#January 2026

Wow new year! Should review all these notes at some ponit for a bit of reflection on the year.

Feels like problem after problem I'm solving in pursuit of the "perfect system" but it will never be perfect lol

Anyway....

#Revisiting lsof

So gemini came up with this, using lsof to listen to open files to find processes that are listening and finding ports. Context is i wanted to find the port that sabnzbd was listening on!

➜ jollof dev-setup (main) ✗ sudo lsof -i -P -n -c sabnzbd | grep -i listen
lsof: WARNING: can't stat() fuse.gvfsd-fuse file system /run/user/1002/gvfs
      Output information may be incomplete.
lsof: WARNING: can't stat() fuse.portal file system /run/user/1002/doc
      Output information may be incomplete.
systemd        1            root 729u  IPv4  17381      0t0  TCP 127.0.0.1:631 (LISTEN)
sshd      138238            root   6u  IPv4 246743      0t0  TCP *:22 (LISTEN)
sshd      138238            root   7u  IPv6 246745      0t0  TCP *:22 (LISTEN)
systemd-r 138246 systemd-resolve  12u  IPv4 264651      0t0  TCP *:5355 (LISTEN)
systemd-r 138246 systemd-resolve  14u  IPv6 264659      0t0  TCP *:5355 (LISTEN)
systemd-r 138246 systemd-resolve  23u  IPv4 264667      0t0  TCP 127.0.0.53:53 (LISTEN)
systemd-r 138246 systemd-resolve  25u  IPv4 264669      0t0  TCP 127.0.0.54:53 (LISTEN)
.postgres 138341        postgres   6u  IPv6 270343      0t0  TCP [::1]:5432 (LISTEN)
.postgres 138341        postgres   7u  IPv4 270344      0t0  TCP 127.0.0.1:5432 (LISTEN)
syncthing 138406          jollof  13u  IPv6 270360      0t0  TCP *:22000 (LISTEN)
syncthing 138406          jollof  29u  IPv4 267453      0t0  TCP 127.0.0.1:8384 (LISTEN)
Prowlarr  180815        prowlarr 340u  IPv6 407811      0t0  TCP *:9696 (LISTEN)
Radarr    180819          radarr 359u  IPv6 396267      0t0  TCP *:7878 (LISTEN)
Sonarr    180822          sonarr 366u  IPv6 405579      0t0  TCP *:8989 (LISTEN)
nzbget    180836          nzbget   3u  IPv4 389849      0t0  TCP *:6789 (LISTEN)
python3.1 180925         sabnzbd   7u  IPv4 403487      0t0  TCP 127.0.0.1:8080 (LISTEN)
markdown- 192067          jollof  25u  IPv4 448020      0t0  TCP *:8336 (LISTEN)
➜ jollof dev-setup (main) ✗

Can see above its 8080

#NixOS Module System - Auto-injected Args

In flake.nix I explicitly create pkgs-unstable:

pkgs-unstable = import nixpkgs-unstable { ... };
commonSpecialArgs = { inherit inputs commonGroups pkgs-unstable; };

But in modules like nixos-base.nix, I use { pkgs, pkgs-unstable, ... } - where does pkgs come from?

pkgs is automatically provided by nixpkgs.lib.nixosSystem. When you call it, NixOS evaluates nixpkgs for your system and injects pkgs into all modules.

specialArgs is only for extra args NixOS doesn't know about (like pkgs-unstable).

Other auto-injected module args: config, lib, options, modulesPath

Docs: https://nixos.org/manual/nixos/stable/#sec-writing-modules

#Inherit in nixos

Brief refresher on inherit (i always forget)

nix-repl> let a=10; in {b=a+20;a=a;}
{
  a = 10;
  b = 30;
}

nix-repl> let a=10; in {b=a+20; inherit a;}
{
  a = 10;
  b = 30;
}

nix-repl>

#Bash piping args

Well I'm finally getting round to trying to get my head round -s and -- in bash!

So.... -s is to take commands from stdinput. so with piping, can see without this it expects a file

➜ joelyboy ~ bash echo hi /run/current-system/sw/bin/echo: /run/current-system/sw/bin/echo: cannot execute binary file
➜ joelyboy ~

ahhhh, not this didn't work as -s expects from STDIN but i passed them as arguments! so it just opens a shell

➜ joelyboy ~ bash -s echo hi

[joelyboy@desktop-work:~]$ ^C

here is a proper working example:

➜ joelyboy ~ echo 'echo hi' | bash -s
hi
➜ joelyboy ~

Now, as we know in bash double quotations "" expand args, so below will try to expand $1, and find nothing!

