Infrastructure Documentation
A network only the founder understands is an operational risk. A new technician takes weeks to get up to speed; a 3 AM incident turns into chaos; a growing ISP gets acquired and the buyer asks for documentation that does not exist.
What infrastructure documentation is
Network documentation covers three layers: IPAM (IP addressing and VLAN management), physical/logical inventory (equipment, cables, topology), and operational procedures (runbooks, maintenance checklists, contingency plans). Documentation that lives in the founder's Visio file does not count — it must be accessible and kept up to date.
What RASYS does
- phpIPAM as source of truth — prefixes, VLANs, devices, RIRs, ASN, locations.
- Structured IPAM — prefix hierarchy, per-service IP pools, traceable allocations.
- Topology diagrams with Draw.io (exportable/editable) or Mermaid (Git-versionable).
- Operation runbooks — equipment replacement, manual failover, step-by-step troubleshooting.
- Maintenance checklists — planned windows with validation steps before and after.
- NOC procedures — escalation path, who calls whom, when to page on-call, how to open a ticket with the carrier.
- Documented DR plan — defined RTO/RPO, covered scenarios, documented restoration runbook.
- Existing infrastructure survey when there is no baseline documentation — we interview the team, inspect remotely, and build the baseline.
Tools we work with
phpIPAM (IPAM and inventory), Draw.io, Mermaid, Markdown in Git (versionable documentation), Confluence/Notion when the customer already uses them.
When it makes sense to talk to us
No team member can onboard a new technician in under a week; a recent incident exposed the lack of a runbook; opening a new PoP or expanding and documentation is out of date; a buyer or auditor requested network documentation; you want to reduce key-person dependency.
Talk to us — initial conversation, no commitment. See also: Configuration Backup.
FREQUENTLY ASKED QUESTIONS
How long does the initial survey of an undocumented network take?
It depends on size and complexity. A network with 5-10 edge devices and 2-3 PoPs: 2-4 weeks to reach functional documentation. Larger networks or those with lots of legacy without any records: we assess after an initial remote inspection access. We start with what impacts operations most and cover the rest in iterations.
Is phpIPAM mandatory or do you work with other tools?
phpIPAM is what we use for IPAM — it is open source, has an API, and is straightforward to operate for an ISP team. For physical/logical inventory (equipment, racks, cables, locations), we work on top of markdown versioned in Git. If the customer already has NetBox, Nautobot, or another working tool, we build on top of what exists. What we avoid is an Excel spreadsheet as the single source of truth — no versioning, no API, and it goes stale within weeks.
Does the documentation stay alive day to day?
Yes. We work on a monthly plan — customers stay with us for years, it's not a one-off project that ends. Documentation is maintained alongside the rest of the operation: every config change generates a ticket that includes the matching update, and Mermaid diagrams live in the same repository as the infrastructure, so any infra change goes through the diagram update. Documentation nobody maintains is pointless.
Do you only document, or do you also fix what you find?
During the survey, we flag everything we find that is inconsistent or problematic. Fixing what we find is a separate scope — we document the findings and you decide what to tackle first. Many customers use the survey as input to prioritize their technical backlog for the following quarter.