Hola.. Soy una chica de 19 años, vivo en la calle y ando buscando trabajo. Tengo los 2 carnet, hablo ingles nativo/fluido pero no he terminado la secundaria. No tengo economia para pagar la secundaria (hasta las publicas son caras, tuve que salirme porque ya no pude con todo los gastos). He aplicado a todo y siempre me dicen que no porque no tengo diploma. Hasta call centers y comida rapida (mcdonald y burger king) me han rechazado por no tener diploma. Alguien sabe algun trabajo/empresa/lugar que te acepte sin diploma aqui en la ciudad de Panamá??

Hace unos años estuve en un proceso de selección y me quedé con la duda de como es trabajar alli. Que tal la cultura? Salarios? Beneficios?

I have been aggressively migrating from AWS permanent credentials to OIDC in GitHub Actions, mainly for deploying to ECS.

I know GitHub Actions were supporting OIDC for a while now. But the pressure on compliance is the reason for this migration.

If you are new to OpenID Connect (OIDC), it allows GitHub runners to mint short-lived (15–60 min) STS tokens on-the-fly with zero stored secrets.

Here’s a quick breakdown of how it works, the Terraform/OpenTofu setup, and the subtle gotchas that I faced.

1. How It Works Under the Hood

1. When your workflow job starts with id-token: write, GitHub's OIDC service generates a cryptographically signed JSON Web Token (JWT).
2. The aws-actions/configure-aws-credentials action sends this JWT to AWS STS via sts:AssumeRoleWithWebIdentity.
3. AWS validates GitHub's signature, checks your IAM Role's Trust Policy (to ensure the token came from your exact repo and branch), and returns temporary STS credentials.

2. The Infrastructure Setup (Terraform / OpenTofu)

You only need two AWS resources: an OIDC Provider and an IAM Role with a Trust Policy.

hcl # 1. The GitHub OIDC Identity Provider resource "aws_iam_openid_connect_provider" "github" { url = "https://token.actions.githubusercontent.com" client_id_list = ["sts.amazonaws.com"] thumbprint_list = [ "6938fd4d98bab03faadb97b34396831e3780aea1", "1c58a3a8518e8759bf075b76b750d4f2df264fcd" ] } # 2. IAM Role with Scoped Trust Policy resource "aws_iam_role" "github_deploy_role" { name = "github-actions-deploy-role" assume_role_policy = jsonencode({ Version = "2012-10-17" Statement = [{ Effect = "Allow" Principal = { Federated = aws_iam_openid_connect_provider.github.arn } Action = "sts:AssumeRoleWithWebIdentity" Condition = { StringEquals = { "token.actions.githubusercontent.com:aud" = "sts.amazonaws.com" } StringLike = { # Restrict exclusively to your repository & branch/tags "token.actions.githubusercontent.com:sub" = "repo:your-username/your-repo:*" } } }] }) }

3. The GitHub Actions Workflow

In your .github/workflows/deploy.yml

name: Deploy to AWS on: push: branches: [ main ] jobs: deploy: runs-on: ubuntu-latest permissions: id-token: write # CRITICAL: required to request the OIDC JWT contents: read steps: - name: Checkout code uses: actions/checkout@v4 - name: Configure AWS Credentials via OIDC uses: aws-actions/configure-aws-credentials@v4 with: role-to-assume: arn:aws:iam::123456789012:role/github-actions-deploy-role aws-region: us-east-1 - name: Verify Authentication run: aws sts get-caller-identity 

Three real-world gotchas that will save you hours

If you get Error: Could not assume role with OIDC: Not authorized to perform sts:AssumeRoleWithWebIdentity, check these 3 things:

