Bonum Certa Men Certa

Using a Single-Board Computer to Monitor IPFS

IPFS lights-based monitoring on self-hosted SBC
IPFS lights-based monitoring on self-hosted SBC (blue is for status, green and red for upstream and downstream payloads)



Summary: IPFS is light and simple enough to run from one's home, even on a low-voltage machine, and the code below can be used as a baseline for monitoring IPFS activity 24/7




#!/usr/bin/python3
# 2019-04-22
# 2020-11-07



from blinkt import set_pixel, show from random import randint,random,shuffle,randrange from time import sleep import argparse import signal

def solid(r,g,b,s): while True: for pixel in range(8): set_pixel(pixel, r, g, b) show() sleep(0.1)

def random_lights3(): while True: for pixel in range(8): r = randint(0, 255) g = randint(0, 255) b = randint(0, 255) set_pixel(pixel, r, g, b) show() sleep(0.1)

def random_lights2(): while True: p=range(8) p=sorted(p, key=lambda x: random()) for pixel in p: r = randrange(0, 255, 16) g = randrange(0, 255, 16) b = randrange(0, 255, 16) set_pixel(pixel, r, g, b) show() sleep(0.1)

def random_lights1(): while True: p=range(8) p=sorted(p, key=lambda x: random()) for pixel in p: r = randrange(0, 255, 8) g = randrange(0, 255, 8) b = randrange(0, 255, 8) set_pixel(pixel, r, g, b) show() sleep(0.1)

def spacer(r,g,b,seconds): while True: for pixel in range(8): set_pixel(pixel, r, g, b) next = (pixel+1)%8 set_pixel(next, 0, 0, 0) show() sleep(seconds)

def reversed_spacer(r,g,b,seconds): while True: for pixel in reversed(range(8)): set_pixel(pixel, r, g, b) prev = (pixel-1)%8 set_pixel(prev, 0, 0, 0) show() sleep(seconds)

def cylon(r,g,b,seconds): while True: for pixel in reversed(range(8)): set_pixel(pixel, r, g, b) prev = (pixel-1)%8 if prev < pixel: set_pixel(prev, 0, 0, 0) show() sleep(seconds) for pixel in range(8): set_pixel(pixel, r, g, b) next = (pixel+1)%8 if next > pixel: set_pixel(next, 0, 0, 0) show() sleep(seconds)

def pulsed_bar(r,g,b,seconds): steps=8 while True: for fade in reversed(range(steps)): r2=r*(fade+1)/steps g2=g*(fade+1)/steps b2=b*(fade+1)/steps # print (fade) for pixel in range(8): set_pixel(pixel, r2, g2, b2)

show() sleep(seconds) for fade in range(int(steps/1)): r2=r*(fade+1)/steps g2=g*(fade+1)/steps b2=b*(fade+1)/steps for pixel in range(8): set_pixel(pixel, r2, g2, b2)

show() sleep(seconds*0.5)

def ipfs(r,g,b,seconds): steps=4 # how many stages in gradient brightness=0.5 # how bright the lights will get bluebright=100 # the brightness of the blue light in the middle (0-255), albeit overriden by input dim=1 # increase to dim down the lights run = 0 # running count for periodic file access while True: # run always (until interruption) run=run+1 # first, open from files the required values, which change over time if (int(run) % 50 == 1): with open(r'~/RateIn', 'r') as f: # open from file the IN value # print(r) lines = f.read().splitlines() r=int(lines[-1]) # read the value # r=int(map(int, f.readline().split())) # prototype, for multiples (stale)

with open(r'~/RateOut', 'r') as f: # open from file OUT value # print(g) # show values, debugging lines = f.read().splitlines() g=int(lines[-1])

with open(r'~/Swarm', 'r') as f: # open from file Swarm value # print(g) # show values, debugging lines = f.read().splitlines() bluebright=int(lines[-1])/2 # print(bluebright)

for fade in reversed(range(steps)): # fade in effect # print(g2) # show values again, debugging # print(r2) r2=r*(fade+1)/steps/dim g2=g*(fade+1)/steps/dim b2=b*(fade+1)/steps/dim

