I work at etlelite.com
Wednesday, July 20, 2022
Tuesday, July 23, 2019
Install LEMP, Nginx, php-fpm, mysql, and mysql drivers on Digital Ocean Ubuntu 18.04
2 sudo apt update
3 # update apt
4 sudo apt install nginx
5 # Install nginx
6 sudo ufw app list
7 # Show apps available in Uncomplicated firewall
8 sudo ufw allow 'Nginx HTTP'
9 # Allow NGINX
10 sudo ufw status
11 # Show if UFW is working
12 systemctl status nginx
13 # Show if UFW is turned on status/systemctl
14 curl yourip
15 #verify nginx is running, and you are getting html
16 sudo apt install mysql-server
17 # Install Mysql Server
18 sudo mysql_secure_installation
19 # Set up a secure mysql install
20 sudo mysql
21 # Setup root user appropriately in mysql, see /root/.mysql_history
22 sudo add-apt-repository universe
23 # set up repo for php-fpm
24 sudo apt install php-fpm php-mysql
25 #Install Both PHP-FPM and the PHP-FPM drivers
26 vim /etc/nginx/sites-available/example.com
27 # Setup your server block
28 sudo ln -s /etc/nginx/sites-available/example.com /etc/nginx/sites-enabled/
29 # enable your server block
30 sudo unlink /etc/nginx/sites-enabled/default
31 #disable the default
32 sudo nginx -t
33 # test new nginx config
34 sudo systemctl reload nginx
35 #restart and reload nginx
36 vim /var/www/html/info.php
37 #create phpinfo page
38 curl http://yourip/info.php
39 #verify page is working, make sure the mysql and mysql are on the page
Wednesday, November 22, 2017
Monday, October 9, 2017
Some combination of these Will get omreport installed on a dell PowerEdge server with CentOS7
Omreport needs to be installed on your dell poweredge server, so you can check various RAID battery lives and vdisks and what nots.
wget -q -O - http://linux.dell.com/repo/hardware/latest/bootstrap.cgi | bash yum install dell-system-update yum install srvadmin-all yum install omreport yum install libcmpiCppImpl0 openwsman-server openwsman-client sblim-sfcb pciutils cat /etc/redhat-release wget -c http://downloads.dell.com/FOLDER02645266M/1/OM-SrvAdmin-Dell-Web-LX-7.4.1-1341.RHEL7.x86_64_A00.tar.gz mv /etc/redhat-release /etc/redhat-release.bak; echo "Red Hat Enterprise Linux Server release 7.0 (Maipo)" >> /etc/redhat-release tar zxvf OM-SrvAdmin-Dell-Web-LX-7.4.1-1341.RHEL7.x86_64_A00.tar.gz ./setup.sh rm /etc/redhat-release; mv /etc/redhat-release.bak /etc/redhat-release wget -q -O - http://linux.dell.com/repo/hardware/OMSA_7.4.0/bootstrap.cgi | bash ./setup.sh yum install srvadmin-all
Wednesday, October 4, 2017
Install Mac App Store Applications from the command line.
I recently, and hastily made some scripts for recovery of a mac, as I did a clean install of High Sierra (Which I don't recommend for right now, I have some problems with it, that I feel will easily be remedied as time passes. Mainly some issues with messages).
But I found a handy location, called mac that let's you install Mac App Store Lications from the command Line.
If we learned anything today, it's that Lication is the best way to shorten the word Application.