1. Case Sensitivity in the sub claim: AWS IAM condition strings are case-sensitive. If your GitHub repo or username uses mixed casing (e.g. MyOrg/Repo), make sure your IAM sub condition matches the exact casing GitHub sends in the token. Using wildcard matching (repo:MyOrg/Repo:*) helps avoid exact ref string mismatch issues.
2. Job-Level vs. Workflow-Level Permissions: Always set permissions: id-token: write on the specific job, not just globally at the top of the YAML file. Some runner configs don't inherit top-level permissions to nested jobs.
3. CA Thumbprints: Don't dynamically query GitHub's leaf certificate for thumbprints in Terraform—they change frequently with CDN updates. Use GitHub's official intermediate root CA thumbprints:
   • 6938fd4d98bab03faadb97b34396831e3780aea1
   • 1c58a3a8518e8759bf075b76b750d4f2df264fcd

Summary

• No stored secrets in GitHub settings.
• No key rotation schedules to manage.
• Granular security. You can restrict deployment roles to specific branches or environments.

Are you already using OIDC for your pipelines, or are you still relying on IAM users? Curious how folks here handle multi-account / cross-account OIDC setups.

Estimada comunidad, queria consultar en donde y aproximadamente el costo de prueba de Paternidad, y si estas son funcionales para presentar como evidencia en un juicio.

Muchas gracias gente.

These SQL injections have been coming in daily to this site which has still seemingly managed to avoid infection simply because of the changed database prefix. This one uses the “author not in” query bug inside some crazy nested query to place the payload in the database.

This is all related to the WP2Shell bug which was reported over a month ago. The site had auto updates disabled and still managed to avoid getting infected because of some seemingly silly security practices like changing table names and database prefixes. As far as I can tell all of these failed. I was alerted to the issue because the site got locked in a loop of timeouts, perhaps related to these malformed queries. Unfortunately I’m going to have to do some kind of a forensic workup even though there were no clear indicators of compromise because requests to the batch endpoint resulted in many 207 codes which means they may have succeeded.

But when I saw the giant block of 0s and 1s and hex codes in the logs, my heart dropped into my stomach. Stay safe out there people and leave auto updates on. It’s worth the chance of breaking your site every once in a while.

Recently I had an idea for a research tool, I realized I'd need to be able to find websites programmatically and the webs data is behind the largest corporations, not something we reliably have access to for being able to develop around.

So it got me thinking, how lightweight could you first build a crawler if you only cared about getting meta data (titles, description and the url)? Then it got me thinking, could we not share the load, create a distributed peer to peer network and make this data decentralized?

So what I have built is the open web directory. A network that crawls and builds a yellow pages like directory for web pages online.

You can setup your own node and contribute to the network by simply running:

git clone https://github.com/idev-games/the-open-web-directory.git cd the-open-web-directory npm start 

And open port 80 for that device by port forwarding.

I think this is going to be a fascinating experiment, it's early days so expect bugs and issues but give it ago and let's see if we can index the web.

Check out more info here:

https://github.com/iDev-Games/The-Open-Web-Directory

The site listed on github is the live front end (you can host the html files in the public folder anywhere to open another frontend on the web).

I was just notified that my website was down, I tried to go to the site and I am getting the 404 error, I do not know what or where to go. I have it through cloudflare and designed it with readdy.ai. It worked fine and now all of a sudden its down, could anyone help me with this? Thanks

edbuild.net

Hello, I have been thinking about a project I‘d like to try out, where I have like a Database, an api server and a web server. It‘s not too large, and mostly to learn about how they all interact. The issue is, I would prefer not having to pay for multiple servers to host everything. Is there a way to simulate how it would be in real life on a single machine?

Hi everyone,

I’m looking for a good-quality laptop backpack mainly for work, commuting, customer visits, and occasional business travel.

I’d like something practical and durable, but still professional-looking rather than a hiking or tactical backpack.

My main priorities are:

- good protection for the laptop, preferably a separate padded laptop compartment

- comfortable shoulder straps and back panel

- good internal organization for charger, cables, mouse, headphones, documents, etc.

- quick-access pocket for keys / phone / wallet

- space for a water bottle

- durable materials and good-quality zippers

- some water resistance would be a plus

- preferably a luggage pass-through for attaching it to a suitcase

- professional / minimalist design

- preferably something that will last for many years

It will mainly be used for everyday work, but occasionally I’d also like to use it for 1–2 day business trips.

