Your First Greenbone/OpenVAS Scan
Your first Greenbone/OpenVAS scan should answer a small, well-defined question about a system you are permitted to assess. The goal is not to produce the longest possible vulnerability report. It is to learn how a scan target is selected, how the scanner decides which checks to run, how results become evidence and what needs human verification before anyone makes a security decision.
The naming can be confusing at first. OpenVAS is the scanning engine associated with Greenbone’s vulnerability management stack. Greenbone Community Edition includes additional components for managing vulnerability data, scans, tasks and reports. A learner often interacts with those components through the Greenbone Security Assistant web interface, rather than typing every scanner command directly. An installed scanner without healthy feeds or a working manager is not a complete, ready-to-use assessment environment.
This walkthrough explains the sequence using one lab machine or another explicitly authorized host. It avoids scanning third-party systems, exploit validation or production-wide discovery. Menu labels can vary by Greenbone release and distribution; the current official Greenbone documentation should take precedence over screenshots from older OpenVAS tutorials.
Before configuring software, establish who owns the target and what the scan may touch. For a first exercise, choose one virtual machine in a lab that you control, ideally with a straightforward set of services whose versions and configuration you can inspect. Record the target address, intended test window, business impact if it becomes temporarily slow and how you will stop a scan if necessary. A private address is not proof of permission: a shared corporate network may contain other people’s equipment that has not been authorized for assessment.
Write a short statement of purpose, such as identifying whether the lab host exposes any obsolete services or missing security updates. Avoid selecting broad ranges because they are easy to type. Treat every discovered host as a potentially distinct owner, especially in cloud networks and shared facilities. If the system is production or safety-critical, arrange a maintenance window and follow the organization’s formal change process instead of copying settings from a classroom demonstration.
Decide what you already know about the target. Its operating system and applications, installed security patches, open listening ports and owner documentation make the results easier to interpret. You will need these facts later to distinguish a reliable vulnerability finding from a banner guess. This early asset work is part of a credible vulnerability assessment, not administrative busywork.
The Greenbone Community Edition is a group of related services. The scanner executes vulnerability tests against the systems in scope. The Greenbone manager maintains objects such as targets, tasks, scan configuration and reports. Its web assistant presents workflows to the operator. Greenbone’s published architecture also describes the role of update feeds that supply vulnerability tests and associated information; if those feeds have not finished loading, scan coverage and results may be incomplete.
This is why the first quality check is operational rather than security-related. Confirm the installation is a supported, stable release for the chosen platform. Review the current official installation instructions for your distribution or the Community Containers installation. Third-party operating-system packages may follow a different upgrade cadence from the upstream Greenbone stack. Never assume that a command from an old blog post still produces a compatible setup.
On a Kali-based training host, Greenbone’s official documentation describes verifying the installation and feed health before accessing the local web assistant. For container-based deployments, the published compose configuration includes several services, persistent volumes and separate feed-related components. Whichever route is chosen, use the supported guide for the actual version you installed. Do not paste an unknown setup script into a production machine or expose an administrative web UI beyond its intended access boundary.
Check that the manager, scanner and web assistant are healthy; that the vulnerability test feed has loaded; that storage and system time are sane; and that you have a properly secured administrative account. A scan that starts with stale or missing test data may look successful while silently missing important checks. Capture the version and feed status in your lab notes so any later comparison uses the same baseline.
In the Greenbone Security Assistant interface, create a target with a descriptive name, such as the approved lab virtual machine, and enter only its authorized address. Greenbone’s workflow separates target definitions from tasks, allowing the same narrowly defined asset to be assessed more than once with different permitted scan configurations. That distinction is useful for an audit trail: the scope should remain clear even if the testing approach changes.
Verify the address carefully before saving it. A typo could redirect the assessment to an unrelated device. Decide whether host discovery and port selection match the target. If a firewall suppresses ping, a missing-host result is not necessarily evidence that the target is absent; it may be a discovery configuration issue. Rather than broadening to every address nearby, investigate the approved host and its network behavior.
Some tasks may use login credentials to inspect the operating system from inside. A credentialed assessment can improve package and configuration visibility, but the test account should be specifically authorized, limited to necessary access and protected throughout the exercise. Beginners can start with an unauthenticated lab scan to understand visible services, then compare it with a separately approved authenticated assessment. Record the limits of each rather than suggesting that one mode detects every vulnerability.
If the interface offers target exclusions, treat them as part of the written scope, not a convenience to improvise later. An asset owner may permit one host but not a neighboring database, embedded controller or printer. Clear inclusion and exclusion rules reduce the chance of a technically successful but unauthorized scan.
Greenbone provides scan configurations that determine the vulnerability tests used and their behavior. Its documentation describes choices including a broad Full and fast configuration, but selecting the broadest option automatically is not responsible assessment design. Choose a configuration appropriate to what you are trying to learn, to the approved target and to the environment’s tolerance for traffic and testing.
