ReliabilityOps ExpertiseToolsField NotesContact
Tools & Resources

Things I built that you can use.

Free, offline-first operations tools and a monitoring product. Everything read-only or safe by design, most needing no account and no install. Built from real incidents, for DBAs and SREs.

FLAGSHIP · FREE

SQL Field Kit

An offline triage toolkit for SQL Server Always On / HADR and PostgreSQL. Failover audits, an AI analyst that turns a paste into a runbook, and a full SSMS-style training simulator. All in the browser, connecting to nothing. The tool a locked-down team can actually adopt.

Triage → AI Analyst → Training Sim → PostgreSQL → AG Fundamentals →
AI

AI Analyst

Paste an error log, wait stats, or DMV output. Get severity, likely cause, what to check next, and the fix. Nothing stored.

Free · runs on your paste
TRAINING

Failover Simulator

Real HADR incidents replayed in an SSMS + PowerShell sandbox. Novice to expert, no lab servers needed.

Free · runs in-browser
MONITORING · SAAS

StatusOwl 🦉

Uptime and status-page monitoring I built and run. Know before your users do. Checks, alerting, and a clean public status page.

The Field Kit family

The same offline, read-only philosophy across the stacks you actually run. SQL is live; the rest are rolling out.

LIVE

SQL Server 🗄️

HADR triage, failover audits, AI analyst, training sim. The flagship.

Free · in browser
LIVE

PostgreSQL / RDS 🐘

Replication, WAL, vacuum, and RDS/Aurora triage. The Postgres companion.

Free · in browser
IN DEVELOPMENT

Linux Field Kit 🐧

Paste top, dmesg, systemctl, disk and memory output. Triaged with the right next command. A read-only "what's wrong with this box" companion.

Building now
IN DEVELOPMENT

Terraform Field Kit 🏗️

Read a plan, spot the destructive changes before apply, decode state drift, and catch the "this will replace prod" line that scrolls by.

Building now
IN DEVELOPMENT

Docker Field Kit 🐳

Container won't start, image won't build, network won't route. Paste the logs and inspect output, get the likely cause and the check.

Building now
IN DEVELOPMENT

Kubernetes Field Kit ☸️

CrashLoopBackOff, pending pods, OOMKills, and the events nobody reads. Paste kubectl describe and get triaged.

Building now
IN DEVELOPMENT

PowerShell Toolkit ⚡

Read-only diagnostic scripts and one-liners for Windows and SQL. Pull the status of a service, a session, an AG, without memorizing the incantation.

Building now

Utilities

IN DEVELOPMENT

Log Analyzer 🔎

Paste a wall of log output, SQL error log, syslog, application log. And get it grouped, ranked by severity, and pointed at the lines that matter. Pattern-first, offline, no upload.

Building now
IN DEVELOPMENT

Query Writer 📝

Describe what you need to check, "is this AG healthy," "what's blocking session 52," "which logins are orphaned". And get the read-only script to run. A safe recall layer for the queries you always half-remember.

Building now
IN DEVELOPMENT

Incident Command Kit 🎯

A browser-based coordinator that walks a solo on-caller through roles, timeline capture, comms templates, and a severity decision tree. Then exports a clean postmortem skeleton. Nothing stored. Pairs with the AI Analyst.

Building now
IN DEVELOPMENT

SLO / Error-Budget Calculator 📉

Enter request and error counts, get burn rate, budget remaining, and alerting-threshold suggestions. Fully client-side. The kind of thing every SRE team needs and can never get past procurement.

Building now
IN DEVELOPMENT

Runbook Builder 📗

Turn messy notes into a structured, testable runbook. Preconditions, steps, rollback, verification. Extends the "turn a paste into a runbook" idea end to end.

Building now

Learn & open-source

The tools above are mine. Read-only and safe by design. The projects below are established open source that run real infrastructure. They're how you go deeper.

⚠️ These touch production systems. Read the docs before you point them at anything real. Unlike the kits above, they are not read-only sandboxes.
INCIDENT & ON-CALL

Response tooling

Battle-tested incident management and on-call.

Netflix Dispatch → Grafana OnCall → PagerDuty Incident Response docs →
CHAOS & RESILIENCE

Fault injection

Break it on purpose, before it breaks you.

Chaos Mesh (CNCF) → LitmusChaos (CNCF) → Toxiproxy (network faults) →
SLOs & ERROR BUDGETS

Reliability targets

Define and enforce what "reliable" means.

Sloth (Prometheus SLOs) → Pyrra (SLO management) → OpenSLO (the spec) →
OBSERVABILITY

The reference stack

Metrics, traces, logs, pipelines.

OpenTelemetry → Prometheus + Alertmanager → Vector (pipelines) →
DOCTRINE · START HERE

Learn the craft

Free, foundational SRE reading.

Google SRE Book & Workbook → awesome-sre → awesome-chaos-engineering →

Downloads & references

⬇️
Invoke-HadrAudit.ps1
Read-only PowerShell fleet audit. One CSV per section + findings, across a server list
📘
AG Fundamentals
How Always On availability groups actually work. The why behind every check
📝
Field Notes
Failover war stories, HADR deep-dives, and reliability tooling philosophy