#Have the full Xcode installed for this to work
#Install mas
brew install mas
#Declare your lications, with line breaks in between
LICATIONS="Pages\nKeynote\nNumbers\nGarageBand\nKindle"
#for loop, looping through lications
for i in $(echo -e $LICATIONS)
do
#Install the lication, by finding and printing it's corresponding number, this is how mas works
mas install $(mas search $i | ggrep -P "^\d+ $i$" | awk '{print $1}') &
# Note that I eneded this line, with an ampersand switching it to the background, as to install in Parallel
#If you feel like it's doing to many at one time, remove the ampersand
done
Thursday, September 21, 2017
Thursday, June 8, 2017
Things I learned about R, Stats, DataScience or Machine Learning Today, 1496931039
R feels like an easy language to just run with. For a while I felt I groked it just about as well as I needed. But any time I get that confident, there is for sure something coming to prove me wrong. And this was no exception. Most days I can knock out code in R no problem, with very little interruption. Mostly I just import data via a csv or some other format and just start rocking. Well, today I had to do something basic I had never done before, just add a row to a data frame. Should be simple enough, right? Well it wasn't, it was very strange. I just wanted to add a vector with 31 one Integers, to a data frame. There has to be a simpler way to do this, but I turned the vector into a data frame and used rbind to add it. That wasn't the tricky part that was simple enough, but it kept adding NAs rather than my vector. It turns out I had the wrong class for my vector, I needed factors. The finished code for adding the row is:
for(i in factorCols) { levLeng <- length(levels(nashHouse[,i])); fLevels[i] <- levLeng; }
fLevels.df <- data.frame(t(as.character(fLevels)), stringsAsFactors = T)
colnames(fLevels.df) <- resCols
rbind(results, fLevels.df)
I used a for loop in my code to build my vector, which I believe is generally frowned upon in R, I will work to create a more functional version moving forward. Also I made this:
pie(c(10,10,1,2), labels = c("Naps", "Cartoons","Research","Coding"), main = "How I solve Machine Learning, and Data Science Problems", col = c("darkslategray1","darkslategray2","darkslategray3","darkslategray4"))
Clearly I'm not a person to trust, on any of the subjects in the post title.
Sunday, June 4, 2017
Star Trek Stats
I grabbed some Star Trek data from, http://www.thestartrekproject.net/ , and started making some graphs in R. Here are deaths, in shirt color:
Here are kills by Species:
It does not surprise me at all that humans are the most violent. As a white male, I know my race is the most prolific in violence.
Here's an interesting one. It shows Total Death's per victim...You'll notice Worf at 4 and Picard at 3...what do you want from me, it's Star Trek.
Here is the R, to make the above chart:
library(readxl)
library(ggplot2)
stardeath <- read_xls(path = '~/Downloads/all-deaths.xls')
char <- data.frame()
species <- unique(stardeath$Victim)
for(i in species){
new <-data.frame(amount = length(which(stardeath$Victim == i)), 'species' = i)
char <- rbind(char,new)
}
#We only want species with greater than 2 deaths
morethanone <- char[which(char$amount > 2),]
ggplot(morethanone, aes(species, amount)) +geom_bar(stat = 'identity', aes(fill = species))+ theme(axis.text.x = element_text(angle = 60, hjust = 1)) + labs(y = "Total Deaths",x = "Victim")
Also you should tottally get the DS9 DVDs
, cause DS9 was the best.
Nutaku, Sad Panda Studios, Crush Crush, and Theobromine Poisoning of bears.
Recently I started playing a nutaku.net / Sad Panda game, Crush Crush. Because a) I have an interest in creating a game like it, and 2) I'm a pervert. It's a dating sim, and an idle game. It does have the ability to poke fun of itself, for example one of the potential dates is a Bear. At on point the bear requests 50 boxes of chocolates.
After some research I realized that Bears have a lower threshold for chocolate consumption that humans, much like dogs. I wrote the following, to support:
I have been playing one of your games, "Crush Crush". Within it there is a character who appears to be a bear. One of the objectives in the game is to give "Bearverly" 50 boxes of chocolates. While this might be acceptable in a human who has a higher Theobromine resistance, bears like dogs have a much lower tolerance. If each box of chocolates is approximately 14 oz, and half of that box is actually chocolate, that would be approximately 21.875 pounds of chocolate.
"In 2014, four American black bears were found dead at a bait site in New Hampshire. A necropsy and toxicology report performed at the University of New Hampshire in 2015 confirmed they died of heart failure caused by theobromine after they consumed 41 kilograms (90 lb) of chocolate and doughnuts placed at the site as bait."
If you take the amount of chocolate and doughnuts these bears consumed, per bear, you would see the number is very close:
22.5 lbs / 1 Bear
There is an LD50(Lethal Dosage 50 percent, a term for when 50 percent of lab animals die, when given that amount of a hazardous material) for human consumption of Theobromine which is 1000mg/1kg, however there is no LD50 for Bears that I can I find. In 28 grams of chocolate(1oz), there is approximately 60mg of Theobromine. A large bear can weigh somewhere in between 272 and 408 kilograms. Meaning a lethal dose if the for a human threshold, somewhere in between 272 grams and 408 grams. And while 50 boxes of chocolate may only hold around 42 grams, with the bears lower threshold you can see the issue. And if any of them were dark chocolate, it nearly quadruples the amount of theobromine.
