And now it's time for a story of how Pinterest scaled to 11 million users with only 6 engineers.
No BEAM this time, though Pinterest isn't a chat app either. They managed to scale that much and stay so lean thanks to the architectural decisions (mostly around keeping it simple).
P.S. Also, Pinterest is probably the best social network out there (if you could call it a social network).
#architecture
There were quite a few nice bundles at HumbleBundle lately!
This one is about cloud as well as ML pipelines design, which should be interesting.
Also, these are O’Reilly books and O’Reilly is famous for their high quality.
#books #bundle
Do you run databases in Kubernetes?
Even if you don't, I bet you may run database migrations there. How do you do that?
This article on "The New Stack" makes a case for GitOps approach to the database migrations in Kubernetes.
*tl;dr*: It's Atlas Operator, there's no alternative.
#kubernetes #databases
For some folks it’s easier to digest information in video format.
I am myself guilty of watching way too many YouTube videos.
Thus, today I want to share with you a bundle of Linux Video Courses from Packt & HumbleBundle.
#books #bundle
A late Sunday digest is out!
https://open.substack.com/pub/catops/p/catops-digest-2024-01-21
#digest #newsletter
We usually don't post event announcements. Yet, there's an exception: our direct involvement in the event :)
On 17th of February I'm speaking at DevOps fwdays'24 conference together with other wonderful people.
You can join the conference either offline in Kyiv or online from anywhere in the World as long as you have the Internet connection.
Also, I have a promo-code CCD4090099 that gives 15% off for any type of tickets!
UPD: if you would like to talk at the conference, you have the last chance to submit your talk via the CfP form!
#event
For today’s Donations Monday you can support our snipers to start your week on a good note.
#donations #Ukraine
Good news in Terraform world!tofu test
syntax is the same (or pretty similar) to terraform test
syntax in 1.6+
Also, it's first GA release of opentofu (about 5 months from forking)
Let's see how it will go. Currently, I can said that there bunch of folks who bombard everything with "please support tofu in N" (before there were any releases) but don't do anything to contribute to projects.
Also, in case you're excited about migration to tofu - firstly, you must fix all your TF code stacks, which is nearly impossible in huge organizations. Or make gradual migration => live with both tf and tofu side by side for months.
#terraform
Just recently Denys made a video about the interviewing process into a “hot” AI startup.
They use Python there (surprise, surprise). And just today I’ve noticed a Python book bundle on Humble Bundle.
Also, if you’d rather switch to the management career track all together, there’s another book bundle for you.
I know that some people complain about the quality of Packt books, but I’m a simple man: I see a relevant book bundle - I share it. Now, you can decide for yourself.
#python #management #books
The first CatOps digest of the year 2024 is available here:
https://newsletter.catops.dev/p/catops-digest-2024-01-07
#digest #newsletter
Start the new year with donations to the Ukrainian military!
Back and Alive raise 220M UAH for Ukrainian snipers.
You can join the fundraiser here
#donations #Ukraine
Root Cause Chronicles: Connection Collapse is a hypothetical (or not?) scenario of a production outage.
What I like about such hypothetical scenarios compared to postmortem articles is that they usually focus on the process, while postmortems focus on outcomes. In other words, in articles like this you see something like: “We checked an app A and then an app B, and then this DB and that cluster; and finally we found it!”, while postmortems tend to condence the same process into dry statements like: “After thorough investigation we found that it was a DNS issue”.
Thus, such hypothetical articles better showcase an actual process of the incident resolution, in my opinion.
Bonus: InfraCloud (authors of this article) have a repository with such incident scenarios that you can study.
#sre #oncall #incident_management
**Database Fundamentals.**
Because it just fundamentals, it can take a few hours to read and understand + mandatory breaks :)
It's one of the best articles I've seen on general DB topics, with a huge amount of links and notes to go deeper into Rabbit Hole. Definitely recommend it to read.
#databases
Protect the Sky fundraiser has 5% left to close its goal!
It was 10% two weeks ago, when I previously posted this fundraiser.
Thus, I believe it’s possible to close it before the end of the year!
#donations #Ukraine
How Discord Serves 15-Million Users on One Server.
I think this article was posted before on some of DevOps-related channels.
tl;dr: It's BEAM & Elixir. Also, every time you see an article titled "how a chat app scales to crazy number of users", it's always BEAM.
Yet, in their case, Discord had to make some performance analysis and architectural optimizations with some data offloading. That part is interesting.
P.S. I'd love to learn Elixir someday, but I have zero experience with it or Erlang, or FP in general. So, if you're in the same position, and you're looking for a learning buddy, please, let me know :)
#elixir #architecture
For today’s Donations Monday I’d like to share a fundraiser for recon drones by Come Back Alive.
