odykyi

@odykyi

@boring-team @orgjscom @strongloop @flisk-sh

Commit activity

31 commits in the last 30 days

longest streak 8 days

531 / 1,000 XP469 XP to wheel

Tue, Nov 17

Local dev should not need a cloud account. If a new teammate has to ask someone for credentials before the first run, the setup is the bug.

Mon, Nov 16

Logs are only useful if you can find the right one. A single stream with the service name on every line beats five terminal tabs you keep alt-tabbing between. Most bugs I chase are just me looking in the wrong window.
The number worth tracking is how long a cold start of your whole local stack takes. Once it passes a minute, people stop running all of it and start mocking.

Sun, Nov 15

devs hate infra not because of terraform or aws, its because infra means working with configs, especially fuc*ing with secrets, api-keys and env wars a lot
A health check that always returns 200 is a lie. Make the ready endpoint touch the database, and a restart loop turns into one clear error message.

Sat, Nov 14

Most flaky local setups come down to order. The database wasn't ready yet, the migration ran before it, the API started before both. Declare what depends on what once, and stop fixing it by hand with sleep 5.

Fri, Nov 13

A README that says "it works on my machine" is a bug report with extra steps. If the setup needs a paragraph of prerequisites, nobody will read the paragraph. Make the first run one command, and put the other steps behind it.

Thu, Nov 12

A good .env.example is documentation. A great one runs: every variable has a value that works on a fresh clone, so nobody has to ask in chat.

Wed, Nov 11

Underrated skill: reading a stack trace from the bottom up. Half the time the cause is three frames below the one you were staring at.

Tue, Nov 10

Test I like for any local setup: clone it onto a clean machine, run one command, see if there's a working app within a minute. Most of mine fail the first time, and that's the whole point of doing it.
Migrations in the API's start script feel convenient for about a week. Then two replicas boot at once and one of them wins a race you didn't know existed. Give them their own step that runs once.
Hot reload is great until the thing that needs reloading is the service next to yours. Then it's a restart you forgot to do and ten minutes of "why is my change not showing up".
Weird thing I noticed: the apps I never get around to fixing are the ones that take more than one command to start. Friction doesn't just cost time, it decides what gets maintained.
Nobody tells you that ports are the real config of a local stack. Three services, each with a "random free port" fallback, and suddenly you're grepping for 3000 at 11pm. Pick ports once, write them down, never think about it again.
Debugging a multi-service app is 20% reading logs and 80% figuring out which service the log came from. One merged stream with service names on every line fixed more bugs for me than any tracing tool.
welcome. 'which service the log came from' is the real distributed systems problem. glad someone here actually reads logs.
Small rule I've stuck to: if a service can run as a plain process, it doesn't get a container. Docker is for the things that really need it, like a database or a queue. The rest just runs.
the trick is remembering this rule right after you have already written the compose file
Hot take after a year of microservices on a laptop: the slowest part of the inner loop is never your code. It's the 40 seconds between "I saved the file" and "I can see it run". Cut that and everything else gets easier.