Configuration also governs concurrency: how many hosts and vulnerability tests run at the same time. Higher concurrency may shorten a scan, but it can consume bandwidth and processing resources or degrade a fragile system. For a first lab experiment, one controlled host and conservative, known-safe settings allow you to observe progress and troubleshoot configuration without introducing unnecessary load. Do not enable aggressive or disruptive checks on systems whose owners have not specifically approved them.
The scanner’s output is also shaped by the vulnerability feed and by the available evidence. A test may infer a version from a banner, query a service directly or rely on local package information when credentials are supplied. These methods have different levels of confidence. The user interface may summarize them with a severity score, but that score does not turn an inference into a verified exploit path.
Document the chosen scan policy and why it was selected. If you later compare results after changing a configuration, account for that change; a lower issue count can mean you performed fewer checks, not that the host improved. Reliable trend reporting requires comparable scopes, feeds and scan methods.
Once the target and scan configuration are ready, create a task in the scanning interface. In Greenbone’s published operating workflow, Tasks is where an operator connects a named task to an existing target and scan configuration. Give the task a purpose-based label that makes sense to another analyst, such as a baseline assessment of the authorized lab host, and preserve the configuration details in its description or your assessment notes.
Review the target and selected scan configuration one more time before starting. Confirm that no unrelated ranges or old broad targets were carried forward. Begin the task, then monitor its status rather than leaving it running unattended and assuming that the presence of a progress indicator means everything is healthy.
A task may wait while the manager schedules scanning resources, spend time detecting available hosts, or report errors because a scanner service or feed is unavailable. If the job appears stuck, inspect the relevant service status and task error details. Do not immediately increase concurrency or widen the target scope. Problems at this stage are often installation, network reachability, configuration or feed issues that should be resolved deliberately.
The interface may show intermediate findings while a scan is running. These are not necessarily the final report. Wait for the completed status or document explicitly that the scan ended early. If the target becomes unstable, stop the task and notify the asset owner according to the agreed plan. A responsible scan is one that stays within scope and can be explained, including when it does not complete normally.
The completed report should be read in several passes. First, verify what was actually scanned: the intended host, its reachable services, task configuration, scan time and whether the checks completed. A report produced against the wrong address or with a failed discovery stage is not a valid assessment of the intended system, however detailed the vulnerability list may appear.
Second, inspect findings that matter for your stated purpose. Look for the affected port or component, the vulnerability test used, the observed evidence, references and the scanner’s confidence or quality-of-detection indication where provided. Understand the difference between a verified package issue and a result inferred from a service version. A CVE identifier describes a reported weakness, not proof that every asset displaying a related version string can be compromised in the same way.
Third, compare the finding with independent asset data. Check a vendor advisory, installed package build, configuration or patch status using your authorized access. Operating-system vendors sometimes backport a security fix without changing the banner in the way a simple network check expects. Conversely, a service may be exposed through another path that the scan did not test. Both possibilities demonstrate why automated output requires human review.
Severity is a prioritization aid, not a business-risk determination. Consider whether the vulnerable component is reachable, which users and data it serves, whether compensating controls are present and what will break if a patch is applied. Explain uncertainty rather than exaggerating confidence. A finding that is false positive or no longer applicable should be recorded with evidence; it should not simply disappear without a rationale.
The useful outcome of the exercise is a short, actionable report. Identify the authorized asset, scope, scan date, relevant configuration, major findings, validation method, remediation owner and a sensible next action. Separate immediate corrections from items that need maintenance scheduling, more evidence or risk acceptance. A strong report helps the people responsible for the asset make decisions without reading every scanner detail.
For a lab example, choose one genuine finding, verify it against the software inventory and apply a controlled patch or configuration correction. Run a second approved scan using comparable scope and settings, then explain whether the finding was resolved, reclassified or simply no longer detected. Keep the before-and-after artifacts; they demonstrate a full vulnerability management cycle more effectively than a screenshot with a score.
Do not judge scanner effectiveness by the number of red findings alone. A deliberate baseline may identify no high-risk issues because the test host is patched; that is a useful result when the scan actually covered the planned controls. A very long report may instead reflect noisy, duplicated or weakly supported findings. Be able to explain which results require validation and which are outside the assessment’s scope.
This work also has a delivery dimension. Scanner setup, authorized access, report validation and rechecks consume time that a simple address count does not capture. Understanding vulnerability assessment scoping and effort helps explain why an appropriately controlled professional engagement requires more work than starting a task and exporting a PDF.
At the end of the exercise, you should be able to describe each step in your own words: why the particular target was allowed, which component managed the task, how the feed affected coverage, which scan configuration was chosen, what evidence supported a finding and how a responsible person would verify remediation. That chain of reasoning is more important than memorizing a specific set of menu labels.
Greenbone Community Edition is a capable learning environment when its components are healthy and the scan is tightly scoped. Its official documentation is the appropriate source for changing installation steps and exact menu paths. Keep your practice limited to owned or explicitly authorized systems, record configuration and evidence, and make results useful for decisions. That is the difference between having launched an OpenVAS scan and having carried out a defensible first vulnerability assessment.