Would a bear eat 50 boxes of chocolate in a setting. It would, see above mention, bear traps regularly use chocolate. Would this much chocolate kill the bear...I think it most likely would.
Please, don't kill the bear.
Regards,
Dusty C.
In return I received back:
Hello there,
Thanks for pointing out to us that very important fact. We definitely did not mean to suggest or condone any purposeful poisoning of an animal, that would be absolutely cruel. Please keep in mind though that this is a work of fiction and imagination. It'll definitely take a bit of a turn from real life situations, and maybe require a bit of suspension of disbelief in parts.
After all, the typical bear also would not date a person but rather most likely maul them to shreds(unless the bear was hand raised by humans from a cub). Like poisoning, we are not suggestion or condoning anyone go up to bears in real life with amorous intentions neither. That wouldn't only be cruel, it would also just be silly.
I'll definitely let our game designers know about you pointing this out though. I'm sure they'll definitely keep it in mind for the future.
Cheers,
~Sad Panda
Tuesday, March 14, 2017
Saturday, December 24, 2016
The command line, an 18 year obsession, Or the one where I doubt myself
Since my first introduction to linux in 1999, I have been obsessed with the command line. With shells, commands and the glue that binds them. As a secondary programming languages, that extend that functionality. It's taken me to some weird places, some side journeys that I had to go on just to get further, like text manipulation, regex, and the movement of data around a file structure, and filetype standards. I work in JSON, and CSVs now primarily. I've gotten to a point where now I'm much closer than doing data science than text manipulation, it might have been a round about way to get there but it still feels like despite the windy path, it was the correct one to take. Particularly when I work in R, which seems to be special fitted for data wrangling but somehow I still manage to get more done in less time with some simple bash one liners than I can from within an R workspace.
With R, in the form of RStudio, I find myself working with GUI tools regularly for the first time in 17 years. I don't know if I like it. I don't know if this is the right course. Is this where I need to be aiming. Stats, Forecasting, and more importantly Machine Learning seems like the future. If not that then certainly the field of AI which is but a hop skip an a jump away. But am I, a lowly admin, who just happens to be good at the command line, at text manipulation, at regex, am I the one to pick up this mantle. It seems far fetched to think of myself as being able to do anything revolutionary or even productive in these fields when there are experts who have trained exclusively in them, and didn't end up there from some twisty side path. Some linux open source inspired secret garden that opened up to them.
The only thing I know for sure, is you should buy something from amazon so I can make some damn money. Whether it's the hands down best security book I ever read, or just pay for more prime:
Made a tool for encrypting keybase messages, with a little better workflow
My friend at necrux.com and myself made this tool to improve keybase workflow.
This week in self promotion 1482573936
Currently reading: Automate the Boring Stuff with Python: Practical Programming for Total Beginners
I wanted to pimp out my hiphopShell again now that it's more stable. I hope to have a song out from the new album before the end of the year. I'll post it here, and probably soundcloud as well. I started working with SuperCollider again today, to write some of the more electronic beats in the album. I might also take a gander at CSound again, not that I have particularly fond memories of it.
I have to get my hands on a new midi controller as well, just to complete the album, and I've been listening to a lot of Mac Lethal to inspire me.
Anyway enough Amazon linking except this final word from our sponser, BUY A DAMN AMAZON DOT:
I'm still hurting for cash, can you tell?
Thursday, December 22, 2016
MakeAmericaMesh.org The trip and money.
I'm still trying to come up with the money to complete my Make America Mesh trip. I'm going to shill a bit for amazon associates the next few days in blog posts, I'm going to keep them informative of course, but they will be heavily amazon linked...just like this one.
Again I am trying to do something good, and if there is something I could write about that you would want to pay for, let me know.
Your money...