# print(g2) # show values again, debugging # print(r2)

# print (fade) for pixel in range(3): # first 3 LED lights set_pixel(pixel, r2/20, (g2*brightness)+(pixel*1), b2/20)

for pixel in range(5,8): # the other/last 3 lights set_pixel(pixel, (r2*brightness)+(pixel*1), g2/20, b2/20) if (bluebright==0): set_pixel(3, 255, 255, 255) set_pixel(4, 255, 255, 255) else: set_pixel(3, 0, 0, 0) set_pixel(4, 0, 0, bluebright)

show() sleep(seconds/r*r+0.1) for fade in range(int(steps/1)): # fade out effect r2=r*(fade+1)/steps/dim g2=g*(fade+1)/steps/dim b2=b*(fade+1)/steps/dim

for pixel in range(3): set_pixel(pixel, r2/20, (g2*brightness)+(pixel*1), b2/20)

for pixel in range(5,8): set_pixel(pixel, (r2*brightness)+(pixel*1), g2/20, b2/20) set_pixel(3, 0, 0, bluebright) set_pixel(4, 0, 0, 0) show() sleep(seconds/g*g+0.1)

def flashed_bar(r,g,b,seconds): while True: for half in range(4): set_pixel(half,r,g,b) for half in range(4,8): set_pixel(half,0,0,0) show() sleep(seconds) for half in range(4,8): set_pixel(half,r,g,b) for half in range(4): set_pixel(half,0,0,0) show() sleep(seconds)

def handler(signum, frame): print("\nSignal handler called with signal", signum) exit(0)

signal.signal(signal.SIGTERM, handler) signal.signal(signal.SIGINT, handler)

# read run-time options

parser = argparse.ArgumentParser(description="Drive 'blinkt' 8-pixel display.") parser.add_argument("pattern", help="name of light pattern: \ random[1-3], spacer, reversed_spacer, cylon, pulsed_bar, flashed_bar") parser.add_argument("r", metavar="r", type=int, help="red channel, 0-255") parser.add_argument("g", metavar="g", type=int, help="green channel, 0-255") parser.add_argument("b", metavar="b", type=int, help="blue channel, 0-255") parser.add_argument("timing", metavar="s", type=float, \ help="rate of binking in seconds") options = parser.parse_args()

pattern = options.pattern.lower() r = options.r g = options.g b = options.b s = options.timing

if pattern == "solid": solid(r,b,g,s) elif pattern == "random3": random_lights3() elif pattern == "random2": random_lights2() elif pattern == "random1" or pattern == "random": random_lights1() elif pattern == "spacer": spacer(r,g,b,s) elif pattern == "reversed_spacer": reversed_spacer(r,g,b,s) elif pattern == "cylon": cylon(r,g,b,s) elif pattern == "pulsed_bar": pulsed_bar(r,g,b,s) elif pattern == "ipfs": ipfs(r,g,b,s) elif pattern == "flashed_bar": flashed_bar(r,g,b,s) else: print("Unknown pattern") exit(1)

exit(0)





Example runtime: run-blinkt-ipfs.py ipfs 0 0 0 0.00

Based on or derived from baseline blinkt scripts; requires the hardware and accompanying libraries being installed on the system.

For the code to run properly in the above form, given that it takes input from files, the IPFS values need to be periodically written to disk/card, e.g. for every minute of the day:

* * * * * ipfs stats bw | grep RateIn | cut -d ' ' -f 2 | cut -d '.' -f 1 >> ~/RateIn * * * * * ipfs stats bw | grep RateOut | cut -d ' ' -f 2 | cut -d '.' -f 1 >> ~/RateOut * * * * * ipfs swarm peers | wc -l >> ~/Swarm

These lists of numbers can, in turn, also produce status reports to be shown in IRC channels. When our git repository becomes public it'll be included (AGPLv3).

Recent Techrights' Posts

SUEPO Central Made a Strike (or Striking) Success
Europe has more than enough qualified patent officials
Wikipedia - Funded by Slop-pushing Companies and 'Broligarchs' - Gave Benefit of the Doubt to Slop, Then Regretted It
Wikipedia sucks. Without slop it'll suck a little less.
 
