arm on SANJAY REGMI

What to do When You Cannot Find 'Microsoft Print to PDF' Option When Trying to Print to PDF

Fri, 10 Apr 2026 12:21:22 -0700

Sometimes you’ll come across Windows 11 computers that do not have the very handy “Microsoft Print to PDF” option in the list of printers to select. This option is very handy to save files as PDF and to split big PPDF by individual pages. Here are the steps you could try to get that option back and working.¹²

Go to the search button on Windows, search for “Windows Features” and check under “Turn Windows on or off” box if “Microsoft Print to PDF” is turned off. If turned off, turn it on, and you’re good to go. If it’s turned on and still you do not see “Microsoft Print to PDF option, proceed below.
Open Command Prompt as an admministrator and run the following:

net stop wuauserv
net stop bits
net stop cryptsvc

ren C:\Windows\SoftwareDistribution SoftwareDistribution.old
ren C:\Windows\System32\catroot2 catroot2.old

If you get “Access denied when running the two last commands, run the top three commands again.

net start wuauserv
net start bits
net start cryptsvc

DISM /Online /Cleanup-Image /RestoreHealth
sfc /scannow

The above two commands WILL take time. Prepare to wait.

Reboot, open Command Prompt as admin and:

DISM /Online /Enable-Feature /FeatureName:"Printing-PrintToPDFServices-Features" /All

Check if “Microsoft Print to PDF” is now available when using a PDF Reader (say Foxit).

🍟

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested on Windows 11 Home Premium Edition (Build 26200) (64-bit) (Release ID 25H2), BIOS Version F.29, Released 05/07/2024, HP Laptop 17-cp0xxx ↩︎
Sourced from: https://techcommunity.microsoft.com/discussions/windows11/microsoft-print-to-pdf-missing/4416986 ↩︎

Two Quick Steps to Upgrade Windows Home to Windows Pro with a cheap license

Tue, 17 Mar 2026 13:23:29 -0700

Step 1: Purchase a low cost Windows license from here.

Step 2: On the machine to be upgraded, start Command Prompt with admin privileges. Run this: changepk.exe /productkey VK7JG-NPHTM-C97JM-9MPGT-3V66T. This will require you to reboot the machine and install the files necessary to get you into Professional license. If you are not going to buy a license and are okay without activation, you can use it. But if you bought a license go to the next step.

Step 3: On the machine that has been upgraded, start Command Prompt with admin privileges. Run this: changepk.exe /productkey YourPurchasedSerialNumber. This time you’ll not have to reboot the machine. Go to Settings –> Activation to see if your machine is now Windows Professional and is activated. It should be.

Works with Windows 11.

🍊

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Fix Upgrade Issues After Using Pkgbase on FreeBSD 15.x

Tue, 03 Feb 2026 22:21:41 -0800

I recently upgraded my nas from 14.x to 15-RELEASE and then converted to pkgbase because I just loved the idea of being able to do pkg update and pkg upgrade for all upgrades moving forward. After the upgrade, the version I was on was 15.RELEASE-p1. After about a month or so, when I noticed p2 was out, I ran pkg update and pkg upgrade, it still kept me at p1. Investigating, I figured when I ran the lua based script to upgrade to pkgbase, it didn’t (or I forgot) to say yes for FreeBSD base repository. If you’re like me, here’s the instructions to fix it quickly.

Open this file, I’ll use vim

vim  /usr/local/etc/pkg/repos/FreeBSD-base.conf

Change enabled: no to enabled: yes
Save it.
Make sure you did your edit correctly.

# cat /usr/local/etc/pkg/repos/FreeBSD-base.conf            
FreeBSD-base: { enabled: yes }

Now do pkg update and pkg upgrade and you should get into the latest upgrade.
Reboot (sadly, you’ll have to reboot to be able to use the new kernel)
After reboot, confirm you’ve upgraded by running:

freebsd-version -kru

All the entries listed by the command should be the latest version and patch levels.

🍷

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Fastest Way to Copy Ext4 Formatted Data to FreeBSD Using WSL2 as an Intermediary

Sun, 25 Jan 2026 17:56:52 -0800

I run immich on a very old dell laptop with 100mbps ethernet with USB connected disks for data. The disks are formatted ext4 and the dell runs fedora. (Immich doesn’t run on FreeBSD yet. :( ). When the disk was full, I wanted to use my FreeBSD server as the data drive over NFS. And retire the USB disks. This required copying about 500GB of data from the USB disk to my FreeBSD server before I could switch over.

The first natural idea was to use use rsync. I quickly figured though that the max speed of copy was 5MB/s probably because of 100Mbps ethernet port.

The second idea then was to bring the USB disk to the FreeBSD server and mount it there. Because FreeBSD doesn’t directly have NTFS drivers in the kernel for ext4, I used fusefs-ext4fuse. That bumped the speed to 13MB/s over USB 3.0. That was still too slow. I also learnt the FreeBSD has fusefs-ext2 but it only works for ext2, even though ext2fs has a native kernel module. I seem to have used fuse-ext2 -o async,noatime /dev/da0s1 /mnt/usb which again gave me 12MB/s, too slow a speed.¹

While researching, I figured there was another way that was very strange. It suggested using WSL2. Very curious, I decided to give it a shot. This is how I did it.²

Connect the disk to a Windows box.
Because Windows doesn’t understand ext4 natively, run this in PowerShell (Admin mode) to see if it’s connected:

get-ciminstance -query "select * from Win32_DiskDrive"

The output will be something like:

DeviceID           Caption                             Partitions Size          Model

---

\\.\PHYSICALDRIVE0 Samsung SSD 840 EVO 250GB           3          250056737280  Samsung SSD 840 EVO 250GB
\\.\PHYSICALDRIVE1 KINGSTON SNV3S1000G                 5          1000202273280 KINGSTON SNV3S1000G
\\.\PHYSICALDRIVE2 TOSHIBA External USB 3.0 USB Device 1          1000202273280 TOSHIBA External USB 3.0 USB Dev…

Note the DeviceID. In my case, it was connected as PHYSICALDRIVE2.

So to mount it in WSL2, you’d run, in the same Admin PowerShell Console:

 WSL2 --mount \\.\PHYSICALDRIVE2 --PARTITION 1

If you get an error like this, its just that you’re not running the command in PowerShell Admin mode:

Failed to attach disk '\\.\PHYSICALDRIVE2' to WSL22: Error: 0x80040315
Error code: WSL2/Service/AttachDisk/MountDisk/WSL2_E_ELEVATION_NEEDED_TO_MOUNT_DISK

Once the above command runs without error, open a WSL2 window and type this:

lsblk

Based on the output of lsblk, you could then mount the disk in WSL2 like this:

sudo mount /dev/sd1 /mnt/usb_disk

assuming /sd1 is your partition and the mount point is /mnt/usb_disk (you need to create the mountpoint before running the command).

Once I did that, I ran an rsync from within WSL2 to my NAS. I couldn’t believe my eyes, when the speed of copy went to 50-70 MB/s!

Apparently this is because WSL2 runs a proper/genuine Linux kernel so there’s no other layer to disrupt the flow.

Pleasantly surprised that Windows and WSL2 helped me today with my Linux to FreeBSD data transfer! Wonder what this says about WSL2’s development and progress so far…

😳

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

After this post went out, vermaden found me on the intertubes and showed that there was no problem coping from ext4 to freebsd (kudos to him!). Then I figured I’d used, fuse-ext2 -o async,noatime /dev/da0s1 /mnt/usb which is slow at 12MB/s instead of mount -t ext2fs /dev/da0s1 /mnt/usb which is fast at 60MB/s. For reading purposes, you could use ext2 drivers to correctly read ext4 formatted data. vermaden suggests to use lklfuse -o type=ext4 from the fusefs-lkl package where you get optimal results. ↩︎
WSL version: Linux 6.6.87.2-microsoft-standard-WSL2; PowerShell: 7.4.2; FreeBSD: 15.0-RELEASE-P1; LAN is gigabit ethernet. ↩︎

How to Get the Best Throughput - Both Upload and Download - From Remote Servers When Connected via Tailscale and Using Pfsense Firewall

Thu, 15 Jan 2026 16:29:04 -0800

If you’ve had a VPS with Hetzner in Europe (very competitive hosting rates) and you’re a Tailscale aficionado you’ve probably used Tailscale and been unhappy with upload/download speeds in kilobytes when connected via Tailscale over SSH. It turns out one reason for the low throughput is that you use pfSense as your firewall, and it can be difficult for Tailscale to punch a hole through pfSense. As a result, it falls back to its DERP servers, which are shared resources for all Tailscale users and often have poor throughput. With that said, here’s a quick change to apply on pfSense to alleviate the problem.

Diagnosis

Let’s find out whether your ssh connection to the remote servers is using DERP. Run this from your local machine:

tailscale ping <remote-server>

If the results show via DERP, then its confirmed, you’re not going through a direct connection. So proceed to the fix.

The Fix ¹

Open your pfSense portal and go to Firewall → NAT → Outbound.
Under Outbound NAT mode, select Hybrid Outbound NAT, then click Save.
On the same page, under Mappings, click the first Add button so the new rule is above the Automatic Rules.
Configure the rule and enable it:
- Protocol: UDP
- Static Port: enable it
- Click Save, then Apply Changes.
On the local machine, open a new terminal and run step 1 again. This time, you should see pong from ... in the output. That confirms you have a direct connection.
Check throughput by copying a few files with scp, you should see the throughput is much better than before.
Consider rebooting pfSense after step 7.
Tailscale suggests logging in to the Tailscale admin console and going to Access Controls. Find the nodeAttrs section and add randomize-client-port inside the curly brackets, like this:

"attr": ["funnel", "randomize-client-port"],

🤯

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested works on pfsense community edition version 2.8.1-RELEASE ↩︎

The Easiest and Fail Safe Way to Set Up Scan-to-Email on Your Scanner With Office 365

Wed, 14 Jan 2026 14:28:21 -0800

You just bought a shiny new printer, say an HP, and you want to set up Scan to Email so that when you scan from the multifunction device (printer/scanner), you receive an email with the scanned attachment.

Nowadays, printer manufacturers ship printers with lots of software and often require software to be installed on computers. Installing software on individual machines can be tedious for a small business if you have a bunch of computers.

The usual way to do this is to buy an extra email account and use it. But with two-factor authentication and Office 365 not enabling app passwords by default, it has become more complicated. (I found out the hard way that gmail app passwords don’t work for setting up some HP printers.)

Here’s a quick and easy method that doesn’t get mentioned enough and that many people still don’t know about. It’s called the Microsoft Direct Send method, and it works as long as you have Office 365 licenses. (You’ll not need to buy a new license for this to work.)¹

Stage 1 – Preparation

Find your public IP address. Go to whatismyipaddress.com and copy the IP address. For example, your IP might be 1.2.3.4.
Find your current MX record for the domain. For example: example-ca.mail.protection.outlook.com.
Update your existing SPF record to include your public IP address, like this:

v=spf1 a mx ip4:1.2.3.4 include:spf.protection.outlook.com ~all

Stage 2 – Printer setup

Go to your printer’s Scan to Email settings. Create a new profile and use the following:
Display Name: Choose a name
Email address: For example, scanner@example.ca (you do not need a valid mailbox or a paid license)
SMTP Port Number: 25
Secure connection: None
Server Authentication: None (or “Server does not require authentication”)
Save and test using a valid email address. You should receive an email into the valid email address.

Benefits of this approach

No need to install software on each machine to scan. The whole office/users can use it without any further changes.
Even if poeple change computers, they are upgraded or replaced, this works.
Will work for multiple Scanner/Multi fuction devices.
No password to maintain, never expires.
Does not require any license.

Swap Growing on Your Freebsd Server? Check Your php_fpm Settings - Step 2

Sat, 10 Jan 2026 19:39:00 -0800

After tuning for swap space by reducing MySQL memory usage, the other area to look into would be php_fpm settings.

The settings are located here if you’re using a freebsd jail:

/usr/local/bastille/jails/<your jail name>/root/usr/local/etc/php-fpm.d/www.conf

Edit the file so your last five values are like this:

[www]
user = www
group = www
listen = /var/run/php-fpm.sock
listen.owner = www
listen.group = www
listen.mode = 0660
pm = dynamic
pm.max_children = 3
pm.start_servers = 1
pm.min_spare_servers = 1
pm.max_spare_servers = 2

The meanings of the last five values are:

pm = dynamic # Worker count adjusts based on load
pm.max_children = 3 # Maximum 3 workers (caps memory usage), you might have a lot more by default
pm.start_servers = 1 # When the server starts, start with only one worker
pm.min_spare_servers = 1 # Keep at least 1 idle worker ready
pm.max_spare_servers = 22 # Do not create more than 2 idle workers

📢

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Swap Growing on Your Freebsd Server? Check Your Database Settings - Step 1

Fri, 09 Jan 2026 15:03:24 -0800

On a FreeBSD machine with a bunch of wordpress website jails that work with one database jail, the system might continuosly use up swap space if your VM has say 4G or less memory.

If the swap is continuously increasing, figure what is using swap:

top -o swap -n 20

And if it is a SQL process, open the following file, provided your db jail is called db0, you are using bastille as your jail application and you’re using Unix socket:

/usr/local/bastille/jails/db0/root/usr/local/etc/mysql/my.cnf

and make following edit to the file to make this variable 256 Megabytes (if the existing value is larger than that). Or even 128MB. Simple Wordpress sites work without issues with that much memory:

[mysqld]
innodb_buffer_pool_size = 256M

After that, restart your jail:

bastille restart db0

Now, if you run the top command, your swap should have immediately decreased. And then it’ll continously decrease over the next few hours.

If you want to decrease swap usage even further, go over to Swap Growing on Your Freebsd Server? Check Your php_fpm Settings - Step 2.

😤

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Setup Website Monitoring Properly So You Know if Your Database Connection Goes Down

Sun, 04 Jan 2026 00:14:03 -0800

I monitor my websites with UptimeRobot on the free plan. It’s good, the timer checks every five minutes and for most hobby projects, it’s more than enough. The monitor URLs were setup as for example, https://prasna.ca. A few days ago, I realized there was ‘database connection’ error but since I use redis and CloudFlare, simple curl commands give status 200 and we never found out AND UptimeRobot was happy as well so we didn’t know the site was down.

I figured, if you setup the URLs in UptimeRobot like https://prasna.ca/wp-login.php, it catches the db error. And the reason is that URL is not cached and that UptimeRobot seems to be smart enough to find that its a db error and sends ‘website down’ alerts.

♨️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Exploit the benefit of security and stability of FreeBSD by running webservers in jails along with reverse proxy jail, database jail and a redis jail

Thu, 01 Jan 2026 16:07:29 -0800

This holiday season I read up on FreeBSD jails. I’ve always wanted to be able to setup secure and stable websites. And Jails were the answer.

I’ve used Bastille as a jail tool, there are many out there and you can even do it without a tool, the FreeBSD handbook has enough to set them up.

My architecture for version one was a reverse proxy for all website jails. Then each website jail would have the whole stack - the database, the php components and the wordpres website. It took some time to build one but then I quickly realized, running MySQL databases for every website was a bit too overkill as MySQL took a lot of RAM out of my Hetzner VM with 4GB of memory. Having said that, if you need absolute security, you should do it this way. But I found a more efficient method for my small websites.

Version two consisted of a proxy jail for all websites to reverse proxy, one database jail for all website jails to use, one redis jail for all websites jails to use. This was a much better approach. After a lot of trial and error, I built two scripts. One to setup the base system with whatever websites you currently have. Then a second script to just add more websites. Both are ansible playbooks and the way I did was to copy paste the scripts on the console of the VM via SSH. This creates all required files and you can just run the .sh file.

The script has been designed to run on a freshly spun FreeBSD instance. At Hetzner, you first buy a VM, mount a bootable Freebsd iso. Then reboot the machine. The reboot kicks off booting from the mounted iso. So you install freebsd with ZFS, add a user and put them into wheel group, shutdown the machine. Once the machine is shutdown, unmount the ISO and reboot. You now can ssh into the box using your newly added user. Then do a ‘su’, become root and copy paste the script below.

The script takes care of pkg update, sets up Bastille properly and then builds the entire foundation. It takes cares of setting up of all components, sets up gzip and http2 and redis, certbot etc. At the end, you can just setup the wordpress sites from your browser.

Scripts

Foundation script Instruction: SSH into the host, copy paste the script into console, run the sh file.
Add new Wordpress website script Instruction: SSH into the host, copy paste the script into console, run the sh file.

Some specs

Host system is using FreeBSD 15.0-RELEASE.
All jails are also using FreeBSD 15.0-RELEASE.
Tested on Hetzner VM with 4GB of memory in the second half of December 2025.

Notes

Install Tailscale, enable tailscale setup –ssh so you can securely connect to the server passwordless. You can then stop the regular SSH from running (Yeah, tailscale doesn’t need that). That way, the thousands of bots trying to SSH to your server will not have anything to do at least on your server.https://sanjayregmi.com/posts/how-to-install-tailscale-on-freebsd/
Use ZFS snapshots for safety, backup.
Hetzner snapshots are very fast, take about one minute for 20GB of data. Use it - it costs extra but it’s worth it.
Lock the VM from inadvertent deletion. Hetzner console has the feature and its easy to find and setup.
Setup DragonFly Mail Agent (DMA) on the FreeBSD system so it alerts you via email. It’s quick and easy to setup.
Always take ZFS snapshots before a major update. That’s one of the beauties of ZFS.

