Snapshot
What the project solved
Important internal information was spread across shared drives, email threads, and old Word files with no real structure. Engineers were spending real time hunting for basics that should have been easy to find.
Management wanted a SharePoint fallback, so the new platform could not be the only copy of the data. The server also had to stay internal-only, which ruled out a public SaaS wiki.
The result was a searchable BookStack deployment with a consistent shelf, book, chapter, and page hierarchy, plus an export pipeline that pushed a readable backup into SharePoint every night.
Outcome: Hundreds of legacy documents were consolidated into one internal system without losing the management-required fallback path.
Build
How it was built
01
Infrastructure & Hosting
Provisioned a dedicated Ubuntu Server VM on VMware ESXi to host BookStack. The app, database, and reverse proxy were containerized with Docker and Docker Compose, and the server was placed in a DMZ so it stayed accessible to internal users without unnecessary exposure.
Docker
Docker Compose
BookStack
Ubuntu Server
VMware ESXi
DMZ
02
Migration Automation
Wrote custom scripts in Node.js and Python to batch-process hundreds of Word and PDF files, extract structured content, and create the right shelf, book, chapter, and page hierarchy through the BookStack REST API.
Node.js
Python
BookStack REST API
Batch Processing
Document Parsing
03
Business Continuity & Backup
Nightly database dumps to local storage, plus a scheduled pipeline that exported every book as HTML and PDF into SharePoint. Two independent copies of everything.
SharePoint
Automated DB Dumps
Export Pipelines
Scheduled Tasks
04
Rollout & Adoption
Worked with the team to settle on a structure before the migration started, then ran a short walkthrough for non-technical staff and wrote simple usage notes so the handover after launch was easier.
Internal Training
Documentation Taxonomy
Usage Guides
Screenshot
System view
