Taming the OOM Killer: Process Prioritization for Memory-Constrained Linux Systems

Posted on Fri 18 April 2025 in DevSecOps • Tagged with linux, oomkiller, memory, system-administration, devsecops, process-management, hardening

In memory-constrained environments, the Linux OOM Killer decides what lives and what gets killed. This guide shows how to protect critical processes like sshd and mysqld using oom_score_adj values, with a script that applies them reliably and securely. Make memory pressure predictable and survivable.


Continue reading

Catching a Nation-State Proxy: OSINT Lessons from the Twitter Frontlines

Posted on Thu 17 April 2025 in Threat Intelligence • Tagged with osint, threat-intelligence, phishing, venezuela, twitter, surveillance, devsecops

In 2012, I uncovered a state-aligned Twitter proxy tied to Venezuela’s ruling party. It mimicked Twitter, redirected traffic, and risked phishing user credentials. This post breaks down the OSINT methods I used to uncover it — and why threat intel teams still need to watch for subtle, state-run infrastructure.


Continue reading

The 208.5-Day Kernel Bug: A Lesson in Uptime, Overflow, and Operational Risk

Posted on Wed 16 April 2025 in DevSecOps • Tagged with kernel, bug, Linux, uptime, overflow, devsecops, integer-overflow

A 2012 Linux kernel bug caused CPU lockups after 208.5 days of uptime due to an integer overflow in sched_clock(). Affecting RHEL 5 and 6, it exposed the risks of long uptimes, underscoring the importance of timely patching, uptime observability, and operational risk management in DevSecOps.


Continue reading