The analyzer has detected an exception class inaccessible to other builds. If such an exception is thrown, the external code will have to catch instances of the nearest accessible parent class such as the base class of all exceptions, 'Exception'. This hinders exception handling since the code of other builds cannot identify the problem precisely.
Lack of clear identification of exceptions poses an additional security risk because some specific exceptions may require specific handling rather than general handling. Insufficient logging and monitoring (including exception identification) are pointed out as a separate category on OWASP Top 10 Application Security Risks 2017: A10:2017-Insufficient Logging & Monitoring.
Consider the following simple example taken from a real project:
internal sealed class ResourceException : Exception
{
internal ResourceException(string? name, Exception? inner = null)
: base(name, inner)
{
}
}
To correctly handle a given exception, the exception class must be declared as 'public':
public sealed class ResourceException : Exception
{
internal ResourceException(string? name, Exception? inner = null)
: base(name, inner)
{
}
}
This will enable other builds to catch and handle this particular exception.
Keep in mind that declaring an exception class as 'public' may be insufficient if it is a nested class, as in the following example:
namespace SomeNS
{
class ContainingClass
{
public class ContainedException : Exception {}
....
}
}
Here, the exception class is nested into 'ContainingClass', which is implicitly declared as 'internal'. Therefore, 'ContainedException' will also be accessible only within the scope of the current build even though it bears the 'public' specifier. The analyzer recognizes situations like that and issues the warning.
This diagnostic is classified as: