DAST browser-based crawler

Introduced in GitLab 13.12.

caution
This product is an early access and is considered a beta feature.

GitLab DAST’s new browser-based crawler is a crawl engine built by GitLab to test Single Page Applications (SPAs) and traditional web applications. Due to the reliance of modern web applications on JavaScript, handling SPAs or applications that are dependent on JavaScript is paramount to ensuring proper coverage of an application for Dynamic Application Security Testing (DAST).

The browser-based crawler works by loading the target application into a specially-instrumented Chromium browser. A snapshot of the page is taken prior to a search to find any actions that a user might perform, such as clicking on a link or filling in a form. For each action found, the crawler will execute it, take a new snapshot and determine what in the page changed from the previous snapshot. Crawling continues by taking more snapshots and finding subsequent actions.

The benefit of crawling by following user actions in a browser is that the crawler can interact with the target application much like a real user would, identifying complex flows that traditional web crawlers don’t understand. This results in better coverage of the website.

Using the browser-based crawler should provide greater coverage for most web applications, compared with the current DAST AJAX crawler. The new crawler replaces the AJAX crawler and is specifically designed to maximize crawl coverage in modern web applications. While both crawlers are currently used in conjunction with the existing DAST scanner, the combination of the browser-based crawler with the current DAST scanner is much more effective at finding and testing every page in an application.

Enable browser-based crawler

The browser-based crawler is an extension to the GitLab DAST product. DAST should be included in the CI/CD configuration and the browser-based crawler enabled using CI/CD variables:

  1. Ensure the DAST prerequisites are met.
  2. Include the DAST CI template.
  3. Set the target website using the DAST_WEBSITE CI/CD variable.
  4. Set the CI/CD variable DAST_BROWSER_SCAN to true.

An example configuration might look like the following:

include:
  - template: DAST.gitlab-ci.yml

dast:
  variables:
    DAST_WEBSITE: "https://example.com"
    DAST_BROWSER_SCAN: "true"

Available CI/CD variables

The browser-based crawler can be configured using CI/CD variables.

CI/CD variable Type Example Description
DAST_WEBSITE URL http://www.site.com The URL of the website to scan.
DAST_BROWSER_SCAN boolean true Configures DAST to use the browser-based crawler engine.
DAST_BROWSER_ALLOWED_HOSTS List of strings site.com,another.com Hostnames included in this variable are considered in scope when crawled. By default the DAST_WEBSITE hostname is included in the allowed hosts list.
DAST_BROWSER_EXCLUDED_HOSTS List of strings site.com,another.com Hostnames included in this variable are considered excluded and connections are forcibly dropped.
DAST_BROWSER_IGNORED_HOSTS List of strings site.com,another.com Hostnames included in this variable are accessed but not reported against.
DAST_BROWSER_MAX_ACTIONS number 10000 The maximum number of actions that the crawler performs. For example, clicking a link, or filling a form.
DAST_BROWSER_MAX_DEPTH number 10 The maximum number of chained actions that the crawler takes. For example, Click -> Form Fill -> Click is a depth of three.
DAST_BROWSER_NUMBER_OF_BROWSERS number 3 The maximum number of concurrent browser instances to use. For shared runners on GitLab.com we recommended a maximum of three. Private runners with more resources may benefit from a higher number, but will likely produce little benefit after five to seven instances.
DAST_BROWSER_COOKIES dictionary abtesting_group:3,region:locked A cookie name and value to be added to every request.
DAST_BROWSER_LOG List of strings brows:debug,auth:debug A list of modules and their intended log level.
DAST_BROWSER_NAVIGATION_TIMEOUT Duration string 15s The maximum amount of time to wait for a browser to navigate from one page to another
DAST_BROWSER_ACTION_TIMEOUT Duration string 7s The maximum amount of time to wait for a browser to complete an action
DAST_BROWSER_STABILITY_TIMEOUT Duration string 7s The maximum amount of time to wait for a browser to consider a page loaded and ready for analysis
DAST_BROWSER_NAVIGATION_STABILITY_TIMEOUT Duration string 7s The maximum amount of time to wait for a browser to consider a page loaded and ready for analysis after a navigation completes
DAST_BROWSER_ACTION_STABILITY_TIMEOUT Duration string 800ms The maximum amount of time to wait for a browser to consider a page loaded and ready for analysis after completing an action
DAST_BROWSER_SEARCH_ELEMENT_TIMEOUT Duration string 3s The maximum amount of time to allow the browser to search for new elements or navigations
DAST_BROWSER_EXTRACT_ELEMENT_TIMEOUT Duration string 5s The maximum amount of time to allow the browser to extract newly found elements or navigations
DAST_BROWSER_ELEMENT_TIMEOUT Duration string 600ms The maximum amount of time to wait for an element before determining it is ready for analysis

The DAST variables SECURE_ANALYZERS_PREFIX, DAST_FULL_SCAN_ENABLED, DAST_AUTO_UPDATE_ADDONS, DAST_EXCLUDE_RULES, DAST_REQUEST_HEADERS, DAST_HTML_REPORT, DAST_MARKDOWN_REPORT, DAST_XML_REPORT, DAST_AUTH_URL, DAST_USERNAME, DAST_PASSWORD, DAST_USERNAME_FIELD, DAST_PASSWORD_FIELD, DAST_FIRST_SUBMIT_FIELD, DAST_SUBMIT_FIELD, DAST_EXCLUDE_URLS, DAST_AUTH_VERIFICATION_URL, DAST_BROWSER_AUTH_VERIFICATION_SELECTOR, DAST_BROWSER_AUTH_VERIFICATION_LOGIN_FORM, DAST_BROWSER_AUTH_REPORT, DAST_INCLUDE_ALPHA_VULNERABILITIES, DAST_PATHS_FILE, DAST_PATHS, DAST_ZAP_CLI_OPTIONS, and DAST_ZAP_LOG_CONFIGURATION are also compatible with browser-based crawler scans.

