By ClassName I meant the name of the offending class, not a class named
ClassName. Sorry for the confusion there.
You're hitting a failure in a test that ensures that classes in
sanctioned-geode-core-serializables.txt can be serialized and
deserialized. The serialization filter is objecting
It's possible that I ran build/precheckin with a different version of Java.
Is it possible that would change the bits that
AnalyzeSerializablesJUnitTestBase is looking at and cause an unexpected
failure?
On Thu, Oct 4, 2018 at 1:24 PM, Kirk Lund wrote:
> But I didn't add or touch the class Class
But I didn't add or touch the class ClassName -- according to git log,
Jinmei and Patrick created it in the following commit on 1/29/18 -- I
haven't touched this at all on my branch:
GEODE-3915: use ClassName type for cache-loader, writer and listeners
(#1327)
* GEODE-3915: use ClassName type for
It looks like your region attributes contain an instance of a class that
isn't in sanctioned-geode-core-serializables.txt. It's also possible
that you added the class to that file but it didn't get properly copied
to the output directory, so you might check that too.
Output of this test shoul
Sure is! https://github.com/kirklund/geode/tree/GEODE-2644-Appenders-steps3
My branch has no changes to org.apache.geode.cache.AttributesFactory or its
inner class(es) though. I even double-checked with:
$ git log
./geode-core/src/main/java/org/apache/geode/cache/AttributesFactory.java
It just s
I used to see this issue when I make changes in the serializable class or
its members but don't reflect it in the
sanctioned-geode-core-serializables.txt file.
If I am using a custom object in a test or something I add it
as SERIALIZABLE_OBJECT_FILTER property.
Is your branch hosted in github?
R