I needz it...
http://makeamericamesh.org/
Paypal me bsdpunk@gmail.com
You can email me too...I'll answer that shit. (No money necessary)
Everyday Carry PenTest Edition
This is an everyday carry but not in my pockets, in my man-purse, or briefcase as some have taken to calling it. The "bag" in question being a Kenneth Cole Reaction briefcase.
So the tools most notably in here are USB rubber ducky( for quickly injecting malware in hypothetical within reach machines), Yardstick One( for hypothetically jamming wireless signals within it's rating or hypothetically intercepting rolling encryption car fob door opens, and starts), usb SDR( for playing with Software Defined Radio, and not cell signals), Ubertooth(For bluetooth operations), a host of usb fobs, and that concludes one side of the bag.[Most of these can be purchased at Hak5] On the other side of the bag* I have various antennae for the aforementioned devices, an RFID card reader, some blank cards and keychains, and a set of door shims of varying thickness.
Within the briefcase I also have a black tie, because why not. And a mechanical keyboard. The keyboard may seem like overkill but your fingers will thank you, and with the weight loss from upgrading to the apple from the Dell, I don't notice the extra weight.
*Forgot the Turtle Lan Shell
For less potato, but more linux / hacking pics, check out my instagram.
Friday, December 16, 2016
Making Dialog Appear in Unity, Game Dev
function thingCast() {
Debug.Log("Confrontation");
canvasObject.SetActive(true);
}
Script for Sender:
if(Input.GetButtonDown("Fire2"))
{
var hitt : RaycastHit;
Debug.Log("Fire2");
if(Physics.Raycast (transform.position, transform.TransformDirection(Vector3.forward), hitt))
{
Distance = hitt.distance;
if(Distance > MaxDistance){
hitt.transform.SendMessage("thingCast", SendMessageOptions.DontRequireReceiver);
}
}
}
Assign the actual Canvas object to canvasObject in the inspector.
Everything I learned about Linux, Mac OS X, and Hip Hop... In Whatever time period...1481372427
https://github.com/bsdpunk/hipHopTools
And at that point I was like why not try to roll those into a shell:
https://github.com/bsdpunk/hipHopShell
So that shell doesn't work that great right now, but maybe it will in the future. Or maybe I'm done with it I don't know.
I took a challenge to only use Safari for the past week. I only cheated twice. Once was because of a wireless issue I was trouble shooting and I wanted to eliminate the browser as an issue. The other was because I had some chrome saved password I wanted to use. So what did I think of Safari. Well...
There's no reason not to use it other than nitpick things. I don't like that it doesn't play webm's natively. It certainly takes less RAM than Chrome(my normal choice). It also seems to lend itself to managing your tabs better, as in I never had more than 9 or so open, as where with Chrome it can be in the 100s. I don't like the actual physical spacing of the tabs, I want the close tab button to stay in the same place, and I don't like the way the tabs can shift, those seem like terrible design errors on Apple's part. That being said the browser performed well, and I never had any issues in compatibility. Weirdly enough the better RAM performance has me eying firefox more. Like that makes sense or something.
In my efforts to keep track of my time usage, I wrote a bash function:
function snap () { ps aux | awk '{print $3,$4,$1,$2,$11}'| sort -rnk1,1 | grep -v '0.0 0.0' && echo bk && vm_stat && echo bk && osascript -e 'path to frontmost application as text' ; }
Just get's some information about the processes, memory usage and the current in focus window.
ps aux | awk '{print $3,$4,$1,$2,$11}'| sort -rnk1,1 | grep -v '0.0 0.0'
Get all processes. awk CPU%, MEM%, user, PID, and Location of Program, sort it, omit things not taking resources.
echo bk
Separate with a newline, two characters not naturally occurring in english, then a newline.
vm_stat
Get Memory info
echo bk
Separate with a newline, two characters not naturally occurring in english, then a newline.
osascript -e 'path to frontmost application as text'
Get the in focus window
You can get most of my one liners:
https://github.com/bsdpunk/one-liners
I wrote this function to grab videos off of websites, doing a recursive web crawl:
function wet () { wget -nc -c -r -A'*.flv' -A'*mp4' -A'*mkv' -A'*wmv' -A'*webm' -A'*.mpg' -A'*.mov' -A'.gif' -A'.jpg' -A'.wmv' $@ ;}
I think I had something else to write about but now I don't remember. So maybe another update soon.
