What DevOps practices (tools) do you use to deploy infrastructure?

Question

Work organization

Introduction

Hi
Quite recently, I started using DevOps tools in practice and I can roughly imagine what it is for
(configuration management (ansible, cheff, pappet, salstack, etc.), clouds, git, jenkins, monitoring and logging systems, separating environments into dev groups, testing ,staging,prod )
Отчасти я понимаю, что практика DevOps в целом заключается в автоматизации действий, быстром внесении изменений в инфраструктуру, максимально быстром(мультиплатформенном) деплое в новую среду(будь то клауд или голое железо), разработки инфраструктуры для автоматического масштабирования, предоставлении сред под разные задачи ( testing, development, staging, prod) и централизованного максимально детального сбора логов, метрик и service discovery со всего этого добра. К сожалению на данный момент, так как я full-time и один выполняю обязанности Ops, в конторе которая динамично развивается, и до этого не имел представления о том как строится инфраструктура - я не имею целостной картины инфраструктуры - используемой на практике в больших софтварных конторах. Нет никакого Agile, никто даже не ставит задачи, все на словах. На деле, никто не заинтересован внедрять практики DevOps. В общем пока в голове каша. ). Теперь собственно к чему этот набор слов:

Please advise where you can at least superficially (in theory) get acquainted with the finished, working infrastructure scheme in order to generally understand the categories and bundle of all tools that are used and in what sequence (workflow).
Example: I have an ansible playbook that deploys a git repository on a group of servers, but I roughly understand that most likely the "big guys" do this not through the playbook, but through jenkins + docker + webhook, but in practice I look at this I can't :)
Thank you.
Criticism is welcome and encouraged)

PS

ЗЫ: нет сеньора, который скажет, так никто не делает - потому что так и так... Есть группа dev-ов которым плевать на инфраструктуру так же как и руководству - работает прекрасно, не работает - сам дурак. А мне нужен "experience adviсe"(набитые другими шишки и best practice).

Answer 1

- Terraform for cloud infrastructure configuration + chef for automation.
- On docker swarm servers for small projects.
- CI/CD on gitlab

Answer 2

Example: I have an ansible playbook that deploys a gitrepository on a group of servers, but I roughly understand that most likely the "big guys" do this not through the playbook, but through jenkins + docker + webhook, but in practice I don’t look at this can :)

For big uncles, this can be done by hand. There is no one best way, there are certain rules of the game (budget, number of servers, number of components and stack) - and choose the tools that are convenient for you.
Also, the use of docker is not always justified.
Often the most convenient tools are not the best in the world, but those that are comfortable for you and developers (but this does not mean that you do not need to periodically learn new things and try to look at your infrastructure with a clear eye).
Try not to optimize anything, but to solve specific problems, such as "here I spend xx hours weekly on a routine. Does it make sense to spend a week on optimization and then save on it later, or the routine is associated with processes that cannot be automated (for example, due to bureaucracy).

Answer 3

- Ansible
- Molecule
- Gitlab
The roles are edited by the developer, local tests are run, everything is pushed to the turnip, then the ci / cd launch is triggered, the tests are run there, MR, code review, tests, rolling out to prod.