🚥

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Enable Sending Email From Your FreeBSD Machine Using Dragonfly Mail Agent (DMA)

Mon, 22 Dec 2025 19:10:57 -0800

After you’ve setup your FreeBSD box, having the box email you for all sorts of reports, cron updates etc is really useful. You don’t have to login to the machine to see them, you can get them in your email. With DragonFly Mail Agent, the setting of such an email system is quick and easy. Here’s the intructions:

Prerequisites

Replace nobody@gmail.com with your own email address. Make sure you have two factor authentication on and then you can setup an app password.
Replace “replace-with-your-app-password” with your gmail app password.
The following using gmail as your SMTP provider.
Assumes you’re installing this as root.

Part one - setup DMA

Setup of DMA for FreeBSD 14.x

Install DMA

pkg install dma

Backup original config

cp /usr/local/etc/dma/dma.conf /usr/local/etc/dma/dma.conf.bak

Configure DMA

cat > /usr/local/etc/dma/dma.conf << 'EOF'
SMARTHOST smtp.gmail.com
PORT 587
AUTHPATH /usr/local/etc/dma/auth.conf
SECURETRANSFER
STARTTLS
MASQUERADE nobody@gmail.com
EOF

Set up authentication

cat > /usr/local/etc/dma/auth.conf << 'EOF'
nobody@gmail.com|smtp.gmail.com:replace-with-your-app-password
EOF

Secure auth file

chmod 640 /usr/local/etc/dma/auth.conf
chown root:mail /usr/local/etc/dma/auth.conf

Enable DMA in rc.conf

sysrc sendmail_enable="NO"
sysrc sendmail_submit_enable="NO"
sysrc sendmail_outbound_enable="NO"
sysrc sendmail_msp_queue_enable="NO"
sysrc dma_enable="YES"

Set up mailwrapper

cat > /etc/mail/mailer.conf << 'EOF'
sendmail        /usr/local/libexec/dma
mailq           /usr/local/libexec/dma
newaliases      /usr/local/libexec/dma
EOF

Quick Test

echo "Test email" | mail -s "Test from FreeBSD" nobody@gmail.com

Monitor logs

tail -f /var/log/maillog

Setup of DMA for FreeBSD 15.x

Notice the path for dma is different, its simpler in 15.x.

pkg install dma

cp /etc/dma/dma.conf /etc/dma/dma.conf.bak

cat > /etc/dma/dma.conf << 'EOF'
SMARTHOST smtp.gmail.com
PORT 587
AUTHPATH /etc/dma/auth.conf
SECURETRANSFER
STARTTLS
MASQUERADE nobody@gmail.com
EOF

cat > /etc/dma/auth.conf << 'EOF'
nobody@gmail.com|smtp.gmail.com:replace-with-your-app-password
EOF

chmod 640 /etc/dma/auth.conf
chown root:mail /etc/dma/auth.conf

sysrc sendmail_enable="NO"
sysrc sendmail_submit_enable="NO"
sysrc sendmail_outbound_enable="NO"
sysrc sendmail_msp_queue_enable="NO"
sysrc dma_enable="YES"

cat > /etc/mail/mailer.conf << 'EOF'
sendmail        /usr/libexec/dma
mailq           /usr/libexec/dma
newaliases      /usr/libexec/dma
EOF

echo "Test email" | mail -s "Test from FreeBSD" nobody@gmail.com

tail -f /var/log/maillog

Part two - make cron send emails to you

# Create /etc/mail/aliases or add root alias
echo "root: nobody@gmail.com" >> /etc/mail/aliases

# Or if you want all mail to go to your Gmail
cat > /etc/mail/aliases << 'EOF'
# Send all system mail to Gmail
root: nobody@gmail.com
EOF
# Then rebuild aliases
newaliases

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Upgrading Proxmox 8 to 9 on Macbook Pro 2015

Mon, 15 Dec 2025 19:03:56 -0800

I run proxmox 8 on an old 2015 ish Macbook Pro. 16GB memory, silent, 500GB SSD, its a good machine. When Proxmox 9 came out, I wanted to upgrade it to 9. Proxmox has an upgrade guide. Followed it, everything went fine. After reboot and upon ssh, I realized, root was mounted readyonly.¹

A dmesg check revealed that there were a lot of hard disk read errors. Strange I thought. LLMs said the nvme drives on those machines were not so great and that it was probably failing and it failed now. I wasn’t so sure. After a few cleanups and reboots, I decided to use CloneZilla to image the original Mac OS I’d saved. Interestingly, restore succeeded without any issues. Now, Clonezilla is all linux and if it was able to first run without issues and succeed in cloning, the mac ssd wasn’t the issue.

Thinking perhaps the upgrade was the issue, I used the Proxmox 9 ISO to do a full and new install. Midway through the installation, I began to get write errors. That was confounding!

So after some thinking, I tried to install Ubuntu 24.x, same issue. Lastly, I ran the Proxmox 8 install, it installed flawlessly. Then I knew it was a kernel/drivers etc. issue. Took half a day but I’m back using the Proxmox server on my trusty macbook pro.

Learnings - Dont’ trust LLMs. They don’t have the full picture, they go after break fix and sometimes, they’ll waste your time going into circles.

❌

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested on Proxmox ISO version 9.1-1. Macbook pro 2015 ish. Linux pve-a 6.8.12-17-pve #1 SMP PREEMPT_DYNAMIC PMX 6.8.12-17 (2025-11-21T11:16Z) x86_64 GNU/Linux ↩︎

Hard Disk Noise and Seagate IronWolf Pro 12TB

Mon, 01 Dec 2025 15:41:08 -0800

I bought two Seagate IronWolf Pro 12TB hard drives recently for my home nas. They’re considered great drives for 24/7 operation. I’m using them as mirrored ZFS drives in FreeBSD. One thing that took me by surprise was the noise of the disk! The first time I installed them and I was running them on a desk (yeah, too excited to place them in the case), outside the case, the noise was startling. I thought the disks were bad. I had to repeatedly run smartmontools to check that they were good and that the ZFS pools had no errors. The seeks are pretty loud and unsettling and the writes are noisy as well. Since then, I’ve properly installed them into the case and kept teh case about a metre away and I’ve perhaps gotten used to it. I’ll probably move it out but if you have a lot of continuous disk operations during the day at your office, you’d really have to get used to the noise.

A month in, ten TBs or reads/writes and they’re happily chugging along.

camcontrol devlist output: ST12000NT001-3MD101 EN01

♨️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Enjoy Fedora Command Line Input (CLI) colors? Apply it quickly to your Ubuntu and Debian machines

Mon, 10 Nov 2025 13:13:59 -0800

I recently installed Fedora and loved the CLI color combinations (vibrant!). I wanted to apply it to Ubuntu and Debian (Proxmox) servers so here are the commands, just copy paste it into your cli.¹:

The script

#!/bin/bash
# Apply Fedora-like CLI colors and prompt on Ubuntu

# Backup existing configuration
cp ~/.bashrc ~/.bashrc.bak.$(date +%s)

# Append Fedora-style prompt and color aliases
cat <<'EOF' >> ~/.bashrc

# ==== Fedora-style CLI color scheme ====

# Enable color support for ls and grep
export LS_OPTIONS='--color=auto'
eval "$(dircolors -b ~/.dircolors 2>/dev/null || dircolors -b)"
alias ls='ls $LS_OPTIONS'
alias ll='ls -l $LS_OPTIONS'
alias l='ls -lA $LS_OPTIONS'
alias grep='grep --color=auto'
alias fgrep='fgrep --color=auto'
alias egrep='egrep --color=auto'

# Fedora-style prompt
if [ -n "$PS1" ]; then
  if [ "$(id -u)" -eq 0 ]; then
    PS1='[\033[01;31m]\h[\033[00m]:[\033[01;34m]\w[\033[00m]# '
  else
    PS1='[\033[01;32m]\u@\h[\033[00m]:[\033[01;34m]\w[\033[00m]\$ '
  fi
fi

# Optional: Fedora-like LS_COLORS palette
if [ ! -f ~/.dircolors ]; then
  wget -q -O ~/.dircolors https://raw.githubusercontent.com/trapd00r/LS_COLORS/master/LS_COLORS
fi

# ==== End Fedora-style CLI ====
EOF

# Reload configuration
source ~/.bashrc

echo "Fedora color scheme applied. Run 'source ~/.bashrc' if not already active."

😚

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested on Ubuntu 22.04.3 LTS and Debian 13.1. ↩︎

Turn your Proxmox host into a fully functional Apple Time Machine backup server using Samba + Avahi — compatible with all modern macOS versions

Fri, 07 Nov 2025 12:18:06 -0800

The following is the steps to connect a USB external disk to proxmox host and enable Time machine. The instructions work for macOS 10.13 → 15.7.¹

1. Connnect your USB disk and mount it

lsblk                     # find your USB disk, e.g. /dev/sdb
fdisk /dev/sdb            # create GPT + one partition (n → enter → enter → w)
mkfs.ext4 /dev/sdb1
mkdir -p /mnt/usb-timemachine
blkid /dev/sdb1           # copy UUID

Edit /etc/fstab:

nano /etc/fstab

Add:

UUID=<your-uuid> /mnt/usb-timemachine ext4 defaults 0 2

Then:

mount -a

2. Install Samba + Avahi

apt update
apt install samba avahi-daemon avahi-utils -y
systemctl enable --now smbd nmbd avahi-daemon

3. Create Time Machine User

useradd -m time-machine-user
passwd time-machine-user
smbpasswd -a time-machine-user
smbpasswd -e time-machine-user

Prepare the directory:

mkdir -p /mnt/usb-timemachine/timemachine
chown -R time-machine-user:time-machine-user /mnt/usb-timemachine/timemachine
chmod -R 770 /mnt/usb-timemachine/timemachine

4. Configure Samba

nano /etc/samba/smb.conf

Replace all contents with:

[global]
   workgroup = WORKGROUP
   server string = Proxmox Time Capsule
   map to guest = Bad User
   disable spoolss = yes
   host msdfs = no
   server signing = mandatory
   smb encrypt = required
   min protocol = SMB2
   max protocol = SMB3
   server min protocol = SMB2_10
   ea support = yes
   vfs objects = catia fruit streams_xattr
   fruit:aapl = yes
   fruit:metadata = stream
   fruit:posix_rename = yes
   fruit:resource = file
   fruit:locking = none
   fruit:encoding = native
   fruit:zero_file_id = yes
   strict sync = no
   sync always = no
   spotlight = no

[TimeMachine]
   path = /mnt/usb-timemachine/timemachine
   read only = no
   valid users = time-machine-user
   vfs objects = catia fruit streams_xattr
   fruit:time machine = yes
   fruit:time machine max size = 900G
   msdfs root = no

Restart:

systemctl restart smbd nmbd

5. Advertise via Bonjour (Avahi)

nano /etc/avahi/services/smb.service

Paste:

<?xml version="1.0" standalone='no'?>
<!DOCTYPE service-group SYSTEM "avahi-service.dtd">
<service-group>
  <name replace-wildcards="yes">%h Time Capsule</name>
  <service>
    <type>_smb._tcp</type>
    <port>445</port>
  </service>
  <service>
    <type>_device-info._tcp</type>
    <port>0</port>
    <txt-record>model=TimeCapsule8,119</txt-record>
  </service>
</service-group>

Restart Avahi:

systemctl restart avahi-daemon

Verify broadcast:

avahi-browse -a | grep SMB

Expected output:

= eth0 IPv4 pve Time Capsule _smb._tcp local

6. Test Locally

smbclient -L localhost -U time-machine-user

Expected:

Sharename       Type
---------       ----
TimeMachine     Disk

7. Connect from macOS

Finder → Go → Connect to Server
```
smb://pve.local/TimeMachine
```
(or use your Proxmox IP, e.g. smb://192.168.1.8/TimeMachine)
Login as time-machine-user
Open System Settings → General → Time Machine → Add Backup Disk
Select “pve Time Capsule” and use the same credentials
Time Machine will show “Preparing backup…”

8. Verify on Proxmox

ls /mnt/usb-timemachine/timemachine

You should see:

<YourMacName>_<UUID>.sparsebundle

Done!

You now have a fully compatible, Proxmox-hosted Time Machine server that works across macOS 10.13 → 15.7, supports multiple Macs, uses SMB3 encryption, and broadcasts as a genuine Time Capsule.

📛

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested on Proxmox version Linux pve 6.14.8-2-pve #1 SMP PREEMPT_DYNAMIC PMX 6.14.8-2 (2025-07-22T10:04Z) x86_64 GNU/Linux. ↩︎

If You Are Constantly on SSH and Having to Run Long-Running Processes Then Use Tmux

Mon, 03 Nov 2025 21:14:57 -0800

If you’re like me, always running ssh to connect to servers and having to run processes and wanting to know what you ran earlier in one combined session, run tmux. No backgrounding of commands, no using nohup, and even if the ssh session breaks, which most of the time it does if you’re running something that takes a long time, tmux is the best.

Running tmux is easy, just install and run tmux. However, we’re going to set up tmux such that it runs immediately after ssh. Here are the steps for FreeBSD 14.3.

Edit ~/.cshrc, add the following lines at the bottom of the file:

# Auto-start tmux on login (SSH or console)
if (! $?TMUX) then
    if ( -x /usr/local/bin/tmux ) then
        tmux attach -t main >& /dev/null || tmux new -s main
    endif
endif

Run a new ssh session and tmux should start automatically.

Note

To get back to the last session, do:

tmux ls

If you didn’t name your tmux session, then the default command would be:

tmux attach -t 0

🍇

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Combine Lots of Video Clips Into One Very Quickly

Wed, 29 Oct 2025 21:03:16 -0800

Say you have 30 (or any large number) different clips in a folder that you just want to combine (concatenate) based on the timestamps. You do not want to reencode them, just want to connect/combine them into one as if they were text files. The reason you do not want to re-encode them is that it’ll take a long time to process them and you might lose fidelity of the video if re-encode was not done correctly. The following instructions will just concatenate the files and produce a single file as an output in no time even on old hardware. (Ran this in FreeBSD 14.3 but you should be able to adapt it to a linux box or a Mac with homebrew).

Get a list of files in the order you want:

ls -1tr > files.txt. This saves the files we need in the order we need. You can edit this file using your editor if needed.

Create the concatenation list

Run this in /bin/sh: for f in $(ls -1tr); do echo "file '$f'" >> concat.txt; done

Run ffmpeg to assemble into a single file

ffmpeg -f concat -i concat.txt -c copy final-single.mp4

Final check

Now run it and upload it to your Immich server!

💝

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Set Up Your Own Photos Server Instead of Google Photos or Something Else by Using Immich and Tailscale

Mon, 20 Oct 2025 20:40:22 -0800

Last updated: 03 Jan 2026

The solution is a free software called Immich https://immich.app and you can run it at your home (or in a rented server somewhere) on an old machine with some USB drives or set it up on fancy stuff with RAID etc.

There are four parts to it:

Setup Immich Server

Go watch the awesome video that Tailscale folks have produced: https://www.youtube.com/watch?v=guHoZ68N3XM

Setup Immich Server URL with Tailscale

This allows you to access your immich server with a URL that is only accessible to you as it using tailscale VPN.
Tailscale to the rescue again. Watch the same video as above.

Upload your media (for a large amount)

Download and install Immich-go, which allows you to upload large amounts of photos and videos to your Immich server with either your own local media or stuff you downloaded from Google Takeout.
Download and install for your architecture: https://github.com/simulot/immich-go/releases/latest
Basic commands - to upload from Google Takeout (download as zip): ./immich-go upload from-google-photos --server=https://your-immich-server-url --api-key=get-it-from-immich-portal /location-of-your-google-takeout-files
Basic commands - to upload from a folder in your system: ./immich-go upload from-folder --server=https://your-immich-server-url --api-key=get-it-from-immich-portal /path-to-photos-to-upload
For very large files and slow networks, you might want to keep the session longer so the files actually get uploaded. The switch to add to the above is: --client-timeout=8h. Change the 8h (eight hours) to whatever suits you.
If you want to upload as another user (say a family member wants to upload ttheir google takeout), you can login as them and under settings, generate their own API key that you can use in the above commands so the bulk uploads go to their accounts.

Update your Immich server (installed via docker container)

SSH to your server
Find the the path to immich installation, say /appdata/immich/.
run these commands in sequence to stop the docker container.

docker compose stop

docker compose pull

docker compose up -d

Check using app or login to immich and see if the version information is correct.

Set up backup jobs to at least another disk (ideally three copies)

Of course, back up your main data folder. Use rsync or if in a VM, you’ve got better tools.

Enjoy!

Download the app to your phone, set up autobackups, tinker around

🐬

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Finally You Can Install Latest Versions of Mac OS on Old Hardware

Mon, 13 Oct 2025 19:00:00 -0700

I keep a bunch of old hardware. The oldest one is a Toshiba Dynabook SS 2000 DS80P (circa 2001). But I digress. I also have a 2015 Mac stuck at Monterey. I’m always trying to see if my old machines can be used in any way so they’re relevant still today. While researching recently, I came across soemthing called Opencore Legacy Patcher (OLP) that allows you to install latest Mac OSes on old computers! I was jubiliant.

I tried on a Macbook 2011 (Sequoia) and it stayed on for over two hours (Poor thing has ‘only’ 2GB memmory). So I skippped. But on the Macbook 2015, it worked! I ended up installing Sequoia. The instructions are very simple, just head over to https://github.com/dortania/OpenCore-Legacy-Patcher and read the getting stated guide - https://dortania.github.io/OpenCore-Legacy-Patcher/.