Saturday, December 10, 2016
Ourobouros, I almost made a thing
Totally forgot to announce it here but I'm almost finished with a thing...Interestingly enough I just did a full reinstall on my drive so I lost the latest build...but not to worry, I'll just post a thing about how to grep your empty space on your hard drive next...to show how to recover lost files. DOH...WHY DIDN'T I VERSION CONTROL...THOUGHT I DID. DOH.
Friday, December 9, 2016
Reasons I Hate A Certain Hosting Company
I don't hate them for my uncermonious firing. I don't hate them for their customer service, which is actually quite good. I don't hate them because they aren't up to the gold standard of hosting, that is RackSpace (Another former employer of mine) which I hold to be a gold standard of what hosting should be. I have criticisms of their tech, and security. But I don't hate them for that.
I hate them because I met a long time 3 character name. Because of a 16 bytes in unicode-32. I hate them because of 128 bits of utf-32. A long time malvagerite(the irc network I founded, and love), who would have remained a name on the screen if I hadn't have taken my short meaningless commission at that company. The first time we met I was so worried about my stay in that big city. Were things going to turn out well(they wouldn't) would I live comfortably and be ok(I would, but only because of those 128 bits).
I was this weird neurotic outcast, coming into a clove of already clicked people at my work. I didn't really fit in, and I was an asshole about it. I snubbed my nose at them for their brown cherry mx switches("You don't love yourself"), and their acursed use of the nano editor. So of course everybody loved me...no.
But that first paycheck I came to the actual city and I met that, 128 characters. Sen, as he was known in our community would come to be my greatest ally in a world that seemed out to get me. He was loud. He was big. He would shout obscenities in Polish. He had a knack for saying the most vulgar things. He was my friend. When I walked with Sen, no matter which part of the bombed out ghetto, or nicities the city had to offer, I walked with a 10 foot tall bullet proof, body gaurd.
He could be quick to anger, he was scary, and he could blame you for misfortunes on a dime. Sure those weren't the most charming qualities...But he made up for all of that in both genorosity and loyalty. When I walked with Sen I had no one to fear. If I needed food, BAM fucking scrapple on the table, even if I wanted beer, I didn't even have to ask he would provide. If he saw me scrounging for quarters suddenly a dollar would appear.
In less than a month of being there, I lived in the same building as him. This was before his greater qualities shown through. We both took a risk. I wish now that I would have stayed. But I didn't. Sen is dead now.
Typing those words puts me in a state. I recently did a stint in the hospital for over a week, as in I got out today. I had time to think about things in a neutral enviroment. I didn't know how much his death affected me until reflecting over it, in pale clinical lights. With lots of alone time. With my recent projects I have been running non stop, without reflection. I miss Sen, and I feel an overwhelming misplaced hatred for a company who has essentially done no ill will. They let me go because my performance was poor, as any company would do. But at least I can hate them. At least I can pretend it's someone elses fault and not the cruelty of the world that took Sen. I miss you Sen. Malvager misses you Sen.
Tuesday, November 22, 2016
What's real sadness.
It's smoking butts from an ashtray, while cuddling under a blanket your dead grandma made you on the balcony, while you cry because your friend is dead. If you have 15 bucks, send it to https://www.facebook.com/malwina.bartulska.5?fref=ts [Malwina Bartulska] on facebook. If you have 20, send 15 to her, and 5 to me for makeamericamesh.org ( paypal.me/bsdpunk ). Or just send it all to her, that would for sure not hurt my feelings.Either way she could use some healing, and I need some money for my meshnet charity mission. But I'm not going to lie...the next 5 dollars I get goes towards a pack of fucking newports.
Sunday, November 20, 2016
Saturday, November 19, 2016
Meshnets for life, homelab
Special thanks to Shayne
http://makeamericamesh.org/
Also Look at this mother fucker, and donate it to me:
GL-MT300A-Ext, smart mini router, 128MB RAM, MicroSD card, 300Mbps WiFi, OpenWrt pre-installed, Repeater, Tethering, OpenVPN
Thursday, November 3, 2016
LET THE GRAPHING BEGIN
I started working on a top secret project, involving Machine Learning, and predictive analysis.But because of the territory I have been doing a lot of graphing. Just a by product of learning R I guess. tinyLove, is doing well. I would like to state if I was starting over, I would start my R installation from this page:
https://r.research.att.com/
Rather than just bulldozing ahead with brew. I've had a little bit of an issue with one of the more obscure package installs. And I feel like i wouldn't have had to go to the extreme effort, to get that package installed if I would have known about the above site.
