CodeHiRise

Wander in the wonderful world of electronics,
web dev and in between

[CodeHiRise @codehirise]

@HsanC_ Congrats 🎊 its not easy to grow a saas

[CodeHiRise @codehirise]

@EOEboh @umami_software 🔥

[CodeHiRise @codehirise]

@brankopetric00 Use EC2 Auto Scaling Groups for horizontal scaling.
Use CloudFront CDN for caching with VPC origins via an internal load balancer.
Use Redis to cache expensive DB reads and DB read replicas.
For writes, use an SQS queue for handling bursts.

[CodeHiRise @codehirise]

@brankopetric00 Monolith on ECS should be a good starting point. Revisit EKS if there is an absolute need for it. Saving money is crucial for a startup in the early stage.

[Socket @SocketSecurity]

Yep, that works as a lightweight local guardrail. It makes Socket Firewall the default path for everyday installs.

For macOS/Linux users, the equivalent in zsh/bash would be:

alias npm="sfw npm"
alias yarn="sfw yarn"
alias pnpm="sfw pnpm"
alias pip="sfw pip"
alias uv="sfw uv"
alias cargo="sfw cargo"

[CodeHiRise @codehirise]

Day 7 of 7 cloud best practices that can save your production launch 👇

7/7
Tag your cloud resources early.

Every resource should have tags like:
• Environment
• Service name
• Owner
• Cost center

Untagged infrastructure becomes impossible to track as systems grow.
#devops

[CodeHiRise @codehirise]

Spent a few hours debugging a Next.js app on ECS Service Connect failing health check.
Turns out I missed setting HOSTNAME=0.0.0.0 in task definition env vars 🤦‍♂️

#nextjs #aws #devops

[CodeHiRise @codehirise]

Try using iac tools like Terraform/Pulumi for your next project.

@HashiCorp @PulumiCorp

#terraform #pulumi

[CodeHiRise @codehirise]

Day 6 of 7 cloud best practices that can save your production launch 👇

6/7
Use Infrastructure as Code from day one.

Manual cloud changes feel fast early on
Until nobody remembers:
• What changed
• Who changed it
• Why it changed

#terraform #iac #cloud #devops #aws

[CodeHiRise @codehirise]

That’s how cloud environments turn into unmanageable snowflakes.

With Infrastructure as Code everything is version controlled and reproducible down to the exact state.

This makes environment provisioning and management far more predictable.

[CodeHiRise @codehirise]

@tan_stack @TansTack @npmjs colab with @SocketSecurity

[CodeHiRise @codehirise]

@tan_stack @TansTack At this point i think its better to scan every package version by npmjs before publishing.

[CodeHiRise @codehirise]

These values should live in configs, secrets managers, or environment variables not inside your codebase. This will make sure we can easily swap them without costly database backfills.

#cloud #devops #software #development #aws

[CodeHiRise @codehirise]

Day 5 of 7 cloud best practices that can save your production launch 👇

5/7
Never hardcode environment-specific values into your application.

Things like:
• Bucket names
• Queue URLs
• Regions
• API endpoints
• Database hosts

[CodeHiRise @codehirise]

But do not run critical customer-facing systems or stateful apps like databases on infrastructure that can disappear with little warning.

Before using Spot ask:
“What happens if this server vanishes in 30 seconds?”

Cheap infrastructure becomes very expensive during outages.

[CodeHiRise @codehirise]

Day 4 of 7 cloud best practices that can save your production launch 👇

4/7
Not every workload belongs on Spot instances.

Spot is great for:
• Batch jobs
• Background workers
• Temporary workloads

If interruption is acceptable, Spot can massively reduce cloud costs.

[CodeHiRise @codehirise]

Randomly changing outbound IPs will eventually break:
• Third-party integrations
• Firewall rules
• Webhooks
• Enterprise whitelisting

Static IPs save a surprising amount of operational pain later.

[CodeHiRise @codehirise]

Day 3 of 7 cloud best practices that can save your production launch 👇

3/7
Keep your ingress and egress IPs predictable. These are the main entryways and exits to your environment.

[CodeHiRise @codehirise]

Use:
• Public subnets for load balancers and nat gateways
• Private subnets for applications and databases

A single misconfigured security group should not expose your entire infrastructure.

Good network isolation makes your infrastructure secure and keep things organized.

[CodeHiRise @codehirise]

Day 2 of 7 cloud best practices that can save your production launch 👇

2/7
Do not put everything in public subnets.

Your databases, services, queues, and private workloads should never be directly exposed to the internet.

Public access should be intentional, not the default