The biggest thing about upgrading is getting a more secure Operating System or one that is patched continuously.

After the upgrade, Macbook is still god, it’s not sluggish and works like a charm. If you have an old machine, give it a shot.

I did the try to install OS 26 which did install but in the login screen, the trackpad wouldn’t work and the keyboard wouldn’t either. Perhaps it’s a bug. So had to revert back to Sequoia.

If you want to try it out, do clone your mac disk before, instructions here: https://sanjayregmi.com/posts/image-and-restore-macs-using-clonezilla/

:contentment:

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to do disk cloning when Clonezilla doesn't list the external source disk

Tue, 16 Sep 2025 12:32:42 -0700

I wanted to clone an external SSD at 256GB (source) with internal NVME 512GB (destination) using clonezilla. (It had freebsd in it). External SSD was connected via usb-c. The machine was a Dell XPS laptop.¹ Each time I ran clonezilla, it wouldn’t show me the external disk. However, lsblk showed it and to make matters worse, if I chose the image to disk option, the external disk would show! I even tried with the latest Clonezilla live cd, “alternative stable - 20250620-plucky” and it had the same issue.

So at the end, I just did ran a DD command something like:

sudo dd if=/dev/source of=/dev/dest bs=64M status=progress oflag=direct

which in fact worked, even though Clonezilla was refusing to show the source disk.

Of course you need to fix the partitions (extend them) later becuase the source disk is smaller than the destination but the cloning itself worked without any hiccups.

🐎

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Tested using Clonezilla version “alternative stable - 20250620-plucky” with and external SSD disk containing FreeBSD 14.3 with zfs to an NVME drive in a Dell XPS 15. ↩︎

Find DMARC DKIM SPF Nameservers With One Script

Thu, 28 Aug 2025 16:33:52 -0700

If you are setting up Microsoft Office or Google Workspace, you’re constantly having to check DMARC, DKIM and SPF values and then check if the edited values have propagated. Here’s a script that does just that. It even saves main zone records in a new folder with a symlink of the latest output. All output is saved.

This script was developed with a large language model.

Features

Displays DKIM, SPF, DMARC and nameserver values.
Polls against Cloudflare and Google servers.
Saves a little bit more output to a folder with the domain name.
Symlnks the latest file with ’latest’ in the name.

This script is tested on WSL running Ubuntu 24. Please run it at your own discretion.

How to run

Make your script executable. (chmod +x scriptname)
Enter like this on the terminal: ./dns-check.sh example.com

#!/usr/bin/env bash
# dns-check.sh (summary + zone snapshot only)
# Usage:
#   ./dns-check.sh [-r 1.1.1.1] domain.com
# Env:
#   RESOLVER=1.1.1.1
#   RESOLVERS=1.1.1.1,8.8.8.8,9.9.9.9
set -euo pipefail

# ----- args -----
resolver_cli=""
if [[ "${1:-}" == "-r" || "${1:-}" == "--resolver" ]]; then
  resolver_cli="${2:-}"; shift 2 || true
fi
domain="${1:-}"
if [[ -z "${domain}" ]]; then
  echo "Usage: $0 [-r 1.1.1.1] domain.com"; exit 1
fi

# ----- resolvers -----
if [[ -n "${resolver_cli}" ]]; then
  resolvers=("${resolver_cli}")
elif [[ -n "${RESOLVER:-}" ]]; then
  resolvers=("${RESOLVER}")
elif [[ -n "${RESOLVERS:-}" ]]; then
  IFS=',' read -r -a resolvers <<< "$RESOLVERS"
else
  resolvers=(1.1.1.1 8.8.8.8 9.9.9.9)
fi

# ----- deps -----
if ! command -v dig >/dev/null 2>&1; then
  echo "Installing dnsutils..."; sudo apt update -y && sudo apt install -y dnsutils
fi

# ----- dig helpers (caching + single active resolver) -----
dig_quick_opts="+short +time=1 +tries=1"
dig_dump_opts="+noall +answer +time=1 +tries=1"

ACTIVE_RESOLVER=""

choose_resolver() {
  for r in "${resolvers[@]}"; do
    if dig ${dig_quick_opts} @"$r" NS . >/dev/null 2>&1; then
      ACTIVE_RESOLVER="$r"; return
    fi
  done
  ACTIVE_RESOLVER="${resolvers[0]}"
}
choose_resolver

declare -A CACHE_Q   # key "TYPE|NAME" -> +short text
declare -A CACHE_F   # key "TYPE|NAME" -> +answer text

_q_raw() {  # raw +short
  local rr="${1-}" name="${2-}"
  [[ -z "$rr" || -z "$name" ]] && { printf ""; return 0; }
  dig ${dig_quick_opts} @"$ACTIVE_RESOLVER" "$rr" "$name" 2>/dev/null || true
}
_f_raw() {  # raw +answer
  local rr="${1-}" name="${2-}"
  [[ -z "$rr" || -z "$name" ]] && { printf ""; return 0; }
  dig ${dig_dump_opts} @"$ACTIVE_RESOLVER" "$rr" "$name" 2>/dev/null || true
}

q() {  # cached +short
  local rr="${1-}" name="${2-}" k
  [[ -z "$rr" || -z "$name" ]] && { printf ""; return 0; }
  k="${rr}|${name}"
  if [[ -z "${CACHE_Q[$k]+x}" ]]; then
    CACHE_Q[$k]="$(_q_raw "$rr" "$name")"
  fi
  printf "%s" "${CACHE_Q[$k]}"
}
f() {  # cached +answer
  local rr="${1-}" name="${2-}" k
  [[ -z "$rr" || -z "$name" ]] && { printf ""; return 0; }
  k="${rr}|${name}"
  if [[ -z "${CACHE_F[$k]+x}" ]]; then
    CACHE_F[$k]="$(_f_raw "$rr" "$name")"
  fi
  printf "%s" "${CACHE_F[$k]}"
}

# ----- TXT helpers -----
get_txt_raw_lines() { q TXT "$1" || true; }
get_txt_records() {
  local raw line
  raw="$(get_txt_raw_lines "$1")"
  [[ -z "$raw" ]] && return 0
  while IFS= read -r line; do
    printf "%s\n" "$line" | sed 's/^"//; s/"$//' | tr -d '"'
  done <<< "$raw"
}
wrap_txt() { fold -s -w 70 | sed 's/^/    /'; }
ttl_of() {  # parse TTL from cached full answer
  local rr="${1-}" name="${2-}" ttl
  [[ -z "$rr" || -z "$name" ]] && { printf "Auto"; return 0; }
  ttl="$(f "$rr" "$name" | awk 'NR==1{print $2}')"
  [[ -n "$ttl" ]] && printf "%s" "$ttl" || printf "Auto"
}

# ----- paths -----
mkdir -p "$domain" "names"
ts="$(date +"%Y-%m-%d_%H-%M-%S")"
zonesfile="$domain/${domain}_${ts}.zones"
zones_latest="$domain/latest.zones"

# ====== HUMAN READABLE SUMMARY ======
echo "Checking DNS records for: $domain"
echo "Resolver: $ACTIVE_RESOLVER (chosen from: ${resolvers[*]})"
echo "--------------------------------------------------"

# SPF
echo "[SPF]"
mapfile -t spf_list < <(get_txt_records "$domain" | awk 'BEGIN{IGNORECASE=1} /^v=spf1/ {print}')
if (( ${#spf_list[@]} == 0 )); then echo "No SPF record found"
elif (( ${#spf_list[@]} == 1 )); then echo "${spf_list[0]}"
else
  echo "⚠️ Multiple SPF records found (${#spf_list[@]}). Merge into one:"
  for i in "${!spf_list[@]}"; do printf "  [%d] %s\n" "$((i+1))" "${spf_list[$i]}"; done
fi
echo

# DMARC
echo "[DMARC]"
mapfile -t dmarc_list < <(get_txt_records "_dmarc.$domain" | awk 'BEGIN{IGNORECASE=1} /^v=DMARC1/ {print}')
if (( ${#dmarc_list[@]} == 0 )); then echo "No DMARC record found"
elif (( ${#dmarc_list[@]} == 1 )); then echo "${dmarc_list[0]}"
else
  echo "⚠️ Multiple DMARC records found (${#dmarc_list[@]}). Merge into a single record."
  for i in "${!dmarc_list[@]}"; do printf "  [%d] %s\n" "$((i+1))" "${dmarc_list[$i]}"; done
fi
echo

# MX
echo "[MX]"
mx="$(q MX "$domain" || true)"
[[ -n "$mx" ]] && echo "$mx" | sort -n -k1 || echo "No MX records found"
echo

# NS
echo "[Nameservers]"
ns="$(q NS "$domain" || true)"
[[ -n "$ns" ]] && echo "$ns" | sort || echo "No NS records found"
echo

# DKIM
echo "[DKIM]"
policy_recs="$(get_txt_records "_domainkey.$domain" || true)"
if [[ -n "$policy_recs" ]]; then
  echo "Policy (_domainkey.$domain):"
  while IFS= read -r pline; do echo "  $pline"; done <<< "$policy_recs"
  echo
fi

selectors=(s1 s2 selector1 selector2 default google mail k1 k2 mx dkim)
found_txt=0; found_cname=0; any_found=false
dkim_found_names=()

for sel in "${selectors[@]}"; do
  rec="$sel._domainkey.$domain"
  txt_lines="$(get_txt_records "$rec" || true)"
  cname="$(q CNAME "$rec" || true)"

  if [[ -n "$cname" ]]; then
    any_found=true; ((found_cname++)) || true
    echo "Selector: $sel"; echo "  Record: $rec"; echo "  CNAME → $cname"; echo
    dkim_found_names+=("$rec")
  fi
  if [[ -n "$txt_lines" ]]; then
    while IFS= read -r tline; do
      if echo "$tline" | grep -qi '^v=DKIM1'; then
        any_found=true; ((found_txt++)) || true
        echo "Selector: $sel"; echo "  Record: $rec"; echo "  TXT →"
        echo "$tline" | wrap_txt; echo
        dkim_found_names+=("$rec")
      fi
    done <<< "$txt_lines"
  fi
done

[[ "$any_found" != true ]] && echo "No DKIM records found with common selectors."
echo "Summary: DKIM TXT=$found_txt, CNAME=$found_cname"
echo "--------------------------------------------------"

# ===== ZONE SNAPSHOT (full answers; uses cached f()) =====
{
  echo "; Zone snapshot for $domain"
  echo "; Generated: $(date) | Resolver: $ACTIVE_RESOLVER"
  echo
  types=(SOA NS A AAAA CNAME MX TXT SRV CAA DS DNSKEY)
  names=("$domain" "_dmarc.$domain" "_domainkey.$domain" "www.$domain" \
         "autodiscover.$domain" "enterpriseenrollment.$domain" "enterpriseregistration.$domain" \
         "lyncdiscover.$domain" "sip.$domain" "_sip._tls.$domain" "_sipfederationtls._tcp.$domain" \
         "s1._domainkey.$domain" "s2._domainkey.$domain" "selector1._domainkey.$domain" "selector2._domainkey.$domain" \
         "apple-domain-verification.$domain")

  if [[ -f "names/$domain.txt" ]]; then
    while IFS= read -r lbl; do
      [[ -z "$lbl" || "$lbl" =~ ^# ]] && continue
      names+=("$lbl.$domain")
    done < "names/$domain.txt"
  fi
  if (( ${#dkim_found_names[@]} )); then names+=("${dkim_found_names[@]}"); fi

  readarray -t names < <(printf "%s\n" "${names[@]}" | awk '!seen[$0]++')

  for n in "${names[@]}"; do
    echo ";; ===== $n ====="
    for rr in "${types[@]}"; do
      ans="$(f "$rr" "$n" || true)"
      [[ -n "$ans" ]] && { echo ";; -- $rr"; echo "$ans"; }
    done
    echo
  done
} > "$zonesfile"

# latest pointer for zones
if ln -sfn "$(basename "$zonesfile")" "$zones_latest" 2>/dev/null; then :; else cp -f "$zonesfile" "$zones_latest"; fi

echo "Zone snapshot written to: $zonesfile"
echo "Updated latest → $zones_latest"

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Image and Restore Macs Using Clonezilla

Wed, 13 Aug 2025 18:10:53 -0700

If you want to quickly image a Mac’s hard disk (create an exact copy saved as a compressed file) and later restore it, Clonezilla is a simple and reliable choice. This method is especially useful for older Intel-based Macs where internet recovery can be slow or unreliable. Here’s how you do it.

Preprequisites

An external SSD or hard drive to store the image
A USB drive to boot Clonezilla (created with Ventoy) [Ventoy is awesome, keep a USB stick handy with all sorts of ISOs!]

Download links:

1. Build the Bootable USB with Ventoy and Clonezilla

Download and install Ventoy on a USB drive.
Copy the Clonezilla ISO file (amd64) into the USB drive (Ventoy supports booting directly from ISOs).
Safely eject the USB drive once it’s ready.

2. Boot the Mac from the Ventoy USB

Insert the Ventoy USB drive into the Mac.
Power on the Mac and hold the Option (⌥) key during startup.
Select EFI Boot from the startup menu.
Ventoy will load, even if the Mac displays a “security violation” message — you can safely proceed.
From Ventoy, choose the Clonezilla ISO, and when prompted:
- Use GRUB to boot Clonezilla.
- Choose Load into memory and select Large fonts (helpful on Retina displays).

3. Create an Image of the Disk

In Clonezilla, choose Beginner Mode for simplicity.
Select device-image mode (save disk as an image).
Choose your external SSD or hard drive as the destination.
Use compression (zipped folder) to save space — this skips empty disk sectors.
Proceed with the imaging process. The system drive will be saved as a compressed image file.

This will take time, take a break. :)

4. Restore the Image to a Disk

To restore, follow the same initial steps:

Boot the Mac again using Option → EFI Boot → Ventoy → Clonezilla ISO.
This time, select Restore disk from image.
Choose the correct image file you previously saved.
Proceed through the restoration steps in Beginner Mode.

Since clonezilla is restoring using a disk image, you don’t have to worry about partition, etc.

5. Notes

Create the image when the machine’s just been setup.
This process is valuable for older Intel Macs (e.g., 2015 models) that might have problems with Apple’s Internet Recovery.
Although Time Machine is a built-in option, Clonezilla offers a faster, fully offline method and produces a single portable image you can archive. Works well when you use Clonezilla for all your disk imaging and restoration works.

_{Tested on MacBook Pro 2015 with macOS Monterey installed.}

🌱

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Toshiba Dynabook SS 2000 DS80P/2 – the thinnest laptop before MacBook Air

Mon, 23 Jun 2025 20:44:23 -0800

Back in 2006, when I was trawling the second-hand electronics stores in Akihabara, Tokyo, I found this extremely slim and light Toshiba laptop. I remember asking the owner about it and he said it did not work and did not have an operating system.

Steve Jobs would not unveil a similarly thin and light MacBook Air until 29 January 2008.

It ran Windows 2000 originally, it was okay with speed, then ran Windows XP and it was not bad.

Original specifications:
https://dynabook.com/pc/catalog/ss/02061820/spec.htm

I still have the machine. I have recently installed FreeBSD 14.3 (32-bit) on it. Since the system cannot boot from USB, I cloned a FreeBSD 14.3 VM and used nc (netcat) with dd to write the image across the network after PXE booting the Toshiba. It was fun figuring the process out. The combination of nc and dd worked perfectly.

I updated the hard disk to 30GB and added memory so it has 512MB in total.

Below is the full hardware output from the machine.¹

root@tosh:~ # dmidecode
# dmidecode 3.6
Scanning /dev/mem for entry point.
SMBIOS 2.3 present.
39 structures occupying 1187 bytes.
Table at 0x000EC000.