I’m based in Europe, so I’m mainly interested in brands/models that are easily available in the EU without expensive international shipping, customs, or import fees.

I’m not necessarily looking for the cheapest option — I’d rather pay more for something comfortable, well designed, and durable.

What backpacks are you actually using and would recommend?

I’m especially interested in long-term experience: how long have you owned it, what do you like about it, and what annoys you?

Thanks!

Hey guys, just after opinions really. If you’re launching a somewhat large site, with quite a bit of stuff going on, would you rely on your own testing or would you suggest getting a third party tester? It’s mostly for things like UX I’m wanting to test. I know have I would expect things to work but being the sole developer I think I may have gone a bit blind to somethings. I have friends do bits of pieces in terms of checking things but I wonder if the consensus would be to just get an actual tester to go through everything, testing the functionality all works, different variations, ensure the UX is as good as it can be etc or do you think I’m overthinking it?

Este año que pusieron los talleres, les informaron a los panelistas que Cobre Panama era el patrocinador, y se corrio la info hasta los artistas del valley. Y no se han atrevido a subir un post confirmándolo para que cierta gente crea que es fake News.Están ocultando los mensajes de no a la mina en los comentarios.

Mi pregunta es cómo se te ocurre hacer eso para un evento de la gente más woke ( «Woke» (que significa «despierto» en inglés) es un término que describe a una persona consciente y alerta ante las injusticias sociales, políticas y raciales) de Panamá. Es que , que esperan ? Uno quiere ir tranquilo a un evento que tiene NADA QUE VER y meten eso que es un tema delicado en Panamá y es hasta político.

ojo no hay nada de malo en ir ( según mi perspectiva ) por que uno gasto su dinero, y muchos lo ahorraron y el público también es victima de que no digan las cosas de forma clara luego vas y te das cuenta de eso. Y los pequeños artistas que pagaron para estar ahí espero que les vaya bien este año.

Quiero saber qué opinan?

We’ve gone full-Rust with our build pipeline at Outlyne.com and it feels like the React Compiler has really arrived. The last time I posted about using React Compiler here, I talked about how good it’s been for us overall, but also about all its limitations, the biggest of which was that it would bail out on compiling any component or hook that had a try/catch with conditional logic of any kind (if/else, ternary, ??, ||, optional chaining). Another big one was that it would bail on computed object keys, which bit us a few times with assembling classes using clsx (e.g. className={clsx({ [`item-${props.index}`]: props.isItem })}).

Both of those are fixed now. The oxc team shipped official support for the Rust React Compiler on August 4th, and that support includes all of the fixes to the compiler since the v1.0 release of the Babel-based compiler. In our codebase, seven more functions the compiler had been skipping are now compiling, five from the try/catch fix, two from computed keys. And that’s lower than it would be if I hadn’t gone through and contorted eight components to keep conditionals out of their try/catch blocks, including adding a @ts-expect-error from having to drop an optional chain and extracting my auth flow into an async run() that’s immediately awaited inside the try, for no reason but to move the value blocks out of it. All of those workarounds have been rendered obsolete.

There are still bailouts, most notably logical assignment operators (??=, &&=, ||=), but far fewer, and now actively being addressed. A couple weeks ago, oxc shipped a version bump that prevents a bail out from reassigning a destructured prop that then gets read in a nested closure.

Going Rust is also, of course, much faster. The compiler step on our 1,036 files dropped from 14.3s to 0.81s (~17.6×), and the build overall from 22.1s to 9.3s. And we verified parity with the Babel compiler. Across all 1,036 files, there are zero functions the Babel compiler memoizes that the native one doesn’t. The only difference in the built JS before and after are the seven extra compiled functions.

I wrote a blog post with all the details and before/after configs both for folks using @vitejs/plugin-react and for those who use an alternative plugin (React Router framework mode’s plugin in our case): https://blog.master.dev/react-now-rusted-all-the-way-out/