Re: Jeri/SSL export question
Peter Jones <[email protected]> Tue, 31 Mar 2009 15:15:19 -0400
| Newsgroups | gmane.comp.java.sun.jini |
|---|---|
| Message-ID | <20090331191519.GA9986@east> |
On Mon, Mar 30, 2009 at 04:47:34PM -0600, Esmond Pitt wrote: > The client got the stub via a registry lookup. Does the stub have no codebase annotation? That would be another case where BasicJeriTrustVerifier (along with verifiers for subcomponents like SslTrustVerifier) should be able to verify an uncustomized JERI stub without requiring ProxyTrustVerifier, because without a codebase annotation the dynamic proxy class's loader should be a "relatively local" loader trusted by BasicJeriTrustVerifier. > There are no dynamic proxies in the system, only stubs generated by JERI. (A stub resulting from exporting with JERI generally has a dynamic proxy as its outermost object, so I'm not sure you meant to say that.) > So should I not be expecting the ServerProxyTrust method to be > called, is that what it comes down to? Not if the stub can be verified without using ProxyTrustVerifier, like as described above, because BasicJeriTrustVerifier is sufficient. Passing this verification means that the stub can be trusted to implement constraints on remote invocations, like Integrity.YES, ServerAuthentication.YES, and ServerMinPrincipal, so that the client can be sure that invocations on the stub only communicate with the specified principal, regardless of from where or whom the stub was initially obtained. -- Peter -------------------------------------------------------------------------- Getting Started: http://www.jini.org/wiki/Category:Getting_Started Community Web Site: http://jini.org jini-users Archive: http://archives.java.sun.com/archives/jini-users.html Unsubscribing: email "signoff JINI-USERS" to [email protected]