to a search button. By using "$textbox.text" the search function matches that search term to the column that you chose. Finally it returns the data in another text box. In my example I decided that it should return all connected data.

Other functionality I added was a reset button that quickly resets the different buttons and text boxes. In the bottom I have a status bar that gives basic information and it could be used for more detailed error messages for example.

Conclusion

We can with simple means edit datapoints, create a new row or a new column. We can fetch and save the contents of a CSV file. Apart from this we can use our other knowledge of PowerShell to work with the data. Remember to create a backup of any important data before you start experimenting with it. 

Cryptocurrency: Running a Pi Node

Some years ago the cryptocurrency project Pi Network was created, in the beginning it was a closed project where people had to get an invite from existing users. I remember getting mine, which gave a small bonus to me and the person who recruited me.

A lot of time has passed since then and the mining rate has gone down from several coins per day to fragments of a coin per day. One way to currently increase your mining rate is to create what the creators call a Pi Node. The reward is constantly being tweaked and in the future they will determine what the reward will actually be, but currently it has rewarded me with increasing bonus rates of mining.

Just as mining Pi is free, running a node is virtually free as well. You need to have passed the KYC, which costs 1 Pi and you need a computer with internet that will be running. There will be something called a supernode as well, but they have a lot higher requirements of stability, essentially they are nodes for the nodes. This post will thus focus on how you get started with your regular node so that you can increase your mining rate. The content is based on the official information and my personal experience installing and running a Pi Network Node.

Getting started

Download the Pi app for the computer here. It contains the node software, unlike the app in the phone.

Fill in the application to run a node, the team will then see if you are a good candidate and select you.

Nodes are approved based on uptime, connection quality, contribution to the community and security circles.

You need to pass KYC (Know Your Customer).

When you are accepted you can download Docker, as the node requires it to run. You might have to enable virtualization on your intended computer, as it works similar to a virtual machine. I tried running it inside a VM but it didn't work as intended.

Making sure it is running properly

The first time you start the node it will take a while for it to catch up, as there is a lot of data in the blockchain that it needs to synchronize with. When I started mine it needed approximately 3 minutes to catch up. I was also running the optional API.

The Pi app becomes an autostart app automatically after installation, but make sure your Docker also starts automatically in your Windows settings. Inside your Docker program you can also see the containers running, you will see a play button next to the container. Press this if it has stopped.

If you lose power or internet, it can be a good idea to make sure your node is operational again. Normally I am running it on a headless server but I have a mini monitor attached just to make it easy to check regularly. In the future I will also consider getting a monitoring software, so that I can check the crucial services on a GUI from another computer.

If possible leave your node running constantly, on a separate device if need be. I noticed that my rewarded rate increased slightly day over day. It will be interesting to see if there is a cap to this, as the rate increase seemed to stagnate. At the time of publishing this post it has increased the reward for the four days that I have been running it, with updated values each new mining session. In other words, if there is an increase, you will notice it every time you start a new mining cycle.

Only run one node, only have one account, for your own safety. And finally, running a node will use CPU, memory, hard drive and network resources.

AI: Prompt Engineering

The other day I got an advertising on social media about a free certificate in responsible AI use and Prompt Engineering. The course was a collaboration between Microsoft and Founderz who hosted AI Skills 4 All which was educational and interesting.

As a regular user of both Copilot and Edge I decided to give it a go, it was free and took less than 8 hours, so I thought why not?

Basically they brought up the basics of how generative AI work, how to write good prompts and how they leverage the Edge browser. 

The way we should see Copilot is not as the primary agent of choice, but rather our second pilot, that can offload us from certain tasks. We are still responsible for what we do with the tool in that sense. At this stage these generative AI:s, especially Copilot, is better at texts and images than math.

Its strengths are brainstorming, handling large amount of information and creating general ideas for the low risk scenarios. The general downsides of Copilot is that it does not think or feel, it is not always accurate and you can't rely on it as a friend. It also comes with biases of its own.

Improve your prompts

Writing good prompts for AI consists of four main parts.

1. A goal, such as creating a bunch of bullet points detailing the last quarter

2. Context, such as defining a setting where the material will be used, for example a breakfast presentation. You could also specify what type of people that will attend. For example whether it is marketers, competitors or managers.

3. Source of the information, make sure to ask Copilot where to get the data from. This is important because not only is it your job to ensure the quality of the result, but you might know that some sources are better in certain fields. You will notice that the sources are linked in the result as well, if you want to continue your research.

4. Expectations, define what you want out of the material, what sort of result should it create?

Customizing your prompts further

There are additional things that you need to think about in order to create better prompts, such as being rich with details. Remember to add the most important things first, it is valued higher, this is especially true when creating images.