Handle 0x0000, DMI type 0, 20 bytes
BIOS Information
        Vendor: TOSHIBA
        Version: Version 1.40
        Release Date: 04/25/2003
        Address: 0xE0000
        Runtime Size: 128 kB
        ROM Size: 512 kB
        Characteristics:
                ISA is supported
                PCI is supported
                PC Card (PCMCIA) is supported
                PNP is supported
                APM is supported
                BIOS is upgradeable
                BIOS shadowing is allowed
                VLB is supported
                Boot from CD is supported
                Selectable boot is supported
                EDD is supported
                Japanese floppy for Toshiba 1.2 MB is supported (int 13h)
                3.5"/720 kB floppy services are supported (int 13h)
                Print screen service is supported (int 5h)
                8042 keyboard services are supported (int 9h)
                Serial services are supported (int 14h)
                Printer services are supported (int 17h)
                ACPI is supported
                USB legacy is supported
                AGP is supported
                BIOS boot specification is supported
                Function key-initiated network boot is supported

Handle 0x0001, DMI type 1, 25 bytes
System Information
        Manufacturer: TOSHIBA
        Product Name: DynaBook SS 2000 DS80P/2
        Version: PP20080P2F62
        Serial Number: 92036570J
        UUID: b2a01400-c6f9-11d6-8011-d61092036570
        Wake-up Type: Other

Handle 0x0002, DMI type 2, 8 bytes
Base Board Information
        Manufacturer: TOSHIBA
        Product Name: Portable PC
        Version: Version A0
        Serial Number: $$2Y303EZN

Handle 0x0003, DMI type 3, 17 bytes
Chassis Information
        Manufacturer: TOSHIBA
        Type: Notebook
        Lock: Not Present
        Version: Version 1.0
        Serial Number: 00000000
        Asset Tag: 0000000000
        Boot-up State: Safe
        Power Supply State: Safe
        Thermal State: Safe
        Security Status: None
        OEM Information: 0x00000000

Handle 0x0004, DMI type 4, 32 bytes
Processor Information
        Socket Designation: IC3
        Type: Central Processor
        Family: Pentium III
        Manufacturer: Intel Corporation
        ID: B1 06 00 00 00 00 00 00
        Signature: Type 0, Family 6, Model 11, Stepping 1
        Flags: None
        Version:
        Voltage: 1.1 V
        External Clock: 133 MHz
        Max Speed: 800 MHz
        Current Speed: 800 MHz
        Status: Unpopulated
        Upgrade: None
        L1 Cache Handle: 0x0012
        L2 Cache Handle: 0x0013
        L3 Cache Handle: Not Provided

Handle 0x0009, DMI type 5, 20 bytes
Memory Controller Information
        Error Detecting Method: None
        Error Correcting Capabilities:
                None
        Supported Interleave: Two-way Interleave
        Current Interleave: Unknown
        Maximum Memory Module Size: 512 MB
        Maximum Total Memory Size: 1024 MB
        Supported Speeds:
                60 ns
        Supported Memory Types:
                SDRAM
        Memory Module Voltage: 3.3 V
        Associated Memory Slots: 2
                0x000A
                0x000B
        Enabled Error Correcting Capabilities:
                None

Handle 0x000A, DMI type 6, 12 bytes
Memory Module Information
        Socket Designation: Soldering
        Bank Connections: 0
        Current Speed: 6 ns
        Type: SDRAM
        Installed Size: 256 MB (Single-bank Connection)
        Enabled Size: 256 MB (Single-bank Connection)
        Error Status: OK

Handle 0x000B, DMI type 6, 12 bytes
Memory Module Information
        Socket Designation: Toshiba unique
        Bank Connections: 2 3
        Current Speed: 6 ns
        Type: SDRAM
        Installed Size: 256 MB (Single-bank Connection)
        Enabled Size: 256 MB (Single-bank Connection)
        Error Status: OK

Handle 0x0012, DMI type 7, 19 bytes
Cache Information
        Socket Designation: CPU Internal
        Configuration: Enabled, Not Socketed, Level 1
        Operational Mode: Write Back
        Location: Internal
        Installed Size: 32 kB
        Maximum Size: 32 kB
        Supported SRAM Types:
                Other
        Installed SRAM Type: Other
        Speed: 1 ns
        Error Correction Type: Unknown
        System Type: Unknown
        Associativity: Unknown

Handle 0x0013, DMI type 7, 19 bytes
Cache Information
        Socket Designation: CPU Internal
        Configuration: Enabled, Not Socketed, Level 2
        Operational Mode: Write Back
        Location: Internal
        Installed Size: 512 kB
        Maximum Size: 512 kB
        Supported SRAM Types:
                Other
        Installed SRAM Type: Other
        Speed: 1 ns
        Error Correction Type: Unknown
        System Type: Unknown
        Associativity: Unknown

Handle 0x0017, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: MICROPHONE JACK
        External Connector Type: Other
        Port Type: Other

Handle 0x0018, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: HEADPHONE JACK
        External Connector Type: Mini Jack (headphones)
        Port Type: Other

Handle 0x0019, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: DOCKING INTERFACE PORT
        External Connector Type: Other
        Port Type: Other

Handle 0x001A, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: USB PORT
        External Connector Type: Access Bus (USB)
        Port Type: USB

Handle 0x001B, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: USB PORT
        External Connector Type: Access Bus (USB)
        Port Type: USB

Handle 0x001C, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: BUILT-IN LAN PORT
        External Connector Type: RJ-45
        Port Type: Network Port

Handle 0x001E, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: BUILT-IN MODEM PORT
        External Connector Type: RJ-11
        Port Type: Modem Port

Handle 0x001F, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: EXTERNAL MONITOR PORT
        External Connector Type: DB-15 female
        Port Type: Other

Handle 0x0020, DMI type 8, 9 bytes
Port Connector Information
        Internal Reference Designator:
        Internal Connector Type: None
        External Reference Designator: USB PORT
        External Connector Type: Access Bus (USB)
        Port Type: USB

Handle 0x0037, DMI type 9, 13 bytes
System Slot Information
        Designation: PCMCIA0
        Type: PC Card (PCMCIA)
        Data Bus Width: 32 bit
        Current Usage: In Use
        Length: Other
        ID: Adapter 1, Socket 0
        Characteristics:
                5.0 V is provided
                3.3 V is provided
                PC Card-16 is supported
                Cardbus is supported
                Modem ring resume is supported
                Hot-plug devices are supported

Handle 0x0038, DMI type 9, 13 bytes
System Slot Information
        Designation: SD CARD
        Type: Other
        Data Bus Width: Other
        Current Usage: In Use
        Length: Other
        Characteristics:
                5.0 V is provided
                Hot-plug devices are supported

Handle 0x0039, DMI type 9, 13 bytes
System Slot Information
        Designation: CF CARD
        Type: Other
        Data Bus Width: 16 bit
        Current Usage: In Use
        Length: Other
        Characteristics:
                5.0 V is provided
                3.3 V is provided
                PC Card-16 is supported
                Cardbus is supported
                Modem ring resume is supported
                Hot-plug devices are supported

Handle 0x0058, DMI type 10, 12 bytes
On Board Device 1 Information
        Type: Other
        Status: Enabled
        Description: USB
On Board Device 2 Information
        Type: Sound
        Status: Enabled
        Description: SOUND
On Board Device 3 Information
        Type: Ethernet
        Status: Enabled
        Description: ETHERNET
On Board Device 4 Information
        Type: Other
        Status: Enabled
        Description: MODEM

Handle 0x0059, DMI type 11, 5 bytes
OEM Strings
        String 1: PP200N0031K0,PCN1110V0Z01/S3A1156D001

Handle 0x0069, DMI type 12, 5 bytes
System Configuration Options
        Option 1: TOSHIBA

Handle 0x0080, DMI type 15, 23 bytes
System Event Log
        Area Length: 124 bytes
        Header Start Offset: 0x0000
        Data Start Offset: 0x0000
        Access Method: General-purpose non-volatile data functions
        Access Address: 0x0003
        Status: Valid, Not Full
        Change Token: 0x31121416
        Header Format: No Header
        Supported Log Type Descriptors: 0

Handle 0x0081, DMI type 16, 15 bytes
Physical Memory Array
        Location: System Board Or Motherboard
        Use: System Memory
        Error Correction Type: None
        Maximum Capacity: 512 MB
        Error Information Handle: Not Provided
        Number Of Devices: 2

Handle 0x0082, DMI type 17, 23 bytes
Memory Device
        Array Handle: 0x0081
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 256 MB
        Form Factor: Other
        Set: Unknown
        Locator: System 0
        Bank Locator: CSA 0
        Type: SDRAM
        Type Detail: Synchronous
        Speed: Unknown

Handle 0x0083, DMI type 17, 23 bytes
Memory Device
        Array Handle: 0x0081
        Error Information Handle: Not Provided
        Total Width: 64 bits
        Data Width: 64 bits
        Size: 256 MB
        Form Factor: Other
        Set: Unknown
        Locator: Toshiba unique
        Bank Locator: CSA 2 & 3
        Type: SDRAM
        Type Detail: Synchronous
        Speed: Unknown

Handle 0x0090, DMI type 19, 15 bytes
Memory Array Mapped Address
        Starting Address: 0x00000000000
        Ending Address: 0x000000A03FF
        Range Size: 641 kB
        Physical Array Handle: 0x0081
        Partition Width: 1

Handle 0x0091, DMI type 19, 15 bytes
Memory Array Mapped Address
        Starting Address: 0x00000100000
        Ending Address: 0x0001F0003FF
        Range Size: 506881 kB
        Physical Array Handle: 0x0081
        Partition Width: 1

Handle 0x00A0, DMI type 20, 19 bytes
Memory Device Mapped Address
        Starting Address: 0x00000000000
        Ending Address: 0x000000A03FF
        Range Size: 641 kB
        Physical Device Handle: 0x0082
        Memory Array Mapped Address Handle: 0x0090
        Partition Row Position: 1

Handle 0x00A1, DMI type 20, 19 bytes
Memory Device Mapped Address
        Starting Address: 0x00000000000
        Ending Address: 0x000100003FF
        Range Size: 262145 kB
        Physical Device Handle: 0x0082
        Memory Array Mapped Address Handle: 0x0091
        Partition Row Position: 1

Handle 0x00A2, DMI type 20, 19 bytes
Memory Device Mapped Address
        Starting Address: 0x00010000000
        Ending Address: 0x000200003FF
        Range Size: 262145 kB
        Physical Device Handle: 0x0083
        Memory Array Mapped Address Handle: 0x0091
        Partition Row Position: 1

Handle 0x00B0, DMI type 21, 7 bytes
Built-in Pointing Device
        Type: Other
        Interface: PS/2
        Buttons: 2

Handle 0x00B1, DMI type 22, 26 bytes
Portable Battery 
        Location: 1st Battery
        Manufacturer: TOSHIBA
        Manufacture Date: <BAD INDEX>  
        Serial Number: <BAD INDEX>
        Name: <BAD INDEX>
        Chemistry: Lithium Polymer
        Design Capacity: 0 mWh
        Design Voltage: 10 mV
        SBDS Version: Not Specified
        Maximum Error: Unknown
        OEM-specific Information: 0x00000000

⛵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

The orginal battery died, replaced it with a cheap one. :) ↩︎

Fastest Way to Release your Domain for use with Google Workspace

Tue, 03 Jun 2025 09:20:36 -0700

You bought Google workspace through a reseller (say Wix) but let it expire. Now you want to get a new Google Workspace account directly with Google but when trying to setup, you get an error,

"This domain is already is use....".

If you don’t have access to the admin email, there are two good options:

If you bought Google Workspace with a third party, you’ll have to ‘disconnect’ the domain from the Wix account. If you are successful, you’ll need to wait 7 days. That’s too long.

If you have access to your zone editor (which you will if you own the domain), you can fill this form and Google says they’ll get back to you in three days intead of the seven. https://toolbox.googleapps.com/apps/recovery/domain_in_use?visit_id=638914179557282795-311471662&rd=1. This is the fastest way we’ve figured.

Update

Google responded within 24 hours (actually 10 hours) and released it with us having to do nothing extra!

👂

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Reuse My Book Live 2012 for Use With DLNA With a Smaller SSD

Thu, 01 May 2025 13:21:52 -0700

We owned a My Book Live, with 1TB HD as a “fancy” NAS Drive probably released around 2012. Because Windows 11 now comes with default SMB 2.0 and upwards, you cannot use it unless you install SMB 1.1 (SMB 1.1 is insecure and old). I had some older 256GB SSDs and I wanted to replace the existing HD with the SSD and use it primarily to stream music from the drive. I also had saved a 3TB image of the firmware a couple years ago. Here’s how I did it.

1. Extract the Firmware Image

Run a Ubuntu Live Disk (ran version 24.x). Use Ventoy, makes your life a lot easier.

sudo apt install p7zip-full -y
7z x "mybook3tb (2020_10_23 23_11_18 UTC).7z" -o./mybook_image

This produces:

mybook3tb (2020_10_23 23_11_18 UTC).img

2. Write Image to the New SSD

Connect your SSD to the Ubuntu Live machine via USB. You can buy a kit or resue the board and cable from a dead external HD by taking it apart. Find your SSD device (example /dev/sdg):

lsblk

Then flash the image:

sudo dd if="mybook3tb (2020_10_23 23_11_18 UTC).img" of=/dev/sdg bs=64K status=progress
sync

3. Fix GPT and Create a New Data Partition

The data drive is large and it turns out, we can just reduce it to fit the SSD. So here you go:

sudo gdisk /dev/sdg

Inside gdisk, do the following:

d          # delete partition 4
n          # create new partition
4          # partition number
<Enter>    # default start
+220G      # safe size for 256 GB SSD
<Enter>    # accept default type (8300)
w          # write
y

4. Format Partition 4 on MyBook Live

This is the data partition that needs resizing.

Plug the SSD into MyBook Live and power it on.

We’ll need to SSH into it but SSH is not turned on. So go to the portal add ssh after UI so the ULR is /UI/ssh. Enable it.

SSH into the device. You need special commands because it only accepts ssh-rsa:

ssh -o HostKeyAlgorithms=+ssh-rsa -o PubkeyAcceptedKeyTypes=+ssh-rsa root@<IP_ADDRESS>
# Default password: welc0me

Format the partition:

mkfs.ext4 /dev/sda4

5. Mount It

mkdir -p /DataVolume
mount /dev/sda4 /DataVolume
df -h | grep DataVolume

6. Enable Auto-Mount on Boot

echo "/dev/sda4   /DataVolume   ext4   defaults,noatime,nodiratime,discard   0   0" >> /etc/fstab

7. Take the SSD to Mybook, connect it, power it on.

reboot

After reboot, the LED should turn green and the web UI will show Status: Good. But you cannot see the “Public” share. The easiest way to fix this is to go to the Mybook portal –> Utilities –> Factory Restore –> Quick Factory restore. Let it reboot and now your Public folder will be visible, your machine is ready to use with the new SSD Drive.

Note, I just removed the entire Mybook casing, connected the SSD and hung it up using screws and some of the pre punched holes on my network board. :) You can be innovative to reuse the case or create a new case.

Optional

Your machine is using an SSD instead of the regular HD so I had these enabled.

8. Optimizations (Optional)

Disable forced fsck:

tune2fs -i 0 -c 0 /dev/sda4

Disable swap:

swapoff -a
sed -i '/swap/d' /etc/fstab

Confirm /etc/fstab entry:

/dev/sda4   /DataVolume   ext4   defaults,noatime,nodiratime,discard   0   0

This process restores MyBook Live to a working state with a 256 GB SSD, boots correctly, and mounts the data volume automatically.

Lastly, save your music, let Mybook parse through and enjoy listening.

Remember, its a very old device with a very old Linux kernel so don’t use it for any data or sensitive stuff. Security experts would probably say not to run such devices in your network if you’re worried about security.

🐦

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Checklist for Setup of Google Workspace for Micro Small Business

Tue, 01 Apr 2025 09:20:36 -0700

This is a checklist when you setup Google Workspace Email for your small biz so you cover the basics and fundamental security. The steps are in a chronological order for maximum speed.

Preparation

Ensure the domain has never been used with Google Workspace - ask client for verification. Google “remembers” if a domain was setup earlier and even if it wasn’t used, it still gets bound. If you need help, checkout - https://sanjayregmi.com/posts/fastest-way-to-release-your-domain-for-use-with-google-workspace/

Setup

Create account → use unique email (use + sign in your gmail address). First user = admin, auto-licensed.
DKIM → Apps → Gmail → Authenticate email → add TXT → Start auth.
- Validate:
```
dig TXT google._domainkey.example.com @8.8.8.8
```

DMARC → add TXT _dmarc in DNS.

Example:

v=DMARC1; p=reject; rua=mailto:dmarc-reports@example.com; ruf=mailto:dmarc-forensics@example.com; adkim=s; aspf=s

Validate:
```
dig TXT _dmarc.example.com @8.8.8.8
```

Cloud Identity Free → Billing → Get more services → Activate.
Turn off auto licensing → Billing → Subscriptions → Settings. Test with dummy user.
Users → Create all, remove license from first admin, assign to real users, keep 2 admins (for backup).
2FA → Security → Overview → 2-Step Verification → enforcement → On from (choose a date, two weeks from now) → New user enrollment period → 2 weeks → Frequency, allow user to trust the device → Methods → Any except verification codes via text, phone call. Passwords Management: min 15–20 chars, enforce password policy at next sign in.
Passkeys → Security → Authentication → Passwordless → Allow.
Footer → Apps → Gmail → Compliance → Append footer (optional).
Disable forwarding → Apps → Gmail → End User Access → Automatic forwarding → Off.
Final check of DNS values - Run https://sanjayregmi.com/posts/find-dmarc-dkim-spf-nameservers-with-one-script/

Post setup - with users

Setup forwarding of existing email to workspace email.
Setup Bitwarden. https://prasna.ca/use-password-manager-today/

Optional

Add company logo - Account –> Account Settings.

Checklist

Business address
Domain access (Cloudflare, GoDaddy, etc.)
Credit card for billing
Names + usernames + current emails for all users
Aliases needed (info@, sales@, etc.)
Groups needed (support@ → multiple users)
Notes / special rules

Google Workspace email setup request form

Use this to collect data as noted on the checklist. https://tally.so/r/nP70NP

Advanced

If you’re creating more than 50 users, consider using GAM - https://github.com/GAM-team/GAM. It gives you command line access to manage your Google Workspace. It’s incredibly powerful and open source.

🚁

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Mac Terminal Commands Cheatsheet

Fri, 21 Mar 2025 12:55:31 -0700

Just a cheatsheet for terminal commands on a mac.

Find all updates
softwareupdate -l
Install all updates
softwareupdate -i

👢

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Remove Folders With Tilde in Windows

Mon, 10 Mar 2025 14:31:26 -0700

If you have folder names in Windows that start with a tilde (~), they cannot be easily copied or deleted or moved. Here’s the trick to remedy the issue.

