So I wrote a module to do some PGP stuff using BouncyCastle.
Or, more accurately, I used old code we had that was already doing the job
(for like 10 years on Java1.3!) and put it into a module. This module takes
passwords that were encrypted years ago and stored in the DB and unencrypts
them (let's not discuss the best-practices of that, it's a healthy debate in
my own community already; but the peeps want what the peeps want LOL).
This is to say, I have a very old keystore that I need to unencrypt some
very old passwords.
On my local machine I have OSX running Java 1.7. My new module deployed to
Karaf 3 stellar.
On my TEST and PRODUCTION boxes I have Redhat running Java 1.6 but with
Karaf 3 running on Java 1.7.
The new module would not deploy on either of those. Kept throwing a
"UnrecoverableKeyException: no match".
A lot of tinkering and digging got me into thinking that my local version of
Java (build 8) vs the version on my TEST/PROD boxes (build 9) was the
culprit; but rather than go through the process of getting build 8 on those
boxes I first tried pointing Karaf 3 at the system's Java 1.6 -- to my
surprise, Karaf 3 launched just fine ("ummm... wow, Karaf claims Java 7 is
required... WTH?").
Also, to my relief, my module deployed perfectly.
So then I went through the exercise of getting build 8 on the TEST/PROD
boxes (thinking something about Karaf 3 might still break later and I
wouldn't know why so I better get Java 7 on here now)... however, build 8
did not solve the problem; still no deploy!
So, I have 1 problem and 1 question:
*Problem:* my guess is that Java7-Linux vs Java7-OSX is the problem with
getting BouncyCastle to work. Anyone else aware of whether or not there's
an issue there? (yes, I realize this isn't the BC-forum, I'll be asking
there as well; just fishing)
*Question:* why does Karaf 3 claim it needs Java 7 when that doesn't *seem*
to be true? Am I at risk of some feature not working, or of a future break?
Also posted on SO:
http://stackoverflow.com/questions/25265799/java7-bouncycastle-osx-vs-redhat-unrecoverablekeyexception
Any thoughts?
--
View this message in context:
http://karaf.922171.n3.nabble.com/Karaf3-BouncyCastle-and-Java7-vs-Java6-tp4034727.html
Sent from the Karaf - User mailing list archive at Nabble.com.