➜ joelyboy ~ echo "echo Argument is: $1" | bash -s 'apples'
Argument is:
➜ joelyboy ~

Whereas this works! as $1 is NOT expanded before piping

➜ joelyboy ~ echo 'echo Argument is $1' | bash -s 'apples'
Argument is apples
➜ joelyboy ~

So, interestingly then, by default bash takes first param as script path to run and then $2, $3 etc as positional args to pass to the script. I.e. bash testie.sh 'hiya' runs testie.sh with $1 as 'hiya'

BUT, if -s is passed the first one from stdin is the script and then all args following are positionals to the script taken from stdin

Finally, when we see -- it just means to not treat -some_opt stuff as options. Like this where -apples I guess bash thinks is an option (although interesting it didn't complain or throw an error)

➜ joelyboy ~ echo 'echo Argument is $1' | bash -s -apples
Argument is
➜ joelyboy ~ echo 'echo Argument is $1' | bash -s -- -apples
Argument is -apples
➜ joelyboy ~

#lsof & ss deep dive (socket files)

lsof columns (man page):

FD suffixes (docs):

• u = read+write
• r = read only
• w = write only
• cwd = current working dir
• txt = program text (code)
• mem = memory-mapped file

COMMAND column: Yes, systemd means the process named "systemd" (PID 1, the init system). It holds the socket FD because systemd created it via socket activation.

ss columns (man page):

u_str LISTEN 0 4096 /run/docker.sock 13739 * 0 users:(("dockerd",pid=1879,fd=3))

users field: users:(("dockerd",pid=1879,fd=3),("systemd",pid=1,fd=595))

• List of processes with this socket open
• Format: ("command",pid=PID,fd=FD_NUMBER)
• Multiple entries = multiple processes sharing the socket (socket activation)

Why 2 entries for docker.sock? Systemd creates the socket (socket activation), then passes FDs to dockerd. Both hold references.

Docs:

• lsof: https://man7.org/linux/man-pages/man8/lsof.8.html
• ss: https://man7.org/linux/man-pages/man8/ss.8.html

#API/HTTP requests client

So having used postman in the past and wanting to move to a vim based giga-chad 💪 TUI will review the options:

• atac: crashed on importing cloudflare openAPI schema (it is large but not great....)
• posting: more promising but cloudflare uses openapi spec 3.0.3 and posting based off openapi-pydantic package which likes 3.1.x
• resto: 281 stars... last release 2022
• resterm: lets go: still struggling with errors trying to import metabase openAPI json (which IS a valid openapi 3.1 so I assume same problem as the others, not updated for latest)
• openapi-tui:
• atac (UPDATE): trying v0.22.1 from unstable nix packages: same issue with metabase API....
• resterm (UPDATE): trying v0.21.3 from unstable nix packages: same problem
• openapi-tui: DOES work for v.3.1 openapi BUT i try the :send command and nothing happens....

Ok, this seems to be a common theme... found a link here that explains exclusiveMinimum which seems to be plaguing these TUIs

So... lets try a down-converter for openapi?

#Nixarr setup

Decided to go with nixarr (lets give it a try).

At some point i would like to review using configarr or whatever its called to sync config files...

Because i keep forgetting (and they changed the ports)... its:

• radarr: 7878
• prowlarr: 9696
• sabnzbd: 6336 (changed from 8080)

• Step 1 * so go to sabnzbd:6336 to setup.

Newshosting is the ... download server?

Then just add host new.newhosting.com - (then get username and pass - saved in bitwarden, can login to dashboard here: https://controlpanel.newshosting.com/customer/)

important to note that on the newhosting dash here it states 30 max connections so in the server config i set the connections to 30

• Step 2 * Using nzbgeek as the ... indexer?>

Just have to go to prowlarr and add indexer...

Can get API key from here on nzbgeek but i think its the same as the password?

• Step 3 * Have added in prowlarr apps here radarr so that radarr can pull the indexer from prowlarr (i.e. just nzbgeek at the moment)

• Step 4 * Go to radarr download client and add sabnzbd

#Recyclarr

So this is not in true NixOS style but its nice to have a tool to configure trash guides that is NOT through the UI.

Configarr worth looking at after for a more "complete" declarative option

#Radarr/Sonarr Quality System (Claude)

The quality system can be confusing. Here's how it all fits together:

1. Quality Profile Structure:

2. Quality Groups (expand a ranking):

3. Custom Formats:

4. How matching works:

The Hierarchy:

The Big Picture - How It All Fits Together:

1. Source = WHERE did the video come from?

Bluray IS a source - it means "ripped from a Bluray disc"

NF/AMZN are service tags - they tell you WHICH streaming service, but the source is still "WEB"

2. Remux = a FLAG, not a source

Remux only applies to disc sources. WEB-DL is already "not re-encoded" but we don't call it remux.

3. How Radarr builds a "Quality":

A "Quality" is just Source + Resolution combined. Radarr has ~30 built-in qualities.