Run this:

ren ~folder newfolder

But if you have a lot of folders inside a folder that has tildes AND the folder name with tilde is the same, you need to do a search and replace. Here’s the script:

for /r "c:\folderwithtildes" %F in (.) do (if exist "%F\~folderwithtilde" ren "%F\~folderwithtilde" folderwithouttilde)

Caution

As with all things that make changes, please use caution, run the script on sample data first.

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Erase Your Macbook M1 Harddisk the Easiest Way

Wed, 05 Feb 2025 15:48:53 -0800

Starting with Macbook M1s, you cannot simply delete partitions in the hard disk. If you are one of those that installs linux and edits partitions, you might get into a stage where you cannot create/edit partitions anymore. Diskutil commands also will not help you. As a last resort, you might want to reset the hard-disk and its not easy as in the old models. To make matters worse, you’re probably in recovery mode.

Here are the steps to erase the hard disk (reset it) successfully.

Steps

Start machine in recovery mode. For a Macbook M1, keep pressing the power button until the BIOS tells you its loading startup options.
Open terminal.
Type “resetpassword” and enter.
This opens a new dialog box. Click on the top left of the menu bar to find “Erase Mac”. The hard disk is reset, all partitions are deleted (all data is lost!) and the machine is now ready for a fresh install of the operating system.
Reboot the machine and your new install will go smoothly.

🏉

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Acer Machine Bios Update Stops Windows From Booting normally

Mon, 13 Jan 2025 12:54:26 -0700

Acer BIOS version R01-B1 has a bug where when you update the machine, Windows 11 will boot in safe mode because it cannot find the hard disk.

The root cause of the problem is that, the BIOS update changes the storage driver to VMD which Windows 11 doesn’t seem to support natively, at least as of this writing.

The solution is to hit DEL to go to BIOS –> Advanced –> Storage Configuration and disable VMD Controller. Restart the machine and it should boot normally.

Acer will probably fix this in another BIOS update but until then, this is the easiest fix. Otherwise, the driver for Intel RST VMD needs to be injected.

The above solution worked on Aspire TC-1760 with Win Home Premium Build 26100. The BIOS release date is 10/27/2022.

🍟

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

What to Do When You Cannot Onboard a Windows Professional OS Into Microsoft Defender After You Have Upgraded Windows Home Edition to Windows Professional

Thu, 05 Dec 2024 21:56:54 -0800

It’s well known that Microsoft Defender (all versions as of Dec 2024) require a Windows Professional license. Many users buy machines that have Windows Home. When we upgrade Windows Home machines to Windows Pro by buying an upgrade key, there are times when the upgrade misses out on installing some key components that are necessary to run the onboarding script for Microsoft Defender from security.microsoft.com. And the onboarding script gives an error like this: “The service name is invalid.”

Possible solution:

Ensure the machine has actually been upgraded to the Professional license. You can check under “About your PC” option.
Open command line as administrator and run:

DISM.EXE /Online /Get-CapabilityInfo /CapabilityName:Microsoft.Windows.Sense.Client

This will show whether “Sense” (which is Microsoft Defender for Endpoint Service’s internal name) is installed or not.

If its not installed, run, in the same window:

DISM.EXE /Online /add-capability /CapabilityName:Microsoft.Windows.Sense.Client

Allow some time for the system to complete the installation.

Run the onboarding script.

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Convert VCF file into CSV the easiest way with no data loss

Fri, 22 Nov 2024 15:54:40 -0800

Applications like Apple mail export contacts in VCF format. Outlook doesn’t understand VCF. So a conversion from VCF to CSV is required. Trying to upload the VCF file to Google will work but many incomplete or badly formatted contacts will be removed. This doesn’t bode well for data integrity.

The best way to convert a VCF with thousands of contacts is to download the free ThunderBird Client. It requires you connect to an email system like outlook.com which you can create new. Afte that, you can import contacts into a new Address Book. The best part here is that Thunderbird imports ALL the contacts, regardless of whether they’re duplicates, or if they don’t have first name and last names. Finally, you can export the address book to CSV format which can be easily and without errors be imported to Outlook.

👺

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Venom, Diabetes, Weight-loss, High Blood Pressure - Wonder How They're Connected?!

Wed, 13 Nov 2024 10:40:55 -0800

What have diabetes, ozempic, high blood pressure medication etc. have in common with the most powerful venoms of animals in the world? Venoms saves lives! Crazy.

Here’s the gift NYT article if you want to read it all it’s a long but very exciting read

Some important points

The discovery of exendin-4 (a peptide in Gila Monster venom) led to the creation of Ozempic and Wegovy.
Unlike poison, which is considered ‘passive’, venom needs to be produced, stored and deployed by animals so it’s a surprizingly complex protein with myriad of peptides doing a lot of things and hence a great recipe for research for new useful medications.
The key to the usefulness of venom for drug discovery is their interaction with cell receptors.
And the most chilling and unbelievable trait and I quote: “Jewel wasps sting cockroaches with a serum that turn them into “zombies” that willingly accompany the wasps back to their lair, to be devoured by their larvae”.

🏮

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Setting Up Microsoft Defender for Endpoint - A First Timer's Guide

Tue, 15 Oct 2024 16:03:29 -0700

Defender for Endpoint is a wonderful cybersecurity tool that works in Macs and PCs and in PCs works with Windows Defender, other Defender flavors that Microsoft produces etc. It has the following benefits:

Core Defender Vulnerability Management
Attack surface reduction
Next-generation protection
Endpoint detection and response
Automated investgation and remediation
Microsoft secure score for devices
Microsoft threat experts
Centralized configuration and administration, APIs
Integration with other Microsoft Solutions
Microsoft Defender XDR

When you buy a license for the first time, you’ll have to wait for 30 mins to an hour for it to be fully deployed. After that you need to download an ‘onboarding package’ which is visible under ‘Onboarding’ card on the Overview page and install it in the machine of your choice.

If you cannot find it there (MS keeps on making changes to the UI all the time), you can try it under Setting –> Endpoints –> Device Management –> Onboarding under security.microsoft.com. Bottomline is you need to wait for upto an hour to see that appear. Hopefully, MS will make it appear much faster.

Also, the onboarding package can be installed remotely into another machine via administrator powershell by typing the script name in full and hitting enter.

And lastly, you get to install Defender into five machines per license.

Microsoft Defender does not support Windows Home editions

Sadly, Windows Home devices cannot be onboarded to the fantastic Windows Defender for Endpoint (see minimum requirements here: https://learn.microsoft.com/en-us/defender-endpoint/minimum-requirements?view=o365-worldwide#hardware-and-software-requirements). If you try to onboard them using the onboarding script, this is what you get:

[Error Id: 15, Error Level: 1] Unable to start Microsoft Defender for Endpoint Service. Error message: The service name is invalid. For more information, visit: https://go.microsoft.com/fwlink/p/?linkid=822807

The only way out is to upgrade the home machines to Windows Professional.

Perhaps this reflects some business or sales strategy on Microsoft’s part, but it does raise a valid question: why is a feature available for Mac and Linux platforms, yet excluded from Windows Home editions? This discrepancy seems puzzling, especially since covering all major systems could offer more consistent user experiences and align with user expectations.

😵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

On Migration From Google Workspace to Microsoft Office - Some caveats and notes

Thu, 03 Oct 2024 16:28:20 -0700

If you’re migrating a large number of clients from Google Workspace over to Microsoft Office 365, there’s a bunch of software that allows you to do that. They require configuration but the task will be much more scalable, you’ll be able to move over hundreds of accounts with minimal time and hiccups. But if you’re doing only a handful of migrations and you have to install the Office 365 thick clients on client machines, doing it manually will probably be faster and cost effective. Either way, if you go the manual route, here’s a few things that need to be noted.

Get a Google Takeout. For a few GB accounts, its quite fast. The emails are exported in MBOX format and Outlook doesn’t like it but you can convert MBOX into PST using third party paid tools (like, Systools MBOX to PST Converter) and it’ll preserve the folder structure.
If current Google Workspace account is being used with Outlook, an OST file is created in outlook that has all the emails. You’ll need to remember to export it as the export creates a PST instead of an OST. And even though Outlook actually creates OST files, it cannot import them. And to make matters worse, if you’ve removed the Google Account, and even though you already have and OST file, Outlook cannot export it! Intricate! OST file is only ‘alive’ when still connected to the IMAP servers.
Upon switching from Google Workspace email to Outlook, when you go in to setup the new Outlook account, Outlook refuses to add the new account because an account with the same already exists (the Google Workspace one). A safer bet, if the usernames are the same (and they generally are because you’re migrating the user from one service to another), you can create a new username in Microsoft Office and make the existing one an alias. So when you go to add the new account, you can add it WITHOUT deleting the Google mailbox because you want a copy of the emails there as well. For example, if the email you’re trying to import is john@example.com, create a new email account in Outlook as john.doe@example.com. Import the account into Outook. And ONLY then make John.doe@example.com an alias of john@example.com. The idea here is to preserve existing emails so if in the next steps our PST migration fails, the user still has their old emails in Outlook to refer back to.
Importing a PST is not 100 percent perfect. It can fail silently and it can fail with an error. Also with Office 365 online, you can only import a PST in Outlook (classic) and not in Outlook (new) as I write this. Additionally, a large PST will take days for all emails to be ingested into Office online and until then, you cannot do a search and find all emails.
When the ingestion of PSTs fails for whatever reason (of course, you can run scanpst on them and see) but if you don’t have time or want to use a better mechanism, you can actually load a PST file. PST files that have failed to be ingested into Outlook online will be loaded and visible this way. On outlook go to File –> Open and Export –> Open Outlook Data File and then provide your PST. Once the PST opens, you can copy and paste entire folders into the main Outlook folders.

😺

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Block a Website Using Pfsense when you have to stop your kids and family from using some popular websites

Fri, 06 Sep 2024 14:33:02 -0700

Let’s face it. You’re supposed to do important work but want to listen to some music you remembered on youtube and then get carried away into watching something more interesting, forgetting you have a job at hand. Kids with homeworks due are the most vulnerable :).

One way to discourage this behaviour is to block websites like youtube or netflix if you have pfsense installed.

Steps to block websites

Login to pfsense portal.
Navigate to Services –> DNS Resolver –> General Settings
Scroll down to the bottom to “Domain Overrides”
Click “Add”.
Under “Domain” enter your URL. For example, “youtube.com”
Under IP Address, enter “0.0.0.0”.
Hit “Save”
Then hit “Apply” changes at the top of the screen.
The changes take effect immediately.

Caveats

All computers and devices on your network need to use your pfsense router as the DNS. On Windows 11, make sure you have “IP assignment” and “DNS server assignment” values to be “Automatic(DHCP).

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Microsoft Outlook Error: The Operation Cannot Be Performed Because the Message Has Been Changed

Mon, 12 Aug 2024 15:42:39 -0700

This mesage appears when editing a draft email in outlook and generally when using IMAP with/without a flaky wifi. The solution is simple:

Solution

Open Outlook, File -> Options -> Mail.
Uncheck all options under “Conversation Clean up”.
Save, quit Outlook and restart Outlook.

Workaround

If you don’t want to or cannot spend any time tinkering as explained above, you can use the online version of Outlook at either office.com or outlook.com. Online versions are not affected by this issue.

If the above doesn’t help

If this doesn’t solve it, repair/reinstall Office OR use the “Microsoft Support and Recovery Assistant” (SARA) tool.

😌

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

CrowdStrike Outage July 2024 - Technical Summary of What Happened

Wed, 24 Jul 2024 14:35:21 -0700

An update to CrowdStrike Falcon Sensor on 19 July 2024 caused about 8.5 million Windows machines (estimated) to hang and show BSOD (Blue Scren of Death). A fix was pushed out later that day but the machines that went down had to be manually fixed (as of this writing). Here’s a technical summary of what happened based on Dave Plummer’s video below:

Falcon Sensor runs on kernel mode! Executables that run on kernel mode need to pass WHQL (Windows Hardware Quality Lab).
Falcon Sensor was designed in such a way that an update to the sensor would be done through another file called a Channel File. This Channel File or simply a configuration file would be updated multiple times a day and would not go through WHQL process.
On the ill fated day, CrowdStrike sent a configuration file that was invalid. (We’ll probably know how this passed through when they publish their findings)
Falcon Sensor read the new configuration file and since it was invalid, threw an error. Since the Sensor runs in kernel mode, it threw a BSOD.

Remedy

For the affected machines, only manual fix is available. The steps are:

Boot the machine in safe mode.
Remove the offending file starting with C-0000291 under c:\System32\drivers\CrowdStrike directory.
Reboot the machine.
On devices with BitLocker encryption, one also needs the keys.

Prevention

There’s a setting in the Falcon Sensor to delay the update for upto 3 months. The default unfortunately is set to immediate update.

Video

Postscript

Here’s an interesting way an Australian Tax Firm used barcode scanners to read the long BitLocker keys! https://www.theregister.com/2024/07/25/crowdstrike_remediation_with_barcode_scanner/

Reference

CrowdStrike July 19 Outage Updates Page:
https://www.crowdstrike.com/falcon-content-update-remediation-and-guidance-hub/

💣

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Failed to Load Kernel 'Kernel' FreeBSD

Tue, 18 Jun 2024 14:35:21 -0700

Sometimes when you release upgrade freebsd, you might get the following error:

Failed to load kenel ‘kernel’

can’t load ‘kernel’

Type ‘?’ for a list of commands…

And the system doesn’t autoboot.

In times like these, you’re supposed to select “Option 6” and select older kernels, boot the system, redo the upgrade and the system should be repaired. In my case, that didn’t work.

There were no other kernels available, there was no “current” kernel either. So the next option for me was to select “Option 8”, “Boot Environment” and select not the default environment but the next one (you can actually cycle through the older environments). You can do this by selecting “E” followed by “A”. You can press “A” repeatedly to go through all the older environments. Once you boot properly (by pressing the “Enter” button on your keyboard), redo the upgrade and the system should repair itself.

However, if you know this error is because of missing bootcode for freebsd, you can run the last command (the gpart one) from here to install the bootcode.

🟧

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Upgrade FreeBSD to a Newer Version or Release

Mon, 10 Jun 2024 10:44:21 -0700

Commands:

freebsd-update upgrade -r 14.1-RELEASE
freebsd-update install
shutdown -r now
freebsd-update install
shutdown -r now

Notes:

“14.1-RELEASE” –> You can use replace it with “14.2-RELEASE” or any other release in that format.
Before running the final “shutdown -r now”, you might have to do an additional “freebsd-update install” if any application is asking for it.

👅

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Keep FreeBSD Updated With Security Patches

Mon, 10 Jun 2024 10:39:41 -0700

To keep FreeBSD updated with all the security patches, run these two commands:

freebsd-update fetch
freebsd-update install

When you run the first command, you can type “q” to get out of the prompt if you do not want to review all the changes so you can run the second command.

💦

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Git Error - Fatal: the Remote End Hung Up Unexpectedly

Tue, 04 Jun 2024 16:54:29 -0700

When you push commits to git, you sometimes might get into this error. I got this when pushing hugo code into git.


git push -u origin main
Enumerating objects: 61, done.
Counting objects: 100% (61/61), done.
Delta compression using up to 4 threads
Compressing objects: 100% (45/45), done.
error: RPC failed; HTTP 400 curl 22 The requested URL returned error: 400
send-pack: unexpected disconnect while reading sideband packet
Writing objects: 100% (54/54), 2.07 MiB | 5.37 MiB/s, done.
Total 54 (delta 14), reused 0 (delta 0), pack-reused 0 (from 0)
fatal: the remote end hung up unexpectedly

This is caused by git buffer settings. Run:


git config http.postBuffer 524288000

Then run the git push -u orgin main command. It should go through. More here: https://stackoverflow.com/questions/15240815/git-fatal-the-remote-end-hung-up-unexpectedly

😤

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

What Is Being an Entrepreneur

Wed, 24 Apr 2024 20:50:55 -0700

It’s been slightly over six months that I’ve been an entrepreneur for the first time in my life. During these months, I realize entrepreneurship is much more about you yourself than anything else.

It’s about discipline, consistency.

It’s about positivity, carrying it all the time, infecting others with it. And in return, being infected as well.

It’s about being aware of those tiny little progressions, celebrating them, using them as fuel, for yet another day.

It’s so much about you and so less about others. At least in these beginnings. Your dream. Your fascination. Your commitment.

And then there are people. People you meet. Converse with. Discover.

Beautiful, thoughtful, kind, generous people. People you’d not meet otherwise.

They inspire. They energize you.

The best part - they let you discover yourself more. Deeper.

Your own self.

🎀

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Disable Internet Access for a Period of Time Every Day or On a Schedule Using Pfsense

Mon, 22 Apr 2024 19:40:42 -0700

Say you want to disable internet access on a specific computer in your LAN/wifi between 11pm to 2am every day and you have the pfsense (https://pfsense.org) firewall properly setup. Here are the steps:

Assign static IP to the machine

Login to pfsense dashboard > Status > DHCP Leases
Find the current IP of the chosen machine
Give is static IP by clicking on the first plus sign (light one) on the right.

Create schedule

Log in to your pfSense web interface.
Navigate to “Firewall” > “Schedules”.
Click on the “+” button to add a new schedule.
Define the schedule for blocking internet access from 11pm to 2am. You’ll need to add two times because one schedule does not go past the midnight. Test a few times and hit save.