Here is an example of the prompt "Create an image of the planet Mars"

This is the result of the prompt "Now create an image of the planet Mars, from the surface of the planet, where you can an early human settlement, do not include any people. The sky should be daytime."

You can also ask the AI to ask you controlling questions back to you before presenting the result. This increased content quality. You could ask Copilot to return five questions about the material as an example.

To shape the content further, ask the chatbot to take on certain personas. You could for instance ask it to act as an PR-expert to create a certain effect with the text.

Keep in mind that if you are logged in, you can save a history of previous chats. You can also attempt to regenerate the reply, by simply asking it to do so, you might then get a variation of the reply.

What to avoid

There are a few things to avoid when you work with the Copilot, these pointers will also help you make better decisions when taking the help from AI. 

1. Avoid math that you cannot check yourself afterwards, this is due to the probabilistic nature of the generative AI. It uses mathematical models to predict what the next "token" in a sentence is, such as a word or another small piece of data. You can in worst case get different results from the same question. Instead try to focus on outputting text and code for example.

2. Don't be too sparse with the details, it will negatively impact your results.

3. Don't upload information, data, images and so on that might be sensitive or personal.

Troubleshooting the output

If an error occurs you can try to figure out what could have been causing the mishap.

For this it is also important to know how Copilot works. First it searches, then a Large Language Model is used to summarize the content and then it is filtered and presented. 

You can ask yourself these questions when troubleshooting the result:

• Is there an issue with how the question was asked?
• Are there problems with the sources used? Can other sources be used instead?
• Was there something wrong at the search stage?
• Was there an issue with the summarizing stage?
• Did something happen when it was presented?

Different ways to use Copilot

One thing that I myself learned from the course was an efficient way to use Copilot in the sidebar together with Microsoft Office in the browser. This is handy as Copilot is not necessarily integrated into the Office programs that you install on your PC, at the time of this blogpost.

Use the sidebar AI in Microsoft Edge to analyse, summarize, correct or improve Word documents that you have open in the browser, when you are working in Office online. You can of course use it on other sites too and let it assist you in evaluating and working with different content. Another hack I like to use is to take a snippet and upload it, and ask questions about the content.

There is also a speech function and if you use it, you will notice that it speaks back to you.

For mobile they have a dedicated app for Copilot as well, it pretty much fills the same role there, a plus for using the app is that you can use the camera to identify objects for example or transcribe text that you take a photo of.

Summary

Let's wrap this post up by saying that Copilot has helped me a lot, both professionally and in private. Personally I use it as an advanced search engine sometimes, with whom I can discuss the results and ask clarifying questions. I also use it to generate PowerShell code these days, but I try to understand the concept before I ask it to produce the code, this is so that I can evaluate the result better. It is a philosophy of mine that it is better to have done something the hard way (or at least have tried) first, because it gives a better intuitions of the subject.

The course raised my awareness of the probabilistic side of the AI, which explains the interesting results that one sometimes gets. Remember to use the up and down thumb to continue to train the AI, so that it becomes better.

Enjoy!

Hardware: Building a gaming computer

Background

Just like the jedi knights of Star Wars needs to build their own lightsaber, we techies should build our own computer to get a greater understanding for the tech we use and because it is an interesting project and a test of our ability.

This is the second proper gaming desktop that I have, the first was a prebuild that lasted me for 10 years (with some repairs included), but this was before my career in IT. Now I wanted to take on the challenge of building my own gaming computer. As a sidenote the primary games I wanted to play with this new computer were Enlisted, Minecraft, World of Warcraft and Skyrim. Later on I also want to try GTA 5 and League of Legends.

Building was the quick part, while budgeting and planning was the slower part of the adventure. This will be a post about my observations and what I learned. 

Preparations

The decision to build a computer started roughly a year ago, or a bit more. I postponed it for two reasons, I needed to save up some money and I wanted to wait for prices to go down. Furthermore I was also not sure about the parts or even the process itself. So I started by putting together many various setups using the online tool at Inet where I would play around, see what the best parts would give me or how expensive some setups could be. Some setups would be around 2500 euros, which is a lot. So a part of the problem was knowing how much to scale down the computer in performance to lower the costs. It took quite a few sample builds before I realized what would be "good enough".

So my lesson for you is that you can save a lot of money on your build by adapting it for the realistic use case. Maxing everything out is probably fun and definitely expensive. It would future proof your build as well. Which is something you want to keep in mind. You could for example focus on getting parts that are modern, but not necessarily the best performers, to allow for reasonable upgrades in the future.  

