Mesa Development

Service

Discovery Implementation

Typical engagement  ·  4–8 weeks

Discovery only helps if what it finds lands in the right class with the right identifier. Most of the work is in the identification and reconciliation rules, not the scan itself — a schedule that runs perfectly and writes into the wrong class has made the CMDB worse, not better.

Teal patch cables plugged into the numbered ports of a network switch.

Discovery answers a narrow question: what is actually here. Everything downstream depends on that answer being right.

What a Discovery run populates A Discovery schedule runs through a MID Server to probe network ranges, producing server and application configuration items in the CMDB. Discovery Schedule MID Server IP Range cmdb_ci_server cmdb_ci_appl CMDB
Discovery Schedule
A schedule decides what gets looked at and how often. Most coverage gaps are a range nobody added, not a probe that failed.
MID Server
The MID Server reaches into the network on ServiceNow's behalf. Where it sits, and which credentials it holds, sets the ceiling on what can be discovered at all.
IP Range
Discovery only finds what it is pointed at, so an accurate estate inventory is a prerequisite — which is why the first deliverable is a coverage report, not a scan.
cmdb_ci_server
The server class. Whether a probe result lands here or in the wrong class is decided by identification rules, not by the probe.
cmdb_ci_appl
The applications running on those servers. This is the layer that makes a CMDB useful to an incident process rather than just an asset list.
CMDB
The destination. A schedule that runs perfectly and writes into the wrong class leaves the CMDB worse than before it ran.

A Discovery schedule runs through a MID Server to probe network ranges, producing server and application configuration items in the CMDB.

  • Discovery Schedule — service
  • MID Server — application
  • IP Range — network
  • cmdb_ci_server — host
  • cmdb_ci_appl — application
  • CMDB — database
  • Discovery Schedule depends on MID Server
  • Discovery Schedule depends on IP Range
  • MID Server depends on cmdb_ci_server
  • MID Server depends on cmdb_ci_appl
  • cmdb_ci_server depends on CMDB
  • cmdb_ci_appl depends on CMDB
What a Discovery run populates

01

Phases and deliverables

Stated before the work starts, so "done" is not a matter of opinion at the end.

  1. 01

    1 week

    Assess

    • Current-state Discovery and CMDB health baseline
    • Credential and MID Server placement plan

  2. 02

    2–4 weeks

    Configure

    • MID Servers deployed and credential store populated
    • Discovery schedules scoped to agreed ranges
    • Identification and reconciliation rules reviewed against sources

  3. 03

    1–2 weeks

    Verify

    • Coverage report against the agreed range inventory
    • Duplicate and unidentified record triage
    • Runbook and handover to your team

02

What gets measured

Baselined at the start and reported at the end. These are the measures themselves, not claims about past results.

  • m01 Percentage of known IP ranges under an active schedule
  • m02 Percentage of discovered CIs landing in the correct class
  • m03 Count of unidentified or duplicate records created per run
Talk about Discovery All services