4. The Full Tree:

Quality Groups explained:

A group bundles multiple qualities as EQUAL rank. Inside a group, no upgrades happen between members.

Quality Group: "WEB 1080p"
├── WEBDL-1080p   ─┐
├── WEBRip-1080p  ─┼── all scored the SAME
└── (any WEB src) ─┘

Won't upgrade WEBRip → WEBDL
But custom formats CAN still prefer one via scoring

5. Where do Custom Formats come from?

Custom Formats are just pattern matchers YOU define (or import). They scan release names and add/subtract scores.

6. NF: Source vs Custom Format - what's the difference?

RELEASE: "Movie.2024.1080p.NF.WEB-DL.x264-GROUP"
                          │
         Parser extracts: │
                          ▼
┌─────────────────────────────────────────────┐
│ Source: WEB-DL        ← Radarr built-in     │
│ Resolution: 1080p     ← Radarr built-in     │
│ Quality: WEBDL-1080p  ← Source + Resolution │
└─────────────────────────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────┐
│ Custom Format "NF" checks:                  │
│   Condition: title contains "NF"? YES       │
│   → Apply score: +50                        │
└─────────────────────────────────────────────┘

Final score = Quality rank + Custom Format scores

"NF" in the filename is just text. The Custom Format "NF" is a rule that DETECTS that text and adds score.

Jargon Summary Table:

Filename example decoded:

Movie.2024.1080p.BluRay.REMUX.AVC.TrueHD.7.1-FraMeSToR
       │    │      │     │    │    │   │      │
       │    │      │     │    │    │   │      └── Release group
       │    │      │     │    │    │   └── Audio channels
       │    │      │     │    │    └── Audio codec
       │    │      │     │    └── Video codec
       │    │      │     └── Remux flag (not re-encoded)
       │    │      └── Source: disc
       │    └── Resolution
       └── Year

Scoring Example:

In your quality profile, custom formats get scores:

When a release matches multiple formats, scores ADD UP. Higher total = preferred.

Cutoff vs Upgrade:

Current: WEB-1080p (score: 500)
Cutoff:  Bluray-1080p (score: 1000)

→ Will upgrade to Bluray because current < cutoff

New release: Bluray-1080p + "Tier 01" format (+1800)
Total score: 1000 + 1800 = 2800

→ Upgrades because new score > current score

Why Groups?

Quality groups treat multiple sources as equivalent. E.g., "WEB 1080p" group:

• WEBDL-1080p (Netflix direct)
• WEBRip-1080p (screen capture)
• AMZN WEB-DL

All ranked the SAME. Won't upgrade WEBRip→WEBDL, but custom format scores can still prefer one.

Docs: https://wiki.servarr.com/radarr/settings#quality-profiles

#Quality profiles (my notes)

Well.... what claude wrote is... verbose....

Will try to simplify my best i can do

#Git diffs

So have played around with diffs in neovim but nevery really understood how it worked....

Got ChatGPT to outline a super simple repo to play around with the DiffView extension.

I think I understand now that

BASE   = common ancestor
OURS   = your current branch (HEAD)
THEIRS = incoming change

Where theirs e.g. is when we pop a stash (i.e. what the stash "wants")

BASE is a bit misleading... it's in most cases (I can see) the starting point of a stash (i.e. what it was before the stash was done). But that is NOT the same as the prior commit to OURS! I guess its more the last time that OURS and THEIRS was the same? (makes more sense below)

This is the example

git init
echo "line 1" > file.txt
echo "line 2" >> file.txt
git add file.txt
git commit -m "base"

# create stash
echo "line 2 - stash change" > file.txt
echo "line 3 from stash" >> file.txt
git stash push -m "stash version"

# create branch change
echo "line 2 - branch change" > file.txt
echo "line 3 from branch" >> file.txt
git commit -am "branch change"

# apply stash (conflict)
git stash apply

(I have added a prefix for the git stuff so it doesnt think theres a conflict here)

XXXX <<<<<<< Updated upstream        ← OURS (branch / HEAD)
line 2 - branch change
line 3 from branch
XXXX ||||||| Stash base              ← BASE (common ancestor)
line 1
line 2
XXXX =======
line 2 - stash change           ← THEIRS (stash)
line 3 from stash
XXXX >>>>>>> Stashed changes

And as ChatGPT puts it...

BASE = where both started
OURS = what I have
THEIRS = what’s coming in
CENTER = what I commit

#Wincmd?

So apparently :tabdo executes for each tab?

Nice! so like

:tabdo :echo bufname()

Goes and prints bufname for each tab (active buffer)

Ahhhh :wincmd = is the same as C-W = a window command...

Interestingly, the reason the the outline escapes from this when its called (and graciously so, wouldn't want the outline taking up half the page horizontally...)

It is the winfixwidth property that forces the win to stay constant (when C-w = is called)