GNU/Linux Will Thrive as Long as It's Modular, Not Monolithic
To IBM, it's all about money. Nothing else matters.
EPO "Cocaine Communication Manager" - Part X - People Are Leaving
"I was happy to be at the EPO in the beginning, but since I realized it's all a big mafia"
IBM's 33 Years as a "Financial Engineering" (Accounting Tricks) Company
In relation to Red Hat, this "financial engineering" involves culling many workers and trying to replace them with slop
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Monday, March 30, 2026
IRC logs for Monday, March 30, 2026
Links 31/03/2026: Rising Costs, Cyberattacks, Novo Patent Expiry
Links for the day
Gemini Links 31/03/2026: American Spring, Distributed Systems Simulator, and Calculus for Electronics
Links for the day
IBM Layoffs and Their Expected Scope in April 2026
Such layoffs impact not only IBM "proper"
SLAPP Censorship - Part 28 Out of 200: Facing Consequences for Impersonation and Worse
It's not "funny". It is moreover libellous.
Links 30/03/2026: South Korea Next to Curb Social Control Media Addiction and Manipulation, Notorious Patents in the US Challenged
Links for the day
Gemini Links 30/03/2026: Going Back to Wrist Watches and Why LLMs in Programming Suck
Links for the day
Did IBM Pay thestreet.com for Puff Pieces? (Like It Did With Forbes)
If so, there is no disclosure
Payoffs of Lifelong Commitments
"The Lifelong Activist"
Links 30/03/2026: "We Can’t Income-Tax Ultra-Elites"; "The Pirate Bay’s Oldest Torrent Turned 22"
Links for the day
Today, Europe's Second-Largest Institution (EPO) Goes on Strike That Can Last Until 2027. Nobody in the Media Covers This!
"We stand with the protesters"
When the Cost (or Time) of Maintenance Exceeds the Value
In recent years it seems like more people learn to remove things from their lives, not add more things
Passage of Wealth Upwards, Blaming the Victims
Tim Sweeney's net worth is 5.1 billion USD according to Forbes
More Media Needs to Tell the Public Slop is a Giant Bubble, It Should Stop Taking "Sponsorship" Money to Inflate This Bubble
If enough of (what's left of) the media changes its tune and quits being a parrot of GAFAM, then we can debate slop like grown-ups
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Sunday, March 29, 2026
IRC logs for Sunday, March 29, 2026
Trying to Hide One's Abuses by Imposing Silence on Critics ("My Profile Was Private")
With enough daylight, sooner or later everyone knows you are a vampire
Fedora Badges System Shows the Demise of Fedora Under IBM
IBM isn't good at keeping what it buys
IBM is Sunsetting Red Hat, It Only Uses the Brand and the Shell
IBM buys or spins off companies as containers for "toxic assets" and debt
Cisco Systems is a Still Weak Spot With Bug Doors
nothing to offer except storytelling
EPO Strike Begins Today and It's the Longest One Yet (Can Last a Year)
Where's the media?
Gemini Links 30/03/2026: Approaching April and Arvelie Calendar
Links for the day
No Daylight Saved
Is there still any practical reason for this ritual?
Microsoft Azure Does Not Have "Hiring Freezes", It Has Had Mass Layoffs Every Year Since 2020
Things are always a lot worse than Microsoft formally or publicly acknowledges
SLAPP Censorship - Part 27 Out of 200: Using the Tor Network to Hide From Consequences
Only 1-2 weeks after the countersuit the Canadian attempted to deplatform several Web sites
The Limits of Inclusion
Inclusion with caution isn't "opinionated"; it's a defence mechanism, sometimes a survival instinct
Almost 20 Years After Microsoft/Novell
The mission has not changed, but the priorities evolve all the time
People Discuss Rumours of Mass Layoffs at IBM Becoming Public in 1-2 Weeks
IBM is killing its brand or its "goodwill"
LLM Slop Kills Sites, as Sites That Adopt Slop Are Doomed
People won't subscribe to such sites and visit them if they recognise it's just slop
Links 29/03/2026: Indonesia Cracks Down on Social Control Media Addiction, China Becomes World’s Scientific Superpower
Links for the day
Fedora at the Mercy of Microsoft Because of Back-Doored Kick-Switch Boot
We'll soon revisit the defamation attacks on Torvalds
Links 29/03/2026: Water Shortages and No Kings Rallies
Links for the day
The Old Days
In the early days of this site (2006) it was mostly just a couple of people, plus comments
Gemini Links 29/03/2026: Return to Gopherspace, "Zen of Marking Playing Cards"
Links for the day
The Real XBox is Dead, So Microsoft is Calling Everything "XBox" Now
It even wanted to run a campaign to convince everybody that XBox is not actually a console
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Saturday, March 28, 2026
IRC logs for Saturday, March 28, 2026