#donations #Ukraine
We don't make posts related to events, unless we are personally invested in those.
Today is one of such examples. I am a part of an organizational committee of the first DOU Technical Conference.
The CfP is still open. However, it's closing tomorrow.
So, don't miss your chance to become a speaker!
The only caveat is that this is an offline event, so you should be able to physically travel to Kyiv on the day of event.
#event
My boss from one of my previous jobs used to say: “Stop starting, start finishing”.
TBH, I wish much more companies adopt this motto. So, let’s lead by example!
A joint fundraiser by Come Back Alive and Kyivstar - “We live here” is almost over! Its goal is to raise 175M UAH and right now they are lacking about 4M UAH to close this goal!
So, let’s help them to finish it!
#donations #Ukraine
Google shares their experience from 20 years of SRE.
This article contains overview and lessons learned from some older incidents. It’s nice, but I can’t say that there is any mind blowing information in there. Anyway, it’s nice that it came from Google.
#sre
Infrastructure as Code and Configuration Management topics are old and boring. It seems like it's almost impossible to have anything conceptually new in this domain, except some drama around licensing.
Yet, in this blogpost Nathan Peck (he works in AWS, IIRC) argues that we could do it differently and that we could do it better.
The core idea is that instead of writing some YAML or DSL to define various resources and then try to group them somehow, we could leverage the same approach that Web technologies took. More specifically, "decorating" objects with properties provided separately. Similarly to what CSS does to HTML.
There's also a discussion of this post on Hacker News
#iac
💡Міністерство оборони України шукає найкращі технологічні рішення для вдосконалення безпілотних систем на фронті.
28-29 січня проводимо Хакатон «Наступ Машин».
Якщо ви — інноватор, інженер чи військовий і маєте ідею щодо розвитку безпілотних систем, реєструйтеся за посиланням: https://forms.gle/Yu7uRrj578ihsXJu6
Серед членів журі Хакатону:
▪️Тарас Чмут (https://www.facebook.com/TarChmut )
▪️Серж Марко (https://www.facebook.com/serg.marco )
▪️Сергій Флеш (https://www.facebook.com/Serhii.Flash )
І це ще не всі – інших анонсуємо згодом.
Перелік челенджів та деталі щодо заходу ви отримаєте після реєстрації.
Наша мета — створення технологічної переваги над ворогом 10:1. Воювати мають машини, а не люди.
Для цього розвиваємо технології перемоги разом! Craft the Future of Unmanned Systems.
Співорганізаторами Хакатону є Генеральний штаб ЗСУ, Міністерство з питань стратегічних галузей промисловості і defence tech кластер Brave1.
You know how it works: people are posting their end-of-the-year results at the end of each year and their predictions for the future at the beginning of each year.
In this video, Viktor Farcic shares his thoughts on the tools to check out (or continue using) in 2024.
#tools #predictions #video
I would like to remind you about the UA Responders foundation that raises funds for tactical medicine.
You can support them here:
https://uaresponders.org/donate/
#donations #Ukraine
A new episode of our voice chat is here! This time we had a single yet broad topic of architecture and the architect role in companies.
The episode is in Ukrainian and available on:
- YouTube
- Substack
- Spotify
- Apple Podcasts
- Google Podcasts
Enjoy!
P.S. Also, CatOps is participating in the DOU Awards "Podcasts" department! I would really appreciate it if you could support us here with your vote!
#voice_chat
Let's start a new year with something entertaining, like talk about hacking train DRM to be able to fix it.
Additional points, if you'll describe in comments which DevOps, Software Engineering, and Legal Practices were violated by the train manufacturer (:
And, don't forget to support Ukraine 🇺🇦.
I was about to leave for the wnter holidays, but then I recalled that I still owe you the last issue of the CatOps Digest of this year.
So, here you are!
https://open.substack.com/pub/catops/p/catops-digest-2023-12-26
I wish you very pleasant holidays and see you next year!
#digest #newsletter
Let's talk a little bit about culture. Culture of sabotage!
That's a nice guide of how to drop productivity to a minimum without being caught. Enjoy!
https://erikbern.com/2023/12/13/simple-sabotage-for-software.html
#culture
A Linux book bundle on Humble Bundle. Now, I know that people have mixed feelings about books by Packt. Some folks say that they are too basic.
So, if you’re unhappy about the Packt books, you may want to check the Software Development bundle that has a couple of books on CI/CD, DevOps, and SRE topics.
#books
If you are running dynamic environments in Kubernetes or any other ephemeral workloads, you need a way to clean up things once these environments are no longer needed or once your tests are done.
k8s-cleaner may help with it. It's a controller that deletes Kubernetes resources, including custom resources, on a schedule. It supports dry run and some options for customizations.
#kubernetes