◇ OpenCyber Academy
Technologist · Chapter 17 · English

Chapter 17 — Security operations and blue team: tooling

Open, free learning for anyone who wants to protect people and technology. No account is required to read this text or its worked answers.

Chapter 17 — Security operations and blue team: tooling

Lesson map

Lesson 17 published. Map, example, EX049.

flowchart TD
  N0["Learning outcomes"]
  N1["17.1 The toolset"]
  N2["17.2 Fit for size"]
  N3["17.3 Workflow support"]
  N4["17.4 Multi-client separation"]
  N5["17.5 Evaluating tools"]
  N6["17.6 Tool health"]
  N7["17.7 Avoiding tool sprawl"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6 --> N7
  N7 --> EX["Worked example"]
  EX --> Q["Exercise EX049"]

Logical drawing

flowchart LR
  N0["Learning outcomes"]
  N1["17.1 The toolset"]
  N2["17.2 Fit for size"]
  N3["17.3 Workflow support"]
  N4["17.4 Multi-client separation"]
  N5["17.5 Evaluating tools"]
  N6["17.6 Tool health"]
  N7["17.7 Avoiding tool sprawl"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6 --> N7 --> Q["EX049"]
[Learning outcomes]
    |
    v
[17.1 The toolset]
    |
    v
[17.2 Fit for size]
    |
    v
[17.3 Workflow support]
    |
    v
[17.4 Multi-client separation]
    |
    v
[17.5 Evaluating tools]
    |
    v
[17.6 Tool health]
    |
    v
[17.7 Avoiding tool sprawl]
    |
    v
[Worked example] --> [Exercise EX049]

Learning outcomes

You will (1) describe the SOC's main tools and their roles; (2) choose tools appropriate to a small managed SOC; (3) configure tools to support analyst workflow; (4) evaluate tools by outcomes rather than features.

17.1 The toolset

Copperleaf uses a SIEM for collection, search and rules; endpoint agents for telemetry and isolation; a case-management system for alerts and notes; and a shared knowledge base of rule pages and client profiles.

Each tool answers a different need on an incident night. The SIEM found the support2 account, the agent isolated CLINIC2-SRV, the case system recorded who approved what and when, and the knowledge base held the rule page and BrightSmile's profile with its contacts. Losing any one of them would have slowed the response at a specific step.

17.2 Fit for size

A small SOC needs tools that are easy to maintain, support many clients and integrate well. Feature-heavy tools that need a dedicated engineer can be a burden.

Copperleaf is a small provider covering about forty clients, so every hour spent tuning a tool is an hour not spent on alerts. A tool that needs constant attention from a specialist engineer would leave the analysts short-handed; one that can be maintained by the existing team fits the size of the service.

17.3 Workflow support

Knerler, Parker and Zimmerman (2022) advise leveraging tools to support analyst workflow. At Copperleaf, alerts open in the case system with enrichment and links to the client profile, so analysts rarely switch screens.

On the BrightSmile night, the alert opened with the enriched host line and links to the client profile and contacts, so Mei did not have to search three systems before phoning Raj. Fewer screens means fewer chances to copy a wrong host name or call the wrong client during the first critical minutes.

17.4 Multi-client separation

Tools must keep clients' data separate and show which client each alert belongs to. Mistakes here could leak one client's data to another.

Mei's search for support2 in chapter 4 shows the risk: two clients had accounts with the same name. If the tools had not carried the client field, she might have isolated a law firm's server or reported the wrong client's activity to BrightSmile. Client separation is therefore checked on every new search view and report template. Access rights follow the same principle, so an analyst who views one client's alert cannot accidentally open another client's raw logs.

17.5 Evaluating tools

Copperleaf judges tools by outcomes: time to triage, missed detections, effort to maintain. A trial of a new endpoint agent measures these on three clients before any decision.

Using the same three measures for every trial lets Raj compare a new tool with the current one on Copperleaf's own work rather than on a vendor's list of features. Three clients is a small sample, so the trial results guide the decision without being treated as proof that the tool will behave the same on every client.

17.6 Tool health

Tools themselves need monitoring: agent versions, SIEM storage and rule run times.

An old agent version might stop sending a new event type, full SIEM storage can drop events, and a rule that takes too long may not finish before the next run starts. Each failure removes detection quietly in the same way as a broken log feed, so these measures go to engineering alongside the data-gap alerts. Mei's onboarding check of BrightSmile's agents includes their version, so an outdated agent on a clinic PC is listed and updated.

17.7 Avoiding tool sprawl

Each new tool needs an owner and a reason; overlapping tools are retired.

An owner is the person who updates the tool, watches its health and decides when it is no longer needed. Without one, a tool keeps running with old settings and nobody notices when it stops working. Two tools doing the same job also double the alerts analysts must reconcile without finding anything new. When the new endpoint agent is adopted for servers, Raj names its owner and records which old agent features it replaces on those hosts.

Access to tools

Analyst accounts on the SIEM and case system use MFA and are reviewed quarterly.

Worked example

In the endpoint agent trial, the new agent detected the same test events but reduced analyst triage time by two minutes per alert because it showed process trees directly in the alert. It also used more memory on old clinic PCs. Raj decides to adopt it only for servers and newer PCs.

Synthetic case. Tools and trial results are invented.

Errors and limits

Vendor demonstrations show best cases; trial tools in your own environment.

Exercises

Synthesis

Choose and configure tools for workflow and outcomes, with client separation and limited sprawl.

Continue studying — chapter 17 of 30

← Chapter 16 (draft) · Course chapter index · Chapter 18 (draft) →