Create an alias for the computer

Go to “Firewall” > “Aliases”.
Click on the “+” button to add a new alias.
Create an alias for the specific computer’s IP address or MAC address. We’ll need this when setting up the firewall rule.
You can actually add more than one IP in the alias field. Either separate them with a space or add them one by one by clicking on ‘save’.

Create the firewall rule

Go to “Firewall” > “Rules”.
Select the appropriate interface (e.g., LAN) where the computer is connected.
Click on the “Add” button to create a new firewall rule.
Set the following parameters:
- Action: Block
- Interface: LAN (or whichever interface is appropriate)
- Protocol: Any (or specific protocols if needed)
- Source: Single host or alias (select the alias you created for the computer)
- Destination: Any
- Description: Provide a description like “Block Internet Access during 11pm to 2am”
- Click on “Display advanced” and select the schedule you created earlier under “Schedule”.

Save and Apply Changes

After configuring the firewall rule, “Save” it.
Apply the changes by clicking on the “Apply Changes” button at the top of the page.

VERY IMPORTANT - Order of the block rule

Ensure your block rule is BEFORE “default allow LAN to any rule” also called the “allow everyone” rule. Rules are processed top down so the block rules need to be applied BEFORE the allow rules.

Test

Test if the rule works. To do a quick test, you can remove the schedule, apply and see if it works. This way, you can check the block rule immediately.

Note: If you’ve installed TailScale (https://tailscale.com) on the client computer to be blocked, the rule explained above might not work.

Notes

You can copy the rules on the Firewall Rules page, create more aliases and add them separately. This is handy if you want to turn off internet in the Roku devices or smart TVs in the house, separately.

⛱️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Develop Locally in Drupal and Host Static Files on your Website

Sun, 14 Apr 2024 18:41:16 -0700

Background

Say you have an old website on Drupal 7 with plugins that don’t work with the latest versions of Drupal. You now cannot migrate over to Drupal 9 or 10 without doing a lot of manual work and or losing the functionality of your current plugins. If your requirements are still fulfilled by the old Drupal 7 website, there’s a solution to keep using it.

Summary of the post below:

Install DDEV on WSL (Windows), then setup your Drupal 7 site inside DDEV (https://ddev.com).
Migrate your existing Drupal site using Backup and Migrate Plugin (https://www.drupal.org/project/backup_migrate).
Then use Static Generator plugin (https://www.drupal.org/project/static) to render static files and upload them to say Cloudflare Pages! (https://pages.cloudflare.com)

This way, you can keep updating content on the Drupal 7 system without any fear of being hacked or corresponding software becoming obsolete while still maintaining the website.

The software stack that makes it possible are WSL, Docker, DDEV and of course Drupal.

Steps

Install WSL2 with Ubuntu.

Just follow this tutorial - https://www.drupal.org/docs/develop/local-server-setup/windows-development-environment/installing-drupal-with-ddev-in-wsl2-on-windows/installing-wsl2-ubuntu

Install Docker and DDEV

Follow this tutorial until “Keep DDEV up to date” heading - https://www.drupal.org/docs/develop/local-server-setup/windows-development-environment/installing-drupal-with-ddev-in-wsl2-on-windows/installing-docker-ddev-drupal-in-wsl2

Install Drupal Project

Under “Install a Drupal Project” heading - run the following steps from the above tutorial:
- mkdir sites
- cd sites
- mkdir my-drupal7-site
- cd my-drupal7-site
- ddev config –project-type=drupal7 –docroot=web
- ddev start
- Now, ddev composer create drupal/recommended-project:9.5.11 will fail. So skip that. We’ll go manual from here.
- Open my-drupal7-site/.ddev/config.yaml in an editor. For Drupal 7, change the “php_version” to “7.0”. Save it.
- Download and unzip your version of Drupal from drupal.org (https://www.drupal.org/docs/7/install/step-1-download-and-extract-drupal); use the “From the command line” instructions.
- After the extraction is complete, run ‘ddev launch’ in the browser. Grab the URL and add /install.php to its end to start the install. The database part is already setup so all you have to do is setup the first user (admin) account.
- You now have a locally working Drupal site.

Migration of the existing Drupal 7 site

Install Drupal Backup and Migrate plugin. Export the database.
Install Drupal Backup and Migrate plugin on your DDEV instance. Restore from the SQL database from your existing Drupal site using the Backup and Migrate plugin.
Now copy over the contents (all files and folders) of the existing Drupal site into the DDEV one.
Your DDEV instance should now be a clone of the existing one.

Note: If your existing Drupal site is less than 100MB gzipped, you can use the Backup and Migrate plugin to fully restore your existing website into DDEV.

Exporting to static site

Follow this on your DDEV instance. (https://sanjayregmi.com/posts/create-static-website-using-cloudflare-pages)

Benefits to this approach of hosting a website

Security Your CMS is not exposed to the internet, only static pages are loaded into your internet facing webserver.
Stop endless cycle of upgrade You can keep using old software and why not if they’re serving your purpose!
Cheaper hosting You can now host your site on a much cheaper plan because it’s all static content. There are free hosting options like Cloudflare Pages(https://sanjayregmi.com/posts/create-static-website-using-cloudflare-pages), Netlify, Github, Vercel to name a few.

💛

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Get an 'A' Rating From Qualys SSL Labs if Your DNS Is Cloudflare and You Use OpenLiteSpeed as Your Wordpress Web Server

Sun, 07 Apr 2024 17:43:31 -0700

By default, when you setup a WordPress site using OpenLiteSpeed (https://openlitespeed.org), you’ll get a “B” rating in Qualys SSL Labs SSL test at https://www.ssllabs.com/ssltest/. And even if you setup your minimum TLS version on Cloudflare to be TLS 1.2, Qualys reports your website still uses TLS 1.1. The following steps allow you to fix the certificate issue and get an “A” rating.

Steps

Ensure you’re setup to use TLS 1.2 as the minimum version on CloudFlare. Go to your domain name in CloudFlare –> SSL/TLS –> Edge Certificates –> Minimum TLS Version –> TLS1.2.
SSH to your server where you website is hosted. Open the OpenLiteSpeed port by typing sudo ufw allow from YOUR_IP to any port 7080. (This command works for Ubuntu.)
Go to the OpenLiteSpeed Server browser at https://YOUR_SSH_SERVER_IP:7080 from your own machine. Login.
Go to “Virtual Hosts”, select your webserver –> SSL –> SSL Protocol. Edit to add these: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256:TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384
Turn on Enable ECDH Key Exchange and Enable DH Key Exchange ().
Save.
Graceful Restart.
Run SSL labs test. Your site should get an “A” now, instead of the original “B”.

Post work and cleanup

Remove the firewall rule now by typing “sudo ufw delete allow from YOUR_IP to any port 7080”. (This command works for Ubuntu.)
Logout of https://YOUR_SSH_SERVER_IP:7080
Check if there are any other ports open by running sudo ufw status on the terminal of your SSH server. Apart from port 80 and 443, there should not be any.
Logout of your SSH server.

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Install Pfsense Firewall With an Existing Netgear MR60 Mesh Wifi With Two Access Points MS60s

Mon, 01 Apr 2024 13:12:43 -0700

I’ve always heard of pfsense as a great opensource firewall with a nice list of features. We have an MR60 with two APs (MS60) for a total of three access points and its a great piece of hardware providing good wireless through the two stories. I wanted to control internet access and MR60 has a module to do that but only for a yearly subscription. I thought this was a perfect time to setup pfsense so this blog is about installing pfsense where there’s an existing Netgear MR60 with two MS60s.

Table of Content

Required Hardware

First the hardware. I found this cool machine, all metal, no fan (QOTOM Q730G5 Barebone Mini PC - Intel Quad-core J4105, AES-NI, 5 Intel 2.5G LAN, 10Watts, Industrial Mini PC Firewall Gateway Router Q730G5 Barebone, No RAM, No SSD) - https://amzn.to/49CKUXZ). I could have used an old machine but I didn’t want to keep a machine on 24/7 that consumed upwards of 80 watts. The TDP for this machine is 10 watts, runs without any sound (I loved that part!) and is just perfect for something like a firewall.

Since this is a baremetal machine, I bought cheap 16GB laptop module (https://amzn.to/4aA0g0H).

I already had an SSD but it takes NVMEs are well. There’s no place to screw in the SSD but I just connected it (one SATA connector is available in the machine) and put it on the top of the board and screwed the whole thing back.

Software Installation

Get pfsense from https://www.pfsense.org/download/, 64 bit. I downloaded the ISO, unzipped it using WinRAR and saved it in Ventoy Flashdisk (https://www.ventoy.net).

Connect your keyboard, a monitor to the machine. It has HDMI for display and several USB 2 and 3 ports. Press F11 to choose to boot from USB.

Select all default options and reboot. If you need help, go through this. https://docs.netgate.com/pfsense/en/latest/install/download-installer-image.html

First configuration of pfsense

Upon reboot, you need to select the WAN port and at least one LAN port. Follow instructions in the prompts. Mine didn’t detect any ports in the device. So I did manual input and provided the names for WAN (igc0 for WAN) and LAN (igc1 for LAN). I used an ethernet port from my existing LAN to connect to the WAN port (Yes, you can do that!) to setup the WAN port.

Upon reboot, connect your LAN port to your computer and make sure you can access 192.168.1.1, the pfsense portal AND access the internet via that connection, provided you’ve connected the WAN port to your existing router. If both work out, you’re now ready for the next step.

I spent a long time to get this basic setup completed. Here’s what I’ve learned. If upon reboot, if you cannot access 192.168.1.1, there’s an issue with your setup. Use the ‘reset’ option in the pfsense menu to reset the setup and start again. This will just reset the configs and and you can reconfigure.

Connect pfsense hardware to your network

Unhook the pfsense from your lab, take it to where your home network switches/internet box is. Unhook the ethernet to MR60 (we’ll set this up later). Hook pfsense so that the WAN (igc0) port is connected to the data port in your ONT or your internet modem. The LAN(igc1) port should now be connected to your switch from where the whole of the house gets their internet.

Check all connected devices have internet.

Re-setup MR60

Bring the disconnected MR60 to your lab. Reset it. (Press the reset button until it starts blinking) After reset is complete, connect it to your lab’s switch that has internet connectivity. Yes, you’ll have to reset the MR60 so you can set it up again because you won’t be able to reconnect it to the network just like that. If you do, you’ll have weird connectivity issues.

Now setup the MR60 as a new setup using the Nighthawk app on your phone.

After setup, all your wifi devices will connected to the MR60. However, they’ll all have 10.x.x.x IPs since MR60 will run its own DHCP by default. If you’re cool with it, you can now stop and reconnect the MR60 to your main router. I wanted one DHCP server and all IPs to be 192.x.x.x so I did this one more step.

Use pfsense as the only DHCP server

Go to routerlogin.net on your browser, login. Go to “Advanced”–> Wireless AP –> Enable AP Mode –> Save. Let MR60 reboot. After reboot, all your wireless devices will be on the same network as your wired machines which is 192.x.x.x.

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Replacement for Ssh - Mosh (Mobile Shell) Is the Modern Secure Shell - use it starting today

Tue, 12 Mar 2024 08:10:43 -0700

If you are a regular user of ssh, you’re aware of some of its issues:

ssh breaks when there’s a disconnection and it doesn’t reconnect. I use tmux as a stopgap so I can go back to what I was doing.
typing on ssh terminal when the remote server is faraway is downright painful becuase the response times are so slow. One has to wait for the characters to be acknowledged by the server.
and when that time you want to hit a CTRL + C becuase you ran the RSYNC command wrong, it doesn’t accept Control C in time. Worse, the SSH terminal then breaks down.

Enter Mosh and it solves all the said problems and some more. It allows for:

roaming while you’re connected to the server (fascinating!)
there’s no lag when typing!
you can put your laptop to sleep and it reconnects when you wake it up!

Now you do have to install ssh, it runs on top of ssh but its a beautiful experience. You’ll never have to do ssh root@foo.bar, you’ll do mosh root@foo.bar. Install it and run it today. Go to https://mosh.org/ to learn more, setup and install. And if you’re like me, don’t miss this 10 min video where one of the creators explains it all in ten minutes. Kudos to Keith Winstein and Hari Balakrishnan!

🔵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Agriculture Drones in Greater Vancouver Region

Wed, 21 Feb 2024 22:42:19 -0800

A month ago at the Tradex in Abbotsford did I see drones being marketed for use in agriculture in the lower mainland. I thought drones for agriculture were a cool modern disruption because it would be so much cheaper to fly an automated drone than to own and fly little airplanes with pilots or huge (and costly) sprayer machines that need to be driven on land. I also thought it would be a good thing for the survival of the business when it goes from retiring parents to their kids - drones are something that the kids are growing up with, are familiar and excited about, while many might not want to touch a tractor! I also heard that there were regulations in place that obstructed the full scale use of drones and that the drone companies were fighting with the government over it.

It’s time countries like Canada understand the shifts in adoption of technology and make it much easier and faster for the farmers to use them rather than obstruct them. Just like many things today, Drones are probably our future, let’s adopt them and figure how best we can add value rather than delay or block them.

✈️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Rysnc Fails When Copying Files With Tilde: rsync: [sender] fstat failed: No such file or directory

Wed, 21 Feb 2024 21:58:08 -0800

Rsync Error:

send_files(71922, /source/TQR6~B)
rsync: [sender] fstat failed: No such file or directory (2)

Trying to rsync between two different machines, I found rsync fails when it finds files that have ~ (tilde) in their filenames. I initially thought this was an issue specific to copying from a Samba Share to FreeBSD zpool. I couldn’t solve the issue entirely but figured escaping the awful tilde was a way to finish copying without rsync crashing. Here’s the command:

rysnc -a --exclude='\*~\*'/source /destination

💢

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Unlocking FreeBSD's Potential: Learn, Use and Elevate Your Experience with Generative AI

Sun, 18 Feb 2024 20:26:34 -0800

The benefits of FreeBSD are numerous. (Just ask chatgpt!). For me, its a very stable system, one that runs on any hardware, old and new, installs very quickly, configurations are in text and the pkg system that installs extra software is simple and powerful. But for a long time, anyone wanting to setup a home server or a VM to try with had to read lots and practice lots and even after practice, if you forget the commands, which happens when you use it infrequently, you have to google feverishly and many a times, the results the search engine finds might not be correct or actually be wrong. Enter ChatGPT and generative AI and things have changed drastically, for the good!

You can now setup a system, configure it, install GUI of your choice, set it up as a Samba Server or connect with your NAS box, setup cron jobs for backups, install the ZFS filesystem adding multiple hard disks and a whole lot more. All you need to do is ask chatGPT when stuck and it’ll generally punch out correct answers or answers that need just a bit of tweaking. This is a boon to the home tinker and those who like to try something different than a Windows machine or an Ubuntu box.

Next time, you get hold of an old piece of hardware, convert it to your server, install FreeBSD and enjoy the beauty of setting it up to the way you like by getting help from ChatGPT.

Some ideas for the FreeBSD home server:

Run cron, backup your websites on a regular basis using wget, cron.
rysnc with your NAS server so you have two copies of your most valauble data.
Use rclone and cron to upload your most important data to backblaze B2 using their free tier.
Install tailscale and access your FreeBSD server from anywhere in the world.
Connect all sorts of old hard drives that you’ve collected over the years, parition with ZFS (just a single command is all that is needed!) and use them as backup drives to copy your data in multiple places.
Configure X Forwarding with SSH so you can if required run programs that need GUI (for example firebox)
Enjoy the simplicity and super fast response times of lynx and elinks when browsing the internet.
SSH from any other machine in the house and find your projects, work on them from the same machine.
Use hugo to setup your website and create posts on your freeBSD machine.

What are some of the tasks you do on your FreeBSD server?! What are you own experiences using ChatGPT for answers to FreeBSD commands and questions? Please share.

PS: And if you’re wondering how you can start with FreeBSD, here’s the first step.

🎵

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Read IDE Hard Disk Formatted With NTFS on FreeBSD

Sat, 17 Feb 2024 22:24:42 -0800

I came across one of my first hard disks when home so I took it out and brought it with me to copy the data out. It was an IDE HD so I bought a cheap IDE to SATA adapter and connected it to my FreeBSD machine to copy the data. (Feeding the power directly from the motherboard’s molex connector was more stable for the old hard drive.) Here’s the steps that worked for me.

Run pkg install fusefs-ntfs to install the NTFS read/write module.
Run echo ‘fuse_load=“YES”’ » /boot/loader.conf if needed to load the NTFS module at boot time, every time.
Create a mount point, mkdir -p /mnt/ntfs.
Run geom disk list to find the name of the disk. Say it is ada0.
Run fdisk ada0 to see which partition is NTFS formatted. In my case, it was the third partition.
Run ntfs-3g /dev/sda1 /mnt/ntfs to load the NTFS folder. Note ada0 is sda1. In my case, it complained of an unclean file but the partition was mounted read-only. Great for copying over the data!
If there are paritions in FAT32, run mount_msdosfs /dev/da0s1 /mnt/fat32 where da0s1 is ada0 and partition 1 and /mnt/fat32 is the mount point provided its been created earlier.

💪

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

ZFS: unsupported feature: com.klarasystems:vdev_zaps_v2

Fri, 16 Feb 2024 21:12:43 -0800

Note of caution:

Read the FreBSD Release Notes carefully at https://www.freebsd.org/releases/14.0R/relnotes/ especially the section called “Upgrading from Previous Releases of FreeBSD”.

