[ http://issues.apache.org/jira/browse/GERONIMO-1563?page=all ]
David Jencks updated GERONIMO-1563:
-----------------------------------
Attachment: GERONIMO-1563-step2.1-v4.diff
GERONIMO-1563-step2.1-v4-openejb.diff
Version that uses substitution groups in the schemas and treats them like anys
in the processing code to get around limitations on xmlbeans substitution group
handling.
There's a remaining problem with xmlbeans validation that I have temporarily
worked around by disabling validation. I'm trying to get more info on what to
do from the xmlbeans team.
I'd like this version to get reviewed with an eye towards committing it.
> Make the JACC implementation pluggable
> --------------------------------------
>
> Key: GERONIMO-1563
> URL: http://issues.apache.org/jira/browse/GERONIMO-1563
> Project: Geronimo
> Type: Improvement
> Security: public(Regular issues)
> Components: security
> Versions: 1.2
> Reporter: David Jencks
> Assignee: David Jencks
> Attachments: GERONIMO-1563-step2.1-v1-openejb.diff,
> GERONIMO-1563-step2.1-v1.diff, GERONIMO-1563-step2.1-v2-openejb.diff,
> GERONIMO-1563-step2.1-v2.diff, GERONIMO-1563-step2.1-v4-openejb.diff,
> GERONIMO-1563-step2.1-v4.diff
>
> Currently we are hardcoded into using our JACC implementation. This means we
> can't use third party authorization/security servers such as Tivoli AM.
> The runtime hardcoding is that the installation of the spec permissions into
> the policy configuration is mixed in with pushing our proprietary
> principal-role mapping into the policy configuration.
> The build time hardcoding is that the only proprietary security configuration
> we accept is our own xml for principal-role mapping, and we insist on it
> being present.
> Some steps for this:
> 1. make separate gbeans for the spec and proprietary access to the policy
> configuration. These should be connected by an interface, and the spec gbean
> should control the proprietary gbean and pass it the contextIds in the
> current application.
> 2. The security builder should be partly namespace driven, with the
> proprietary xml interpretation driven by the namespace.
> 2.a the base security builder should construct the
> ApplicationPolicyConfigurationGBean and hand off to the namespace-selected
> gbean for the proprietary stuff.
> 2.b the proprietary-xml builder should install the "role-mapper" gbean with
> the info needed for e.g. principal-role mapping.
> When we're done with this we should be able to support e.g. IBM pluggable
> JACC implementations that support their role-mapping capabilities by just
> writing an xml format and a gbean that pushes role mapping info into their
> interfaces. The ibm interfaces are explained here:
> http://publib.boulder.ibm.com/infocenter/wasinfo/v6r0/topic/com.ibm.websphere.express.doc/info/exp/ae/rsec_jaccspis.html
> If anyone knows how other app servers configure the non-spec part of JACC
> references would be very much appreciated.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
