weird java error - java.lang.reflect.InaccessibleObjectException?

[n614cd at gmail.com]

Paul,

Is this a custom build or deployment of the IdP?
If yes, and the behavior is inconsistent, like you documented, I would first
check for duplicate java modules or jars. Depending on the order of access
to functions, you can load alternate jars into the classloader.

Tim

-----Original Message-----
From: users <users-bounces at shibboleth.net> On Behalf Of Paul B. Henson via
users
Sent: Wednesday, August 21, 2024 8:40 PM
To: Shib Users <users at shibboleth.net>
Cc: Paul B. Henson <henson at cpp.edu>
Subject: weird java error - java.lang.reflect.InaccessibleObjectException?

I restarted one of my idp's earlier today, which I generally do all of the
time with no problems (other than the expected rush of
NoSuchFlowExecutionException errors right after from those poor people who
had loaded the login page but not yet submitted it or were otherwise in the
middle of a webflow session).

Today though immediately after restarting, I started seeing ELException
errors (which were passed through to clients as a runtime error message on
their web browser), basically saying that it could not call
WarningIterator.hasNext() due to a
java.lang.reflect.InaccessibleObjectException because "module java.base does
not "opens java.util" to unnamed module @4eb45fec". It only seemed to be
affecting some authentications though, in the midst of all of the errors I
did see some successful user logins?

That object is defined in flows/intercept/warning-flow.xml:

    <on-start>
        <evaluate expression="WarningConditionMap.entrySet().iterator()"
result="flowScope.WarningIterator" />
    </on-start>

    <decision-state id="CheckIterator">
        <if test="WarningIterator.hasNext()" then="CheckCondition"
else="proceed" />
    </decision-state>

but I have no idea why all the sudden it couldn't be called? Everything I
could see about this particular error recommends either "fixing your code"
or adding the "--add-opens" jvm argument to work around security
restrictions in the jvm.

Neither of those seems to apply? It's the same code that was always running,
and after looking at it for a bit and having no idea why it was doing that,
I restarted it again and the issue went away, so it was the same code that
was successfully running again.

I guess a solar flare hit a memory DIMM at just the wrong time and screwed
up some internal jvm state while it was starting? Very weird and very
disconcerting not knowing why it happened or if it might happen again out of
the blue at some point...


2024-08-21 16:28:10,012 -
2620:df:8000:4702:0:2:800d:1bab/node0z14eizolsvyr15dii518s9huc11 - ERROR
[net.shibbo leth.idp.profile.interceptor:35] - Uncaught runtime exception
org.springframework.binding.expression.EvaluationException: An ELException
occurred getting the value for expression 'WarningIterator.hasNext()' on
context [class
org.springframework.webflow.engine.impl.RequestControlContextImpl]
        at
org.springframework.binding.expression.spel.SpringELExpression.getValue(Spri
ngELExpression.java:104)
Caused by: org.springframework.expression.spel.SpelEvaluationException:
EL1029E: A problem occurred when trying to execute method 'hasNext' on
object of type 'java.util.LinkedHashMap$LinkedEntryIterator': 'Problem
invoking method: public final boolean
java.util.LinkedHashMap$LinkedHashIterator.hasNext()'
        at
org.springframework.expression.spel.ast.MethodReference.getValueInternal(Met
hodReference.java:159)
Caused by: org.springframework.expression.AccessException: Problem invoking
method: public final boolean
java.util.LinkedHashMap$LinkedHashIterator.hasNext()
        at
org.springframework.expression.spel.support.ReflectiveMethodExecutor.execute
(ReflectiveMethodExecutor.java:146)
Caused by: java.lang.reflect.InaccessibleObjectException: Unable to make
public final boolean java.util.LinkedHashMap$LinkedHashIterator.hasNext()
accessible: module java.base does not "opens java.util" to unnamed module
@4eb45fec
        at
java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(Accessibl
eObject.java:354)

--
Paul B. Henson | Operating Systems and Network Analyst henson at cpp.edu |
California State Polytechnic University Pomona

--
For Consortium Member technical support, see
https://shibboleth.atlassian.net/wiki/x/ZYEpPw
To unsubscribe from this list send an email to
users-unsubscribe at shibboleth.net