If you recently upgraded to FreeBSD 14 but you get the above error displayed before the beastie logo (see image below), here’s a solution for those with the boot loader in the MBR and that use the GPT partition scheme.

First figure if your machine bios is efi or mbr by running efibootmanager -v. If yours is not efi, the message will be “efibootmgr: efi variables not supported on this system. root? kldload efirt?”
Run gpart show and note the disk (ada0, ada1) on which your MBR is written into. We’ll assume its ada0 for the next command below.
Now run gpart bootcode -b /boot/pmbr -p /boot/gptzfsboot -i 1 ada0, where ada0 is your disk.

Running anything like the above on your freebsd box might cause loss of data. Please exercise caution.

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to Install Tailscale on FreeBSD

Thu, 15 Feb 2024 22:33:23 -0800

Tailscale.com’s download page doesn’t list if its available for FreeBSD. But it does work on FreeBSD (tested on Release 14). Here’s how:

Install the package: pkg install tailscale
Edit /etc/rc.conf to add tailscaled_enable="YES"
Run service tailscaled start. If you get an error, reboot.
After reboot, run tailscale login. Grab the url, copy paste in a browser and approve the connection.
You can now ssh into the machine from another ’tailscaled’ machine from within your LAN or outside your home/office.

How to update Tailscale

Once installed, when you run the command tailscale on FreeBSD as root, you get a list of all the commands. Use tailscale update to update.

Key Expiry

You can choose to disable key expiry via Tailscale portal.

The SSH switch

Run tailscale like this: tailscale set --ssh for passwordless ssh logins all managed by tailscale. You only need to run this command once. The beauty of this approach is that you can disable ssh altogether on the system by removing ssh startup from rc.conf. It’s actually safer this way. Longread is available here.

Last updated: 04 Jan 2026

💃

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Create Your Own Secure Network Accessible Over the Internet Using Tailscale

Sat, 10 Feb 2024 22:18:07 -0800

It’s 2024 and there’s a fantastic tool available that can connect Windows, Mac, Linux, BSD machines together into a single network. Of course, that’s possible using samba, ssh, nfs or a whole range of other tools but tailscale makes the process super easy and very quick. And for upto three users, you can have a network of upto one hundred machines, free!

And for those who’re thinking where to use it, here’s a quick use case. If you own multiple machines at home (spouse and kids), managing them will be a whole lot easier if you connect them into one network. Interested? Here’s a quick instruction with video.

🏃

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Youtube's First Video from jawed, Neal Mohan and the Nature of Change

Wed, 07 Feb 2024 13:06:42 -0800

In the first line of the blog post this year by Neal Mohan the CEO of youtube is a link to the very first video that was uploaded to youtube, 18 years ago. ‘jawed’, the creator has a 19 second video where he talks about elephants having ‘really, really, long trunks’. Since his upload, until today, people have watched it over 300 hundred million times (and with Neal providing a link like that, I’m sure its going to rocket even more!), his channel has only that video and there are 4.45M subscribers to it. In terms of earnings, youtubers who have a million subscribers generally earn around 60K USD/year. Let’s step back for a moment. Did ‘jawed’ at that time realize how significant his upload would be?! Did we?!

This comes back to the fundamentals.

About trying new things. About not being afraid to try. About understanding the nature of change. That its by nature we don’t like change but if we have progress in our heads and the dream for progress lying deep in our chests, that we should be able to appreciate, create, propagate change. Its about understanding fear, trepidation, hesitation; that its natural and its okay. Using our heads to relieve the stresses created by our limbic brains. Its about being able to think of the future, what if, what if it does succeed when you have nothing to lose?! (They say part of achieving success is also by dreaming and wishing to be lucky.)

A simple video. One bold person. The beginning of a giant. We have chances for change everyday. Let’s understand its okay. And then proceed to doing it, propagating it. All the best!

🔥

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

What Is Magnetized Target Fusion

Thu, 01 Feb 2024 02:18:58 -0800

Vancouver Canada is an awesome place and did you know that in Richmond that we have a company that’s been working on Fusion Technology since 2002?! Its called General Fusion. But the best part is the Founder and Chief Scientist Michel Laberge explain with pure excitementment what it is all about. All you need to invest is 81 seconds. Watch it here and if you’re like me, you’re really going to enjoy it!

💥

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

On Tea, Intoxication and Meditation

Tue, 30 Jan 2024 14:05:06 -0800

Every time I come to see you
and go away
You make me think.
Lots.

Forgetting myself in intoxication
Of your presence
Of your meditative voice
That tells of stories far and wide
Of kindness, hope, confidence, beauty and strategy.
What a mix, I say to myself!

Time stands still or rather exists only there and nowhere else
All else is forgotten.
Almost like living in a different world.

The flow is uninterrupted.
I dare not disrupt.
For I want the river to flow. Flow in its own accord.
Dancing. Gushing.
Telling stories.

Listening, absorbing,
Being entertained.
Learning.
All that wisdom
That I get for free.
Along with the sky, the lake, the roads and the life below.
We’re listening together.
Being enriched.
Cared for.
Like huddled kids.

While the aroma of tea brings life alive in each and every sip.
The tea made by you.
For us.

💎

Listen to the poem

Your browser does not support audio element.

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Pxe Boot to Install Any Operating System the Quickest Way

Tue, 30 Jan 2024 12:18:06 -0800

If your (old) machine doesn’t have a CD ROM drive or does not boot from USB, PXE boot to re-install a new operating system is the best option if you do not want to buy any more hardware to solve the situation. In the past, setting up a PXE server was a hassle. The internet is littered with linux configurations on tftp server. But if all you want to do is to boot your machine and autoload the ISO, here’s the fastest way with minimal fuss.

On a windows machine, download and install AOMEI Backupper Workstation (trial version is good).
Download your installer ISO. Example, say your old machine is an i386 (ha!) and you want to install linux, here’s the location of an ubuntu mini.iso that works well. Note if your machine has 512MB or less of memory, you won’t be able to load large ISOs so I suggest you pick the mini.iso as they’re less than 100MB in size and will take you through the installation flawlessly.
Open AOMEI Bakupper Workstation –> Tools –> Select your ISO –> Start Service
On the destination machine connected via ethernet, start in PXE mode.

I’ve suffered a lot in the past to run PXE servers. This is the best and foolproof method yet!

Credit: How to Setup PXE Boot Server in Windows Easily and Securely (archive.ph)

💌

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

On Flowers and Memories

Fri, 26 Jan 2024 22:34:48 -0800

Reminds me
Of the times together.

Even in their death,
They’re sitting here,
Gorgeous and beautiful.

Reminding us of the temporal nature,
Of life,
And yet,
Living it all,
Until the end.

Thank you for the flowers.
I don’t think we’ve had flowers here lately.

They remind me,
Of life and death and beauty,
In the essence of it all,
Together, in one shot.

Flowers. Given by you. To us.
Memories.

🌸

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to configure SSH for a passwordless login into your remote machine

Fri, 26 Jan 2024 21:44:06 -0800

On local machine, run

ssh-keygen

This creates two files /.ssh/id_rsa.pub (public key) and /.ssh/id_rsa (private key).
Still on local machine, run

ssh-copy-id remoteusername@remoteserver

This copies your public key to remote server.
Now try doing ssh to remote machine, no password should be required.

if ssh-copy-id fails

Sometimes, ssh-copy-id gives an error and doesn’t copy the files. In that case, use this one liner but replace the remoteuser and servername:

cat ~/.ssh/id_rsa.pub | ssh remoteusername@remoteserver "mkdir -p ~/.ssh && chmod 700 ~/.ssh && cat >> ~/.ssh/authorized_keys && chmod 600 ~/.ssh/authorized_keys"

Ensure

Remote machine is running sshd.
Local machine has relevant terminal with ssh installed.

❗

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

On Text Messages and Emoticons

Mon, 22 Jan 2024 20:22:21 -0800

It’s weird and amazing at the same time. You’re somewhere out there. I know of your existence by these letters you send me by the medium of electrons which themselves if isolated do not exist (Heisenberg principle). And yet your emoticons convey the thoughts and emotions as powerful as you were here.

🐦

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Dream big, really big

Sat, 20 Jan 2024 23:35:11 -0800

Dream big!
So big that sometimes you almost want to suffocate while thinking about it.
Its the only thing that keeps us going
Gives us hope
A reason to move forward
A way to become better
Every day, every hour, every second

Do you have a dream?
Do you dare to dream?
Dream.
Its the first step. Take it.

And when you dream, dream big - really big.
So big that sometimes you almost want to suffocate while thinking about it.

Allow yourself, you’re free to dream, yes!
Its the first step.

❄️

Listen to the poem

Your browser does not support audio element.

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Canadian Government Support to Young Entrepreneurs and Existing Businesses

Sun, 14 Jan 2024 21:49:34 -0800

We did some research into grants, funding and special loans available to startups and existing businesses in Canada when we were establishing Prasna IT. This is a compilation that we believe will be useful to entrepreneurs like us. Here’s a compilation of what we found.

Table of Contents

Loans

Young Entreprenur

Your startup will get 20K funding by Futurepreneur Canada and 40K by BDC.
Loans are collateral free.
Payable over 5 years, interest only for the first year
Program provides free mentorship
And free templates to start your business

How to apply

Apply at futureprenur.ca

A young entrepreneur is one in the age between 18 - 39.

Existing Entrepreneur

If you already have a business, you can avail the following:

National Research Council Canada - https://nrc.canada.ca/en/support-technology-innovation/about-nrc-industrial-research-assistance-programuser-agent

Accelerate growth of small/medium sized businesses by providing funding for innovation

Sustainable Technology Development Canada - https://www.sdtc.ca/en/

Funds cleantech startup 50k-100k non-repayable contribution

SR and ED Tax Incentives - https://www.canada.ca/en/revenue-agency/services/scientific-research-experimental-development-tax-incentive-program/what-are-sred-tax-incentives.html
Strategic Innovation Fund - Funds for proof of concept programs - https://ised-isde.canada.ca/site/strategic-innovation-fund/en

Entrepreneur with at least a 12 month old business

Can get loans upto 250k

Principal postponement upto 12 months Can spend the funds on buying assets, investing in marketing, website or consulting and for line of credit

Apply at bdc.ca

Entrepreneur 24 months old

A variety of loans are available to you

Small business loan of 100k For commercial real estate purchase For technology financing For financing of technology companies

Apply at bdc.ca

Entrepreneur export focussed

checkout https://www.tradecommissioner.gc.ca/funding_support_programs-programmes_de_financement_de_soutien.aspx?lang=eng

Canadian Goverment Grants

Canada Digital Adoption Program (CDAP) - Helps businesses adopt digital techologies
2400 grant to setup an e-commerce sites -
15000 grant to boost business technology and upto 100k zero percent loan.
Checkout https://ised-isde.canada.ca/site/canada-digital-adoption-program/en for more.

On picture, body and soul

Sun, 14 Jan 2024 17:18:23 -0800

I didn’t do anything big,
I just updated my profile picture.
True it was 15 odd years ago that I last changed it.
But you were overjoyed. Ecstatic! You reacted. Lots.
You said you wanted to see more. See not only half a picture but the full one.
The full picture.

I said. Okay, I get it.
I dug in deep.
I tried.
I expressed. And then I shared. With you.
You were not amused.

Not many reactions.
Not like the picture update.
A mere picture update!

I know you read it. Statistics don’t lie. I see you in website analytics.
You visited. And then you left.
Like you came to knock on the door and left immediately after.
Almost like you changed your mind. At the very last moment.
Like you mistook it for someone.

Someone you didn’t know.
Someone you didn’t care for.
Someone you were scared to see. And meet.

And I thought I provided a picture.
A better picture than the digital image I shared.
Than my material face.
Than my material face.

I agree without the material, there’s no place for the soul to exist. And soul is special. Sublime.
I understand without the body, I cease to exist. “I” ceases to exist. (It’ll be a sad day?!)
But am I more than a picture.. I speak out aloud,
I ask myself,
Am I more than the body that carries my soul.

The nurturing I’ve recieved from you.
The teachings I’ve gotten from you.
The sacrifices you’ve done for me.
The knowledge you’ve imbibed in me.
The intellect.
The curioustiy.
Compassion. You’ve inculcated in me.

Am I not more than my picture, I ask myself, out aloud.

💕

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

On leaving and being left

Wed, 10 Jan 2024 21:30:54 -0800

Everybody leaves.
At some point.
Its a revolving door. Its people. Natural. Nothing special or disheartening.

Memories should be kept.
Cherished.
They allow us to be happy (or sad) even after the moments are long gone.

The source of energy is your love. For yourself, for others.
Your dreams. Your being.
When you smell the fresh air or when you touch the first snow.
The fur of the pet when it rubs against your fingers.
The smile of your friend.
The especially good food someone prepared for you.
In your name, in your anticipation.
The invite you didn’t expect.
The hugs and the camaraderie.
The warmth you can feel deep in your heart.
THAT will provide you unlimited energy.
Like energy itself, that which cannot be created nor can be destroyed (btw, do you know of anything of this nature?!).

What you feel as sorrow is self created.
Life’s always good, its serene, its beautiful.
Think I’m bluffing?!

Go out for a walk with who you share your heart.
Wear your nice dress.
Call your friends over for dinner.
Chat. Share.
You’ll be reminded how beautiful life is.
How special you are.

❤️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Install Hugo on FreeBSD, commit to Github and deploy to Cloudflare pages

Tue, 02 Jan 2024 14:38:15 -0800

Prerequisites

Install PHP
Install nginx

Install Hugo

pkg install gohugo (install as root or do sudo)
follow this after install - https://web.archive.org/web/20231202094641/https://developers.cloudflare.com/pages/framework-guides/deploy-a-hugo-site/
remember to remove hugo.toml from the root directory otherwise you’ll get a “not Found” error!

Tip

Install the GUI (maybe xfce or Kde5) so you edit md files, update them on the fly. Copy pasting and updating is easier with the gui and the terminal available. Refer to Hugo Cheatsheat if you need instructions to create the new site.
If you start getting weird errors from Git, just create a new hugo site, copy your contents and config.toml and follow instructions again. IMO Github has a lot to be done before it becomes user friendly. I’ll write aobut it separately someday.

🌶️

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Hugo Cheatsheet

Mon, 01 Jan 2024 21:24:34 -0800

Table Of Contents

Start work on Hugo

hugo new site my-site
cd to my-site (change directory to that folder)
add a theme - direct download, unzip and copy to themes directory under my-site (“etch” is a simple theme)
Start server with drafts visible: hugo server -D& (& to put the command in the background so we can still use that terminal)
Or just type hugo serve to start the server.
If you’re on a headless server, do this:

hugo server --bind 0.0.0.0 --baseURL http://your-machine-ip-on-which-hugo-is-running:1313/ --appendPort=false

Then you can see the rendering in real time by going to any browser on your internal network by typing:

http://192.168.1.99:1313

Content creation and editing

Create new post: hugo new posts/posttitle.md.
Create new page: hugo new newpage.md. Update config.toml file with a new identifier to add the new page menu.
Edit using vim. Path will be content/posts/<posttitle.md> for posts. Menu path will be content/<menu.md>

Push to Git - one time only commands

git init
git remote add origin https://github.com/<your-gh-username>/<repo-name>
git branch -M main
git submodule add https://github.com/LukasJoswiak/etch.git themes/etch (for adding the etch theme, you can choose your own theme and its git url)

Push to Git - each time the website needs to be updated.

git add .
git commit -m “Some Comment”
git push -u origin main

Your password is your PAT and NOT your GitHub password.

Pushing to git with username nad PAT is tedious. Use command once:

git config –global credential.helper store
to type your username and PAT once and your creds will be saved after you enter correctly the first time.

Themes

Themes can be changed on the fly by editing config.toml or hugo.toml. (Note the server might not update immediately, restart a few times if not.)

Here’s a simple hugo.toml for etch theme.


        baseURL = 'https://sanjayregmi.com'
        languageCode = 'en-us'
        title = 'Sanjay Regmi'

        theme = 'etch'
        enableRobotsTXT = true

        googleAnalytics = "Your G Tag"

        [markup.goldmark.renderer]
                unsafe = true

        [menu]
                [[menu.main]]
                identifier ='our-projects'
                name = 'our-projects'
                title = 'Our Projects'
                url = '/our-projects'

                [[menu.main]]
                identifier ='posts'
                name = 'posts'
                title ='Posts'
                url = '/posts'

Emoji Support

Add the following in config.toml
```
  enableEmoji = true
```

Google Analytics

Hugo comes with built-in support for Google Analytics. All you need to do is to add googleAnalytics = “Your Google Tag” in hugo.toml (Newer versions of Hugo suggest using hugo.toml instead of config.toml). However, as of April 2024, some themes are not working with the above instructions. So here’s an alternative approach that works. The instructions are for the “Etch” theme and you should be able to adjust it for yours.

Copy all folders and files from /themes/etch/layout/ and paste them to /layout.
Create a file called analytics.html under layout/partials/ and paste the contents of the script from google analytics, manual install instructions. Save it.
Edit head.html and add {{ partial “analytics” . }} before </head>. Save it.
Now run hugo server (confirm there are no errors. Then load the website locally –> inspect –> network and note gtag appears in one of the requests sent out.
Note you do not need to add anything to hugo.toml as we’ve hardcoded our Google Tag IDs.

Old method - does not work for many themes anymore as of April, 2024

Add the following in config.toml

googleAnalytics = “X-XX” where X-XX is the google analytics code for your website.

