Jenkins is being a pain

[Cantor, Scott]

On 2/26/13 3:08 PM, "Daniel Fisher" <dfisher at vt.edu> wrote:
>
>On Tue, Feb 26, 2013 at 1:58 PM, Cantor, Scott <cantor.2 at osu.edu> wrote:
>
>- does the inclusion of Bouncy Castle in the classpath make it the JCE at
>this point? And do we want that? (I think we don't.)
>
>Just including the jar won't make it the default crypto provider. That
>has to be done by code or JRE configuration. (Unless something has
>changed in Java7...)

That's what I thought, but this failing test's stack trace certainly seems
weird to me:

https://build.shibboleth.net/jenkins/job/java-opensaml-multi-jdk/40/jdk=JDK
-1.7.0/console

testErrorInvalidKeyDecryptionKey(org.opensaml.xmlsec.encryption.support.Sim
pleDecryptionTest)  Time elapsed: 0.04 sec  <<< FAILURE!
org.bouncycastle.crypto.DataLengthException: input too large for RSA
cipher.
	at org.bouncycastle.crypto.engines.RSACoreEngine.convertInput(Unknown
Source)
	at org.bouncycastle.crypto.engines.RSABlindedEngine.processBlock(Unknown
Source)
	at org.bouncycastle.crypto.encodings.OAEPEncoding.decodeBlock(Unknown
Source)
	at org.bouncycastle.crypto.encodings.OAEPEncoding.processBlock(Unknown
Source)
	at org.bouncycastle.jce.provider.JCERSACipher.engineDoFinal(Unknown
Source)
	at org.bouncycastle.jce.provider.WrapCipherSpi.engineUnwrap(Unknown
Source)
	at javax.crypto.Cipher.unwrap(Cipher.java:2466)
	at 
org.apache.xml.security.encryption.XMLCipher.decryptKey(XMLCipher.java:1474
)
	at 
org.opensaml.xmlsec.encryption.support.Decrypter.decryptKey(Decrypter.java:
714)


It's no longer failing, but that was after I slipped the unlimited
strength jars in. But I also don't think this would be fixed by those
jars, which is even more confusing/annoying.

-- Scott