Vulnerability detection

While the browser-based crawler crawls modern web applications efficiently, vulnerability detection is still managed by the standard DAST/Zed Attack Proxy (ZAP) solution.

The crawler runs the target website in a browser with DAST/ZAP configured as the proxy server. This ensures that all requests and responses made by the browser are passively scanned by DAST/ZAP. When running a full scan, active vulnerability checks executed by DAST/ZAP do not use a browser. This difference in how vulnerabilities are checked can cause issues that require certain features of the target website to be disabled to ensure the scan works as intended.

For example, for a target website that contains forms with Anti-CSRF tokens, a passive scan will scan as intended because the browser displays pages/forms as if a user is viewing the page. However, active vulnerability checks run in a full scan will not be able to submit forms containing Anti-CSRF tokens. In such cases we recommend you disable Anti-CSRF tokens when running a full scan.

Managing scan time

It is expected that running the browser-based crawler will result in better coverage for many web applications, when compared to the normal GitLab DAST solution. This can come at a cost of increased scan time.

You can manage the trade-off between coverage and scan time with the following measures:

  • Limit the number of actions executed by the browser with the variable DAST_BROWSER_MAX_ACTIONS. The default is 10,000.
  • Limit the page depth that the browser-based crawler will check coverage on with the variable DAST_BROWSER_MAX_DEPTH. The crawler uses a breadth-first search strategy, so pages with smaller depth are crawled first. The default is 10.
  • Vertically scaling the runner and using a higher number of browsers with variable DAST_BROWSER_NUMBER_OF_BROWSERS. The default is 3.

Timeouts

Due to poor network conditions or heavy application load, the default timeouts may not be applicable to your application.

Browser-based scans offer the ability to adjust various timeouts to ensure it continues smoothly as it transitions from one page to the next. These values are configured using a Duration string which allow you to configure durations with a prefix: m for minutes, s for seconds, and ms for milliseconds.

Navigations, or the act of loading a new page, usually require the most amount of time as they are loading multiple new resources such as JavaScript or CSS files. Depending on the size of these resources, or the speed at which they are returned, the default DAST_BROWSER_NAVIGATION_TIMEOUT may not be sufficient.

Stability timeouts, such as those configurable with DAST_BROWSER_NAVIGATION_STABILITY_TIMEOUT, DAST_BROWSER_STABILITY_TIMEOUT, and DAST_BROWSER_ACTION_STABILITY_TIMEOUT can also be configured. Stability timeouts determine when browser-based scans consider a page fully loaded. Browser-based scans consider a page loaded when:

  1. The DOMContentLoaded event has fired.
  2. There are no open or outstanding requests that are deemed important, such as JavaScript and CSS. Media files are usually deemed unimportant.
  3. Depending on whether the browser executed a navigation, was forcibly transitioned, or action:

    • There are no new Document Object Model (DOM) modification events after the DAST_BROWSER_NAVIGATION_STABILITY_TIMEOUT, DAST_BROWSER_STABILITY_TIMEOUT or DAST_BROWSER_ACTION_STABILITY_TIMEOUT durations

After these events have occurred, browser-based scans consider the page loaded and ready and attempt the next action.

If your application experiences latency or returns many navigation failures, consider adjusting the timeout values such in this example:

include:
  - template: DAST.gitlab-ci.yml

dast:
  variables:
    DAST_WEBSITE: "https://my.site.com"
    DAST_BROWSER_NAVIGATION_TIMEOUT: "25s"
    DAST_BROWSER_ACTION_TIMEOUT: "10s"
    DAST_BROWSER_STABILITY_TIMEOUT: "15s"
    DAST_BROWSER_NAVIGATION_STABILITY_TIMEOUT: "15s"
    DAST_BROWSER_ACTION_TIMEOUT: "10s"
    DAST_BROWSER_ACTION_STABILITY_TIMEOUT: "3s"
note
Adjusting these values may impact scan time as they adjust how long each browser waits for various activities to complete.

Debugging scans using logging

Logging can be used to help you troubleshoot a scan.

The CI/CD variable DAST_BROWSER_LOG configures the logging level for particular modules of the crawler. Each module represents a component of the browser-based crawler and is separated so that debug logs can be configured just for the area of the crawler that requires further inspection. For more details, see Crawler modules.

For example, the following job definition enables the browsing module and the authentication module to be logged in debug-mode:

include:
  - template: DAST.gitlab-ci.yml

dast:
  variables:
    DAST_WEBSITE: "https://my.site.com"
    DAST_BROWSER_SCAN: "true"
    DAST_BROWSER_LOG: "brows:debug,auth:debug"

Log message format

Log messages have the format [time] [log level] [log module] [message] [additional properties]. For example, the following log entry has level INFO, is part of the CRAWL log module, and has the message Crawled path.

2021-04-21T00:34:04.000 INF CRAWL Crawled path nav_id=0cc7fd path="LoadURL [https://my.site.com:8090]"

Crawler modules

The modules that can be configured for logging are as follows:

Log module Component overview
AUTH Used for creating an authenticated scan.
BROWS Used for querying the state/page of the browser.
BPOOL The set of browsers that are leased out for crawling.
CRAWL Used for the core crawler algorithm.
DATAB Used for persisting data to the internal database.
LEASE Used to create browsers to add them to the browser pool.
MAIN Used for the flow of the main event loop of the crawler.
NAVDB Used for persistence mechanisms to store navigation entries.
REPT Used for generating reports.
STAT Used for general statistics while running the scan.