Fleet Security Audit
Auditing 125 production sites across eight servers, then fixing what mattered first
- 125 sites audited
- Servers patched and hardened
- 2 servers shut down
- Industry
- Multi-client web hosting
- Status
- Completed
- Audit scripts
- Normalized dataset
- Remediation workbook
- Workstreams
The Problem
Years of steady growth had left 125 client sites spread across eight EC2 servers and a managed hosting account, each set up a little differently. Without one consistent view of what was running where and what was out of date, there was no way to rank the risk.
The Approach
I built a multi-script audit pipeline with variants for Apache servers, nginx servers, and managed hosting, so every site got checked the same way no matter where it lived. When the audit surfaced an issue that couldn't wait for the final report, I handled containment first and kept the audit moving.
The Architecture
The scripts collect configuration, software versions, and application-level details from every host and normalize them into a single dataset. That dataset feeds a remediation workbook that ranks each finding by severity and groups the fixes into workstreams with effort estimates.
The Results
Leadership got a prioritized remediation plan with real estimates in place of a vague sense of dread. Then the fixes shipped: servers patched and security-hardened, and two shut down entirely. The audit scripts are reusable, so the next audit is a rerun instead of a project.
Built with AWS EC2, nginx, Apache, Bash, WordPress, Security
Not sure what's running on your servers?
I'll audit it. You get a ranked list of what's wrong, what it'll take to fix, and what can wait.
Serving the Central Valley: Sacramento · Stockton · Modesto · Manteca, Tracy & Lathrop