High-Reliability SRE & Hardware Toolkits

Stop Crashing Production with Legacy Tools. Observe Linux from Kernel Ring 0.

Traditional Linux troubleshooting commands (strace, iotop, lsof) were designed decades ago. In modern containerised, high-throughput cloud environments (Docker, Kubernetes, Proxmox VE), using them under load can cause catastrophic outages:

• strace causes 10x–50x latency penalties by pausing processes via ptrace on every syscall, frequently deadlocking production databases.
• top & htop cannot detect CFS CPU Quota Throttling inside containers, leaving engineers baffled when API latency spikes while reported CPU usage is only 35%.
• df -h vs du -sh discrepancies ("Ghost Space") leave sysadmins helpless when unlinked log files remain locked by unclosed file descriptors.
• Commercial APM Agents (Datadog, New Relic) cost $15–$23 per container per month and consume 300MB+ host RAM.

⚡ Enter the SpecInfo eBPF Production Toolkit

The SpecInfo eBPF Production Toolkit injects verified, non-invasive micro-programs directly into Linux Kernel Ring 0 via eBPF and bpftrace. It delivers zero-overhead, microsecond-accurate observability on live production nodes with 0% CPU drain and zero risk of process starvation.

📦 What’s Inside the Box (v1.0 Suite):

1. 🖥️ Interactive Cyberpunk Terminal HUD (specinfo-ebpf)

An enterprise-grade, color-coded interactive command-line dashboard. Run any diagnostic script with a single keystroke, inspect Kernel BTF status, and run automated pre-flight self-tests.

2. 🛡️ 8 Battle-Tested Production Modules:

1. 01_sre_sentinel.bt — Real-time privilege escalation (sudo/su), destructive commands (rm/dd/mkfs), and container execution watchdog with ANSI alert styling.
2. 02_sre_ghost_space.bt — Unlinked file deletion tracker (unlinkat). Catches runaway processes silently holding gigabytes of deleted disk space.
3. 03_sre_inode_drain.bt — Inode allocation rate monitor (vfs_create). Pinpoints infinite file creation loops before df -i reaches 100%.
4. 04_sre_disk_latency.bt — Microsecond storage I/O latency log-histogram. Pinpoints SSD/NVMe tail-latency stalls hidden by average MB/s metrics.
5. 05_sre_oom_watch.bt — Zero-overhead Out Of Memory (OOM) killer alert. Sits idle at 0% CPU until oom_kill_process() triggers, broadcasting instant victim PIDs.
6. 06_sre_tcp_drain.bt — Socket-layer outbound TCP bandwidth monitor (tcp_sendmsg) per process without heavy packet sniffing.
7. 07_sre_container_watch.bt — Container runtime lifecycle watchdog (runc, containerd-shim, dockerd, podman) observing container execution from host Ring 0.
8. 08_sre_cfs_throttle.bt — Container CFS CPU quota throttling & cgroup memory starvation sentinel. Unmasks hidden container latency spikes.

3. 📄 Vector A4 Printable Emergency Triage Runbook (PDF & HTML)

A high-density, 300 DPI Eco-White printable reference card. Designed to be laminated and kept by the server rack or under your keyboard for offline air-gapped triage when networks and AI portals are unreachable.

4. 🛠️ Universal Auto-Installer & Global CLI Integration (install.sh)

One-command dependency setup for Arch Linux, Ubuntu LTS, Debian, RHEL, Rocky Linux, AlmaLinux, and Proxmox VE. Automatically configures global system binary /usr/local/bin/specinfo-ebpf.

💻 Technical Requirements & Compatibility:

• Kernel Support: Linux Kernel 5.4+ with eBPF enabled (Kernel BTF /sys/kernel/btf/vmlinux supported out of the box).
• Supported Distros: Arch Linux / Manjaro, Ubuntu 20.04/22.04/24.04 LTS, Debian 11/12, RHEL / Rocky / Alma 8/9, Proxmox VE 7.x/8.x/9.x.
• Dependencies: bpftrace (automatically installed via install.sh).

🛡️ Solopreneur Commercial Guarantee

Engineered and maintained by Krzysztof Damek at SpecInfo.org. Designed for production SREs, DevOps engineers, and Linux datacenter teams who need surgical clarity without enterprise SaaS bloat.