I do love the mac, I can't live without a tiling desktop however. So I have KWM and KHD installed:
https://github.com/koekeishiya/kwm
https://github.com/koekeishiya/khd
KWM and KHD are mighty powerful. Also a tad more complicated that I would prefer. I miss the simple mod4+# to switch to a space. And I think if I had the will I could edit the config file to be that simple, but I wanted to get a feel for what I thought the designer was going for. And I really appreciate the depth in thought they went for.
Also this evening I was fucking around with Julia, which is a language in it's infancy. I wanted to do something simple but practicle with it, so I made a weather getter with it. One thing I noticed was my curl command picked up a different set of information than the Julia Requests library. I took a gamble that it was the User-Agent making all the difference, and that resetting it in the infant Julia would actually work. That payed off.
using Requests
import Requests: get
cat = get("http://wttr.in/nashville"; headers = Dict("User-Agent" => "curl/7.37.0"))
thedata = transcode(String,cat.data)
print(thedata)
Thursday, September 22, 2016
Everything I have learned about Linux...errr Mac, since Wednesday 1474607836
So I saw this post today:
https://www.reddit.com/r/linux/comments/53vcr9/as_an_experiment_i_ran_linux_for_365_days_without/
Long story short...I used linux for a year, here are my thoughts. As someone who uses linux as a Desktop daily, I never understand what the big deal is. Well, I bit the bullet and bot a Mac on Wednesday. Not because I don't love my trusted linux box, but because the laptop itself was a loaner, and being that I cashed out my 401k, to you know...stay alive, I figured I could use some of it to buy a lappie for myself. I got the cheapest offering, a year old air. so this is:
Everything I learned about using a mac since Wednesday
One thing I have noticed, is that I have sacrificed A WHOLE LOT OF WEIGHT. I don't know what the XPS I was using weighed, but I went from one extreme to the other, and with seemingly no performance cost. To be fair the XPS was several years old and I don't think I ever got a good graphics driver for it. But man this thing flies.
The only thing I am actually having trouble with is Desktop snapping and Window management, in the past I have used Ametheyst, but I have not yet installed it and remember that it wasn't the full flavored solotion I really wanted, so I shall try some new ones, and get back to you.
Also, I have been using brew for package management, and fuck yeah vim is at 8 in brew. Brew...you rock. So I'm going to continue to do these mac updates, with the things I learn.
Monday, August 29, 2016
Straight Crushing it
I want to talk about two companies, that are just straight killing it. Crushing it.
Nutanix is like if a country, that has won more gold medals than the rest of the world, had the largest military, sent a man to the moon, is the youngest kid on the block, and is straight taking it face to face with the largest empire ever created, because they don't want to pay taxes on a war they forced that country to be in.
I don't particularly have a lot of pride in America. But it’s hard to argue with results, and America has been straight crushes it for 200 years. Sure it was built off the backs of subjugated people. And having the ability to keep factories open, after bombing the shit out of the rest of the world. Then moving into the modern era by having more mathematicians on staff to just utterly dominate the information of the world and it's own people.
And Nutanix might be like that. I don't know, I don't really know much about the Nutanix culture or day-to-day affairs. I do know they offer great products, at prices more affordable, than VMware.
The Nutanix VMware rivalry is well documented.
And I wouldn't be raving about Nutanix if there wasn't a reason. The capabilities of Nutanix’s hardware blows everything out of the water. They offer a layer of hardware for SSD cache to alleviate the bottleneck that is IO.
Nutanix is eating VMware's lunch. VMware tried to silently creep in and be your only hypervisor, like some clandestine takeover of the data center. But Nutanix came to the door and said, "I'm better, and sometimes I'm even cheaper."
And why wouldn't you let them in. Nutanix is winning for the foreseeable future.