So when it was finally time to pick out the parts I would use for my real build what I focused on was getting enough performance for my games and getting relatively modern pieces, this allowed me to cut the costs down to 1200 euro. I can still upgrade RAM, processor and add more storage for example.

These are the parts I ordered for my build, two of the small fans were never used.

• Chassi: Chieftec Pro Mini Svart
• Processor: AMD Ryzen 7 8700F 4.1 GHz 24MB
• Graphics card: ASRock Radeon RX 6600 8GB Challenger
• Motherboard: ASUS Prime B650M-R
• RAM: Kingston 32GB (2x16GB) DDR5 5200MHz CL36 FURY Beast AMD EXPO/Intel XMP 3.0
• Storage: Kingston KC3000 M.2 NVMe SSD Gen 4 512GB
• Big fans: Phanteks M25 140mm PWM Svart
• Small fans: Arctic P12 PWM Svart PST Svart
• CPU-cooler: Cooler Master Hyper 212 Black Edition
• Power supply: Chieftec Proton 750W

I chose 32 GB ram to give me ample of headroom, while allowing me to upgrade to 96 GB in the future. The NVMe on 512 GB allows me to quickly boot up the computer, store all of my games at the same time while still leaving me a bit of room. I have a whole slot that I can install a much larger storage into later as well. The CPU felt quite modern and it didn't come with integrated graphics, which means I am 100% reliant on the graphics card. 

Another part of the preparations was to create a bootable USB-drive with Windows 11 Pro, so that I instantly could install it once the computer was ready.

Building the computer

Seeing as this was the first build I had no complete desktop building experience to draw upon. Instead I had a bit of laptop maintenance experience and knowledge from general tinkering and exploring. 

The first step was to gather all the boxes with components and look it over, bringing a general screwdriver and a knife for opening the boxes. The general principle is to complete your motherboard first, before putting it into the chassis. 

So after unboxing what had to be unboxed right away, I put the CPU, RAM and SSD in place. The RAM was the easiest, just make sure to put it facing the right direction and that it fits the slot. The SSD is also quite easy once you know how to keep it in place, in this case I used a locking mechanism made from some type of plastic. The CPU was a bit interesting, make sure to align the corners properly and drop it in straight from above. Then you use the clasp to hold it in place, at this point a plastic cover flew off which I wasn't prepared for. Scary!

Once the CPU was equipped, I first used a default CPU cooler, it already had thermal paste on. But later when I had research the non-stock cooler I equipped it using thermal paste that comes with the cooler in a tube. The CPU cooler was relatively easy to install, the trick is to lower it down on top of the CPU in an even fashion to not create too much bubbles in the thermal paste. Remember to remove the plastic sticker covering the area that connects to the CPU. Fasten the screws a bit at a time, once again making sure to spread the pressure evenly until it sits properly against the CPU.

Next step was to prepare the chassis, I installed the backplate for the motherboard. The secret to installing this backplate is to push it out evenly from the inside, just make sure to align the holes correctly. The role of the backplate is to protect the computer, it can help with both heat and to some degree interference, it also offers extra stability.

Continuing on the journey I now installed the front fans. This was very tricky because it was a challenge to get the front cover off. If I could have traveled back in time, I would perhaps have installed it earlier. I also installed the chassis to the motherboard, along with the power supply. At times it was a bit tricky to see which component that needed which cable, and in some case it was difficult to connect them to the motherboard. Here I also could have done a better cable management job, by running some of the cables behind the motherboard (there is a space there). My advice is that if you are working in a dark chassis, bring plenty of light, a lamp on the forehead would have helped a lot.

Lastly I put in the graphics card, I noted that it looks like it sits upside down in this build and on the backside you only need to remove the sliders that makes room for the card's interface. The graphics card was relatively easy to install, in part thanks to a combination of a good enough chassis and that the PCIe-slot didn't cause any issues. The graphics card needs electricity too, so I connected the power to it from the power supply.

Finally I installed some additional fans (and it was after this that I installed the non-stock CPU cooler).

Setup, first time use and finishing touches

I must admit that the first time I started the computer, I stretched my arm out to the power button, just in case it would explode. But fortunately, it worked on the first go. However, I did not connect it to my main setup at this stage, I had a small touch screen connected just to get through the Windows 11 setup and then of course the USB drive with the operative system.

After making sure it was correctly setup and working, I brought it to my desk and connected it to the ethernet cable and monitor, along with peripherals. 

First thing to do was running Windows Update, downloading specific drivers and update the firmware.

When everything was set and done it was time to download games and start enjoying the product, and what a relief it was to finally be able to play the favorite games again on full graphics. As the games were installed on the desktop I could also remove the from my Framework 13, who had been struggling even at the lowest graphics. 

