jiisanda

Building:

[jiisanda @jiisan_da]

Most SMBs think a standard corporate VPN means they're secure.

They don't realize an unencrypted, malware-ridden personal laptop can just tunnel straight past their firewall.

Next Wednesday (23rd Sept), we're doing a live, zero-BS teardown of a real SMB breach timeline. No product pitch, just a look at what actually stops lateral movement and what fails.

If you're managing remote team security (as a growing company or an MSP) and want to join, reply "Inv" below and I'll DM you the link as soon as it goes live! 🖥️🔒

[jiisanda @jiisan_da]

Learned about Claude Code hooks today. its: when X happens, automatically run Y.
They're not just for after-the-fact stuff like auto-formatting Go code once it's written. They can also run before actions, like guardrails that check a command before it executes.

Anyone using custom hooks? What for?

[jiisanda @jiisan_da]

@4wmturner @blvckitan @FastAPI Yep, for FastAPI's own request spans you can drop opentelemetry-instrumentation-fastapi. Install fastapi[standard], set OTEL_SERVICE_NAME + OTEL_EXPORTER_OTLP_ENDPOINT and you get traces, metrics & logs. Uvicorn's fine, it hooks in at the app level so any ASGI server works...

[jiisanda @jiisan_da]

You won't learn these by reading. Break them on purpose:
→ Kill a health check
→ Set a 1s upstream timeout
→ Cap a pool at 2 connections
→ Replay the same webhook twice

Then read the logs.

Which one has bitten you? 👇

[jiisanda @jiisan_da]

A lot of "backend bugs" aren't in the code. They're somewhere in the path a request takes.

8 traffic flows every backend engineer should be able to trace: 🧵

[jiisanda @jiisan_da]

8. Laptop → Private server
Trace:
Laptop → VPN/Bastion → Private subnet → Server
Understand:
→ Private subnets
→ Keys, not passwords
→ Firewall rules
→ Port forwarding

If port 22 is open to the internet, you're on every scanner's list.

[@kai_infra | IT Infrastructure Engineer @sideb_01]

「eBPFってよく聞くけど、Linuxの何が見えるの？」

topやiostatでは結果は見えても、
「Kernel内部で、今どんなEventが起きているか」までは追いにくいことがあります。
eBPFを使うと、KernelのTracepointなどに処理をHookして、System Call・Network・I/OなどのEventを動的に観測できます。

図はbpftraceで openat() を監視した例👇

sys_enter_openat をTrace
↓
Process名 comm
＋
File Path args.filename

これだけでも「どのProcessが、どのFileへアクセスしているか」をリアルタイムに追えます。

eBPFの仕組みからbpftraceを使ったLinuxトラブルシューティングまで、実機例でまとめました👇
https://t.co/wzXgRmQ3ML

#Linux #eBPF #bpftrace #インフラエンジニア

[Jason Lee @huacnlee]

MinIO 已经 GitHub 公布停止维护了，本地开发，需要用 S3，现在最好的选择是 RustFS 了。

https://t.co/nstAKq1U5R

[jiisanda @jiisan_da]

When do you guys prefer attending webinars?

[jiisanda @jiisan_da]

@RelayAILabs @codewith55 may be correct on support and consistency, but for getting a small project off the ground without AWS-level bills, it does the job.

[bABA @BabaDev_]

MARKETING IS 10X HARDER THAN BUILDING.

Maybe 100x.

Fight me.

[jiisanda @jiisan_da]

@ishiitaa_dev twist: choose polyrepo but try to stitch them together with Git submodules. Now you have 47 repos, a detached HEAD state, and a broken dependency graph.����💀💀