Nutanix is more affordable. By all accounts Nutanix seems to be crazy good and has crazy resilient hardware.
Now to switch gears, paid hosting, the cloud...
The Cloud Saves money, we know this. Some of us don't understand how but it does, so I'll give you a brief intro. Once upon a time I worked at a small search engine. And we needed anywhere between 7 and 30 servers a day, depending on fluctuations. Clearly keeping that hardware on hand was not justifiable as it would typically be sitting dormant in a corner. Renting made fiscal sense and had the added benefit of moving the expense from Capex to Opex. Plus when you're in the cloud, for whatever reason, you just really find ways to squeeze the juice out of those servers. That's it. Economics lesson over.
Let's talk about Amazon. They're offering crazy innovations for goddamn pennies. Seriously look at the services Amazon is offering. Are you even trying other hosting companies? I checked...they aren't.
If brash Nutanix is America, then Amazon is Finland. Amazon's mature for a tech company, maybe not for a normal corporation, but practically old age for a tech company. Finland has the best education in the world, they score high on the happiness index, life is good. Okay, okay we heard some not so great things about their employees not having the best time not too long ago, fair enough. Finland's also practically Eastern Europe. I don't really know if there's actually anything wrong with Finland. I hear they drink a lot, but really that's the kind of quality that endears me to a nation. So whatever, just get past that. I'm pretty sure Amazon's engineers are happy. I mean you can't be straight disrupting the entire paradigm you’re working in, and not feel good. Look at the products Amazon put on the table....no really, fucking look...NO seriously, take a goddamn look. I'll wait:
https://aws.amazon.com/
But I want to talk about two services particularly and, they’re all fucking gems. I'm just choosing these two because they're on opposite ends of the spectrum. But they both do one crazy thing...they save you a fuckton of money.
Lamdba
Okay I'm going to offer you a service, where you just upload code, and you only pay for the compute you use. What do you think?
Are you kidding me with this shit, not only am I only paying for the compute I use, with the right Dev, I don't even need an infrastructure team. I just want to note as someone whose bread and butter is infrastructure, this was very rude of Amazon.
Machine Learning
Okay, so we’re just gonna go ahead and implement some advanced math algorithms for you...So your devs don't have to like know math and shit. Oh what do they do. THEY PREDICT THE GODDAMN FUTURE. But they're hard to implement, or they're something tricky right? No, just upload a CSV with accurate data about the past, and the better your data, the better the predictions. Alright, so what kind of price tag are you talking about on the magic robot that predicts the future? I mean that's got to be expensive right! Well I've used the magic robot, with ginormous CSV files. And it ran me about $4 USD (which I am totally good for Amazon, just having a little streak of unemployment, so that card is maxed out, but like un-suspend my account so I can get that sweet, sweet Alexis skill shirt, plz).
So who else in hosting/cloud has it together. Well let's look at Rackspace, they've got two decisions under their belt, that make them still worthwhile. In all fairness I used to work there and still have a certain fondness for them. Firstly they partnered with Amazon to offer a managed cloud product. And second, they eliminated services not core to their business (Cloud Sites), and they're buying themselves out of stock holder hands’...Presumably so they won't be beholden to stock holders and can be engaging in the kind of shit that Amazon's already doing.
And I have no more nice things to say about hosting so, if you work for some other hosting company, continue read on at your own peril. Companies like Linode, Digital Ocean, and Vultr can be great...for devs only. Or for doing a single box set up. Or even a few box set up. But some of these places you can't even virtualize on the boxes, so don't even think about ESXi, Xen, KVM, Docker, LXC, or any microservices. They all offer around the same comparable features, with maybe a $5 option extra...You know because who wants to spend $10 USD a month on the thing that runs your business…
Oh hey and did I mention that Amazon is straight up cheaper than these guys.
So to hosting companies everywhere I say, Get your shit together. Get all your shit, put it in a backpack, sell it at the shit store if you have to........
GET IT TOGETHER HOSTGATOR, FUCK.
It's like reading a goddamned Ayn Rand novel; everyone else is so comically far behind these guys, it seems like there lack of progress is a sure sign of malicious intent.
Amazon and Nutanix are primarily doing one thing that others aren't. They're innovating. They're taking risks in their fields and the features they offer are unique, and strangely enough cheaper than the alternatives. So everybody else, get it together already.