Flipper Zero: Ducky scripts

Yesterday I received my Flipper Zero and after a long wait I was ready to tinker away with it.

After capturing some infrared signals and replicating them, it was time to test the legendary Bad USB functionality.

The content of this post is based on the GitHub page.

The BadUsb app on the Flipper Zero uses extended Ducky script syntax which is compatible with the classic USB Rubber Ducky 1.0 language but with extra components such as custom USB ID (how the keyboard is identified), ALT-codes (pressing ALT and numbers on the numpad) and the old SYSRQ command. Both \n and \r\n line endings are supported, empty lines are allowed and you can use spaces or tabs for line indentation.

The script is written in simple .txt files and can be copied over to the Flipper Zero using the Windows app for example, and it must be located in the correct Bad USB folder.

To use the Bad USB function, make sure to have a USB-C to USB-A cable that you connect between the Flipper and the computer that you want to run your script on.

Navigate to "Bad USB" in the main menu, select the script that you want to execute and then press run.
Almost instantly it will execute the keypresses that you preprogrammed.

Here is a simple example that helps you start a PowerShell window with administrative privileges.

GUI r
DELAY 500
STRING powershell
CTRL-SHIFT ENTER
DELAY 500
ALT Y
DELAY 500
STRING write-host "This script is running as admin"
ENTER

Spoof keyboard ID

You can set the custom ID of the Flipper USB HID device. ID command should be in the **first line** of script, it is executed before script run. VID and PID are hex codes and are mandatory. Manufacturer and Product are text strings and are optional.

ID, VID:PID Manufacturer:Product        

Example: "ID 1234:abcd Flipper Devices:Flipper Zero".

Delay, comments, holding down

DELAY, one delay value in ms

DEFAULT_DELAY, delay value in ms before every next command

DEFAULTDELAY, alternative to DEFAULT_DELAY

REM, commenting is done by writing REM followed by a single line of text

HOLD, press and hold key until RELEASE command. Up to 5 keys can be held at the same time.

RELEASE, release key

Arrow keys

DOWNARROW / DOWN

LEFTARROW / LEFT

RIGHTARROW / RIGHT

UPARROW / UP

Various buttons

APP, same as MENU, context menu key (right-clicking)

BACKSPACE

BREAK

CAPSLOCK

DELETE

END

ENTER

ESCAPE / ESC

Fx, F1-F12 keys

HOME

INSERT

MENU, same as APP, context menu key (right-clicking)

NUMLOCK

PAGEDOWN

PAGEUP

PAUSE

PRINTSCREEN

SCROLLLOCK

SPACE

TAB

You can combine the following with a special key command or a single character.

CONTROL / CTRL

SHIFT

ALT

WINDOWS / GUI, the Windows key

CTRL-ALT, ctrl plus alt

CTRL-SHIFT, ctrl plus shift

ALT-SHIFT, alt plus shift

ALT-GUI, alt plus the Windows key

GUI-SHIFT, the Windows key plus shift

GUI-CTRL, the Windows key plus ctrl

Strings and delay between keypresses

STRING, print text string, such as writing something in a terminal

STRINGLN, print text string and press enter after it

STRING_DELAY, delay value in ms which is applied once to next appearing STRING command

STRINGDELAY, an alternative to STRING_DELAY  

DEFAULT_STRING_DELAY, delay value in ms, applied to every appearing STRING command

DEFAULTSTRINGDELAY, alternative to DEFAULT_STRING_DELAY

REPEAT   Number of additional repeats  Repeat previous command

WAIT_FOR_BUTTON_PRESS, will wait for the user to press a button to continue script execution. Will wait indefinitely for a button to be pressed.

On Windows and some Linux systems, you can print characters by holding ALT key and entering its code on Numpad.

ALTCHAR, print single character

ALTSTRING, print text string using ALT+Numpad method

ALTCODE, same as ALTSTRING, presents in some Duckyscript implementations

Send SysRq command, which is a way to interact with the computer even in a frozen state.

SYSRQ, single character

Media keys

Some Media/Consumer Control keys can be pressed with "MEDIA" command

MEDIA, used with commands in the list below:

POWER

REBOOT

SLEEP

LOGOFF

EXIT

BRIGHT_UP, brightness up

BRIGHT_DOWN, brightness down

HOME

BACK

FORWARD

REFRESH

SNAPSHOT, take photo in a camera app

PLAY

PAUSE

PLAY_PAUSE

NEXT_TRACK

PREV_TRACK

STOP

EJECT

MUTE

VOLUME_UP

VOLUME_DOWN

FN, Fn/Globe key on Mac keyboard

GLOBE, special key or single character (Fn/Globe key commands for Mac/iPad)