Grab the /layouts/partials/header.html file from your theme. Copy it to /layouts/partials. If the partial folder doesn’t exist, create it. (For some themes like hugo-coder, its the head.html file instead that needs to be edited)

Add the following at the top of /layouts/partials/header.html

  <head>
  {{ template "_internal/google_analytics.html" . }}
  </head>

Save header.html

Insert Table of Content

create a file called `table_of_contents.html’ in layouts/shortcodes. Add the following code:
```
  <div class="toc">
      {{ .Page.TableOfContents }}
  </div>
```
Insert {{< table_of_contents >}} anywhere in your post or page to generate a Table of Contents.

Insert an image

Use this format: where “images” is a folder under “static” and img1.jpg is the image file under “images” folder. Inside the squre brackets, you can add caption but is optional.

Insert YouTube Video

Use this anywhere you want, change the id and title. {{< youtube id=“7qvg5EdL3fQ” title=“W” >}} which produces:
A shorter, newer format is {{< youtube qvg5EdL3fQ >}}

Add last modified date to posts

Add the following to config.toml

  enableGitInfo = true
  [frontmatter]
  lastmod = ["lastmod", ":git", "date", "publishDate"]

This way, the last modified date is taken from git commit dates.

Add categories and tags to your posts

Edit file archetypes/default.md and add the following descriptors:
```
  categories = ['']
  tags = ['', '']
```
Save default.md
Note values for both category and tag should be inside single quotes. More values can be added when separated by a comma.

In your config.toml add the following:

  [taxonomies]
  tag = "tags"    

  [[menu.main]]
  identifier = 'tags'  
  name = 'tags'  
  title = 'tags'  
  url = '/tags'

Save config.toml.
The two descriptors should now appear when you create a new post.

Add headings to your post

Use # and a space to create headings. The more # you use, the smaller the heading will be.

🍄

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Redirect a website in Clouldflare

Fri, 29 Dec 2023 20:52:31 -0800

Login to Clouflare –> select your website –> Rules –> Redirect Rules –> Create Rule.
Set “All incoming requests” to Static, your URL to be redirected to, set Status Code to 301 (permanently moved.
Hit “Deploy”, wait for propagation.

🔴

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Create static website from Drupal 7 and deploy into Cloudflare Pages

Fri, 29 Dec 2023 20:50:12 -0800

To extract a static website from Drupal 7.x so it can be hosted on Cloudflare Pages, follow the instructions below:

Install plugin “Static Generator” from (www.drupal.org/project/static)
Run it.
All static files will be created under “static” folder in the root of your Drupal Installation after you complete initial setup.
Create a new Github repo, upload all files to Github and deploy into Cloudflare. Your website will be live in the URL you setup during deployment.

If you do not know git, use “Git Desktop”, load all the files, and sync it up. As soon as the sync is complete, if you’ve setup Cloudflare pages to autodeploy (default mode), your site will be deployed in Cloudflare near instantly.

🔴

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Resolve "Briefly unavailable for scheduled maintenance" error in Wordpress

Wed, 27 Dec 2023 20:52:43 -0800

Cause - happens when a plugin update gets ‘stuck’ or when trying to update multiple plugins at the same time in WordPress. If it doesn’t go away:

SSH to the server (or FTP)
Navigate to the web root directory. Its /var/www/html for Apache and Nginx generally.
Navigate to the root folder of the website as you might have more than one website running.
Find and delete the file called “.maintenance”. (rm foobar)

🎀

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Add your website to Clouldflare DNS using the free tier

Wed, 20 Dec 2023 20:49:52 -0800

Cloudflare free tier is too cool not to use for your websites. It provides DNS management, free optimizations, DDOS protection, free caching and quick response times. Here’s how to setup a website.

Login to Cloudflare, add your website, follow insturctions.
Change the NS of your domain to the two Cloudflare ones. Wait for 10 mins to 1 hour for propagation.
Cloudflare emails when your NS has propagated.
Go to DNS management, add the A name (link with the IP) and then a CNAME for WWW to your URL without www. Example:
A record, example.com, IP –> Save
CNAME, www, example.com

Leave both fields ‘proxied’

On your webserver, ensure you have certificate setup.
On the SSL/TLS panel, change the mode to “Full (Strict). If you have issues, change them back to ‘Flexible’ and investigate.
On SSL/TS –> Edge Certificates turn on HTTPS, Set Min TLS to 1.2, Turn on TLS 1.3, Automatic HTTPS Rewrites
On Speed –> Optimization –> Content Optimization –> Turn on all free features. Test turning on “Auto Minify” and verify you have no errors when using the website.

🔴

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Lesson for Startups - Frugality Amongst Others

Mon, 04 Dec 2023 00:51:06 -0800

There are only two ways to create revenue. Getting more business or by cutting/limiting your expenses while retaining clients. Frugality is thus very important for a startup where growth is not immediate and one might have to stretch their budgets again and again.

In the Economist article, success of some big Indian companies it describes is because of their focus on profitability, eschewing extravagant spending and nurturing paying customers.

Here are some examples of their frugality:

Building using free software: Zerodha is a software company but uses in-house (probably built with free software) help desk software. This saves them millions yearly.
Nurturing paying customers: Zerodha’s focus on nurturing paying customers is evident through their $2.50 account opening fee, a cost that clients willingly bear despite the potential for financial strain.
Hiring graduates but training them: Zoho rigorously trains recruits.
Advertising: Zerodha does not advertise or provide discounts.

Out here in the West where we’re all about consumption, frugality might not be easy but it definitely helps to build it as part of one’s organizational culture and perhaps in the company DNA.

💝

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Delete all alarms from iPhone

Sun, 26 Nov 2023 20:50:23 -0800

Tell Siri “Deleted all alarms” followed by a “yes”!

🍅

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Install GRAV on WSL based off Ubuntu

Sat, 25 Nov 2023 20:51:24 -0800

Install WSL (powershell command as admin “wsl –login”)

Inside WSL upon installation

install php
download latest version of grav with admin console, expand in your non-root user account
CD to the grav directory
Run “php -S localhost:8000 system/router.php”
Access Grav at localhost:8000/admin on the browser of your windows machine
install any component that Grav complains about (generally PHP libraries)
your site is at localhost:8000

🍉

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Configure FreeBSD on an Old Intel Machine with ATI Graphics

Wed, 01 Nov 2023 01:41:28 -0800

Steps

Default install using Ventoy USB with FreeBSD install ISO
update pkg by running to install any package ex. pkg install vim
pkg install bash
Install video drivers for ati - ref AMD section in https://docs.freebsd.org/en/books/handbook/x11/. This should take the screen to VGE mode with max resolution
Make default shell bash: chsh and edit (vim editor runs) the tcsh out with bash
install kde5, xorg, xfce, firefox-esr, mc, elinks, htop, libreoffice
allow root to login as ssh: edit /etc/ssh/sshd_config and set PermitRootLogin to yes; restart sshd

Tips

Make root writable in single user mode (if you cannot login as root because you messed up the shell or have some corruption:

Commands:

fsck -y
mount -u /
mount -a -t ufs
swapon -a

You can now edit files in /.

pkg breaks after upgrade (error - “ld-elf.so.1” … “not found”)

pkg-static -v
pkg-static install -f pkg
pkg upgrade -f

🔴

Last updated: 16 Jan 2026

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Vim Cheatsheet

Sat, 11 Feb 2023 14:24:52 -0800

This page is frequently updated :).

Didn’t know Vim offered spellcheck, handy when editing your md files on the ssh console; enter the following in Vim command mode:

  :set spell spelllang=en_ca

Paste from clipboard:

SHIFT + Right Click

Edit all files iside the directory

vim \* (open all files inside the directory)  
  :ls (shows all opens files in vim)  
  :w (write the file that is open)  
  :ls (show all files. The files that is open is marked)  
  :b filename (open the second file)  
  repeat until you're done.

🌲

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Koshi (Kosi) River, Koshi Barrage

Tue, 15 Nov 2022 20:51:36 -0800

Wondering what Koshi river in Nepal, the Koshi Barrage and Indo-Nepal relations are all about over this? I don’t have a quick brief here but read/watch through the information below and I hope you’ll have a better view than the often politicized information in regular newsmedia. Write to me if you have comments to share.

What Has Nepal Learned from Unequal Water Treaties with India? Rai Sharad Chandra, Joshi Ratala Dinesh Prasad https://www.researchgate.net/publication/352091045_What_Has_Nepal_Learned_from_Unequal_Water_Treaties_with_India
Institutional Determinants of Indo-Nepal Hydro Cooperation, Sandeep Bhardwaj https://cprindia.org/wp-content/uploads/2021/12/Institutional-Determinants-of-Indo-Nepal-Hydro-Cooperation_Abridged-Report_Sandeep-Bhardwaj.pdf
Basin-scale hydrology and sediment dynamics of the Kosi river in the Himalayan foreland Rajiv Sinha, Alok Gupta, Kanchan Mishra, Shivam Tripathi, Santosh Nepal, S.M.Wahid, Somil Swarnkar https://www.sciencedirect.com/science/article/pii/S0022169419300289
Nepal’s Dangerous Dams by Isabel Hilton, New Yorker, 2015 https://www.newyorker.com/news/news-desk/nepals-dangerous-dams
Kosi - Video during the building of the barrage - https://www.youtube.com/watch?v=I7oEllJpZZM

tags: Koshi River Nepal, Koshi River, Koshi Barrage, Koshi Nepal

🌺

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Sources of Energy in Canada

Tue, 15 Nov 2022 20:50:43 -0800

40pc Petroleum Products
36pc Natural gas
17pc Hydro
1pc Others

76% is from non-renewable sources.

2022 Data - https://web.archive.org/web/20221019161726/https://www.cer-rec.gc.ca/en/data-analysis/energy-markets/provincial-territorial-energy-profiles/provincial-territorial-energy-profiles-canada.html

🍓

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Teach Kids Crypto

Wed, 22 Sep 2021 20:53:04 -0800

Ask them to read in the following order (12-13 yrs old):

Bitcoin Whitepaper: https://web.archive.org/web/20210906010924/https://nakamotoinstitute.org/bitcoin/
After the dry whitepaper, this will get them excited: Silkroad - https://web.archive.org/web/20210917214326/https://www.wired.com/2015/04/silk-road-1/

🐾

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

How to make Nepali Momo

Fri, 30 Oct 2020 20:51:10 -0800

One of the best videos around!

🏮

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Limpiyadhura, Lipulekh

Sat, 20 Jun 2020 20:52:11 -0800

The following is a compilation of some insightful YouTube discussions when the issue of Limpiyadhura and Lipulekh was raging on.

Recommended if you don’t have a lot time

General List

Ratan Bhandari – informative
Bishwa Bandhu Thapa – provides policy insights
Dr. Ramesh Dhungel – informative
Dr. Dipak Gyawali– some good information
Achyut Raj Regmi – interesting to listen
Rukmagat Katuwal – some relevance
Dr. Keshar Bahadur Bhandari – good one
Ashok Mehata- border politics
Nepal China
Anti Indian movement
KP Oli on Nepal-India border
China on Lipulek – some ideas
Indian military in Kalapani
Indian scholars on Nepal India border – thought provoking
Reactive India
SD Muni Ashok Mehata and Rajan Bhattarai – thought provoking
Dr. Bijaya Joli, BJP India – appalling conversations
Saurav with AP News – revolutionary but less pragmatic
Dr. Prakash Chandra Lohani

More interviews of interest

🌷

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

History of Pokhara

Mon, 15 Jun 2020 20:50:59 -0800

Install FreeBSD via pxeboot Using mfsbsd for Systems That Have Less Than 512MB Memory

Tue, 09 Apr 2019 13:30:37 -0800

If you are PXE booting a client machine with an ISO file that is bigger than available memory on your client, the PXE boot will probably fail. This is because PXE boot is copying the ISO into memory. Nowadays, most of the OS install ISOs are more than 512MB, even for FreeBSD. So here’s the process to install FreeBSD on your very old hardware with 512MB memory or less.

Go to https://mfsbsd.vx.sk/
Grab an iso based on your architecture. For example I got the iso from here - https://mfsbsd.vx.sk/files/iso/10/i386/ as my old machine was an i386! (Yes, its not the latest FreeBSD but I can always upgrade).
Setup tftpboot
Once booted, run: zfsinstall -d /dev/ada0
If needed to remove partitions use destroygeom -d /dev/ada0
Reboot
bsdconfig and run network setup so the ethernet card is up and running
pkg install pkg to install the package manager

Configure basic FreeBSD tasks

Install efl and xorg
Edit .xinitc file to include exec englightenment_start startx
Edit rc.conf to include sshd_enable=“YES”; edit /etc/ssh/sshd_config to allow root login and X forwarding

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Split Windows in Text Mode in Freebsd and Linux

Mon, 01 Apr 2019 13:55:36 -0800

tmux is a beautiful console program that allows you to split your windows. Very handy with today’s large screens.

run tmux
CTRL + B “ for horizontal split
CTRL + B % for vertical split
CTRL + B o to move from one window to another
a cheatsheet lives at https://tmuxcheatsheet.com

🦌

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

Why Read Ulysses?

Wed, 13 Feb 2019 13:51:36 -0800

To those who’re used to reading books where there is a plot, events happen and the story moves to an end or a goal, the “stream of consciousness” technique comes rather as a surprise. Some like it, many others find it dense and cannot carry on after a few pages.

If you enjoyed dipping your feet into the consciousness technique in “To the lighthouse” by Virginia Woolf, the one to go next is definitely “Ulysses” by James Joyce. The problem with Ulysses is that its considered one of the most difficult to read books. Searching for why one should read it and how to read it, I came across an article by the Economist which I simply could not read and forget. Of course, it mentions we need three other books to read the one but it should be an amazing experience, regardless.

To those that first read Woolf and want to have a general comparison between the two, this article provides quick info.

[web] · [in] · [fb] · [ig] · [x] · [tt] · [yt]

arm on SANJAY REGMI

What to do When You Cannot Find 'Microsoft Print to PDF' Option When Trying to Print to PDF

Two Quick Steps to Upgrade Windows Home to Windows Pro with a cheap license

Fix Upgrade Issues After Using Pkgbase on FreeBSD 15.x

Fastest Way to Copy Ext4 Formatted Data to FreeBSD Using WSL2 as an Intermediary

How to Get the Best Throughput - Both Upload and Download - From Remote Servers When Connected via Tailscale and Using Pfsense Firewall

Diagnosis

The Fix 1

The Easiest and Fail Safe Way to Set Up Scan-to-Email on Your Scanner With Office 365

Stage 1 – Preparation

Stage 2 – Printer setup

Benefits of this approach

Further Reading

Swap Growing on Your Freebsd Server? Check Your php_fpm Settings - Step 2

Swap Growing on Your Freebsd Server? Check Your Database Settings - Step 1

Setup Website Monitoring Properly So You Know if Your Database Connection Goes Down

Exploit the benefit of security and stability of FreeBSD by running webservers in jails along with reverse proxy jail, database jail and a redis jail

Scripts

Some specs

Notes

How to Enable Sending Email From Your FreeBSD Machine Using Dragonfly Mail Agent (DMA)

Prerequisites

Part one - setup DMA

Setup of DMA for FreeBSD 14.x

Install DMA

Backup original config

Configure DMA

Set up authentication

Secure auth file

Enable DMA in rc.conf

Set up mailwrapper

Quick Test

Monitor logs

Setup of DMA for FreeBSD 15.x

Part two - make cron send emails to you

Upgrading Proxmox 8 to 9 on Macbook Pro 2015

Hard Disk Noise and Seagate IronWolf Pro 12TB

Enjoy Fedora Command Line Input (CLI) colors? Apply it quickly to your Ubuntu and Debian machines

The script

Turn your Proxmox host into a fully functional Apple Time Machine backup server using Samba + Avahi — compatible with all modern macOS versions

1. Connnect your USB disk and mount it

2. Install Samba + Avahi

3. Create Time Machine User

4. Configure Samba

5. Advertise via Bonjour (Avahi)

6. Test Locally

7. Connect from macOS

8. Verify on Proxmox

If You Are Constantly on SSH and Having to Run Long-Running Processes Then Use Tmux

Note

Combine Lots of Video Clips Into One Very Quickly

Get a list of files in the order you want:

Create the concatenation list

Run ffmpeg to assemble into a single file

Final check

Set Up Your Own Photos Server Instead of Google Photos or Something Else by Using Immich and Tailscale

Setup Immich Server

Setup Immich Server URL with Tailscale

Upload your media (for a large amount)

Update your Immich server (installed via docker container)

Set up backup jobs to at least another disk (ideally three copies)

Enjoy!

Finally You Can Install Latest Versions of Mac OS on Old Hardware

How to do disk cloning when Clonezilla doesn't list the external source disk

Find DMARC DKIM SPF Nameservers With One Script

Features

How to run

Image and Restore Macs Using Clonezilla

Preprequisites

1. Build the Bootable USB with Ventoy and Clonezilla

2. Boot the Mac from the Ventoy USB

3. Create an Image of the Disk

4. Restore the Image to a Disk

5. Notes

Toshiba Dynabook SS 2000 DS80P/2 – the thinnest laptop before MacBook Air

Fastest Way to Release your Domain for use with Google Workspace

Update

Reuse My Book Live 2012 for Use With DLNA With a Smaller SSD

1. Extract the Firmware Image

2. Write Image to the New SSD

The Fix ¹