P.S. Nutanix, if you want to send me something for my home lab, that would be wicked awesome(Yeah, fucking right). No but seriously something to wrap around my torso would be great.
Join me next time, when I tell you what I really think of Microsoft.
Sunday, August 28, 2016
Shit I learned about linux, in like, this morning and last night when I didn't sleep, and maybe other times, 147235608
I made my terrible bash script that counts language stuff from github completely automated. I also wrote like a real post, that I'm still proofing. I'm going to probably actually try to promote it, so there's that. This is a lot easier to work with now, though it still lacks any robustness, so one mistep and it's not going to work. Plus, right now it creates and deletes files, which is a horrible way to solve my goal but it was a quick way. So I assume if I have time I'll fix that in the future. I did go ahead and copy and paste a few lines of R so I can get a graph from it as well. https://github.com/bsdpunk/github-lang
Enter your github token, if you have one. Otherwise hit enter: Enter which github user you would like to check percentages on: bsdpunk Total LoC: 367611 AppleScript 696 0.189331% Awk 539 0.146622% Go 6761 1.83917% JavaScript 8323 2.26408% Perl 23081 6.27865% PHP 20528 5.58416% Python 302957 82.4124% R 1926 0.523923% Shell 2800 0.761675%
One of the things I learned today was that you can delete a line and as many lines after it with sed:
sed '/true/,+1d'So match true, delete the line it's on, and the one after it. The Rscript that makes the graph in my github piece is ridiculous simple. R is really just a pleasure to use. I miss doing the draft kings stuff with it.
locs <- read.csv('loc.csv', header=TRUE, sep=",")
png("loc.png")
plot(locs)
title(main = commandArgs(TRUE)[1])
dev.off()
The bash script that does most the heavy lifting, is just a mess. I mean I've done some crappy one liners before, but this just feels bad. I guess I could clean it up.
Here's a little php, to send an email:
php -r '$msg="tasty message";mail("someone@example.com","My subject",$msg);'
And how to force curl, to use a domain, that isn't assigned to an IP(in case you don't want to fuck with your host file):
curl --header "Host:domain.com" 127.0.0.1
Saturday, August 27, 2016
Everything I learned about linux in the last whosit, 1472336806
I use awk a lot. Like every day. I don't know much about it, how to use it, etc. I just know how to set the Field Seperator, which in my head I always called delimiter. Which is with the -F was confusing to me. So I guess that's what I used to know, that and fields. But I had to bite the bullet and write my first, real awk script. It's pretty bad you guys:
BEGIN {
count=0;
FS = ":";
total[count] = "";
lang[count] = "";
final = 0;
}
{
clang=$1;
amount=$2;
#print lang[count];
if (clang == lang[count] || lang[count]=="") {
lang[count] = clang;
total[count] += amount;
final +=amount;
#print final, lang[count];
}
else {
count++;
lang[count] = clang;
total[count] = amount;
final+=amount;
}
#print clang, total[count];
}
END {
print "Total LoC: " final;
for(x in lang){
percent = (total[x]/final)*100;
print lang[x], total[x], percent "%";
}
#print "Total Amount = $";
}
' file
So all of this is to find your total lines of code and what kind of code, on github by cycling through a cleaned up file of your languages. Now I have a lot of one liners I put in the readme.md, to clean up the stuff I grabbed from github, but let's be honest this project is a mess. I didn't really plan to get this deep into github bullshit today, but I did. The lack of planning shows. But if you want to check the mess out it's all posted on https://github.com/bsdpunk/github-lang
So how did my totals come out...
Well if you remove the forks and you take out the HTML and CSS:
Total LoC: 394684 AppleScript 696 0.176344% Go 6761 1.71302% JavaScript 8320 2.10802% Perl 23081 5.84797% PHP 20528 5.20112% Python 332131 84.1511% R 1805 0.457328% Shell 1362 0.345086%Not bad for all my public/open stuff I guess. Maybe it is, I don't know. I'm going to try to post something about PowerShell, and Linux, and Microsoft, and Opensource, soon, maybe. As like an opinion /slash/ neat tricks with PowerShell.
Subscribe to:
Posts (Atom)