There can be quite a variety of reasons that an existing project may not retest (either on an automated schedule, or when you click the restest button). Occasionally these also affect PRs or PR Checks.
If you are finding this happening, please review the following common reasons
- The project has been deactivated in Snyk UI. See project settings for more details.
- For a Project that was imported via SCM, the source file has been moved, renamed, or deleted. See Manifest not found and What happens when a project is deleted for more details.
- .NET framework has been changed (each framework is counted as a different project). See Snyk for .NET and Changing .NET Target Framework for more details.
- If a .NET file no longer references an older framework, the older framework's project will be treated the same as one where the source file has been moved/renamed/deleted. You can choose to keep it in the UI for history, deactivate it, or delete it.
- The PR introduces unsupported changes, such as a change to a version which is either no longer or not yet supported in SCM scanning.
- For example, Python 3.8, or NPM9/workspaces/v3 lockfiles
- For projects requiring in-sync lockfiles, the lock file is out of sync or missing. Try regenerating the lockfile.
- The webhook for project has been removed from the SCM, or could not be created. See here for more details.
- Access Token has expired or credentials are not valid. Please check your tokens to ensure that they have not expired, and the scope is required. See here for more details on setting up SCM integrations.
- SAML/SSO authorisation has been introduced on SCM orgs and Snyk integration has not been authorised for access. Please check your SCM integration settings
- Broker Client is not running or there is a connection issue between Broker Client and Broker Server. See Broker Troubleshooting
- Broker cannot reach your SCM reliably, or runs into rate limit issues. The first step here is to ensure that your Broker client is up to date to take advantage of any recent fixes, and to ensure that you have only one version running, for best stability.
- Kubernetes integrations: In order to maintain the health of the database, any information that relates to a workload that has not been changed or updated for eight (8) days will be removed. This can lead to failure on retesting the workload. (see here)
- Kubernetes integrations: Snyk-monitor has run out of memory or disk space (see pre-requisites). Please check the monitor logs.
- Kubernetes integrations: if your snyk-monitor is still running v1.x, this version has been deprecated and your snyk-monitor must be upgraded to v2.x. Please see this blog post and the installation/upgrade instructions here. You can check the version of Monitor that you are running by navigating to the org settings for your kubernetes integration.
If these do not appear to relate to the issue you're seeing, or you need further assistance with identifying the cause or solution, please reach out to support@snyk.io with the details of the org and project/s involved.