In my view we should go ahead and fix it. How can we break something that never worked?
On Thu, Jan 28, 2016 at 11:00 AM, Valentin Kulichenko < valentin.kuliche...@gmail.com> wrote: > Yes, this happens because in a single JVM the dynamic proxy class is > available for Class.forName, but for multi-jvm case this is not true. We > should additionally write information about the implemented interfaces and > manually recreate the proxy during unmarshalling. But the problem is that > it's not a compatible change. In binary marshaller there is a protocol > version, so should be OK, but we don't have it for optimized marshaller, > right? > > -Val > > On Thu, Jan 28, 2016 at 12:21 AM, Alexey Goncharuk < > alexey.goncha...@gmail.com> wrote: > > > This is correct, I took the original test that existed for Optimized > > marshaller and copied it for Binary marshaller. Was not aware of > multi-jvm > > specifics. Just ran the provided example with Optimized marshaller - it > > does not work either. > > > > 2016-01-28 11:08 GMT+03:00 Denis Magda <dma...@gridgain.com>: > > > > > To my knowledge Alex G. was taking care of this initial issue. > > > > > > This particular one is reproduced only in multi JVM mode. > > > > > > -- > > > Denis > > > > > > > > > On 1/28/2016 2:59 AM, Dmitriy Setrakyan wrote: > > > > > >> Who was originally responsible for fixing the Proxy serialization > issue? > > >> > > >> On Wed, Jan 27, 2016 at 4:07 AM, Denis Magda <dma...@gridgain.com> > > wrote: > > >> > > >> Igniters, > > >>> > > >>> A end user reported on the issue related to proxy > > >>> serialization/deserialization > > >>> https://issues.apache.org/jira/browse/IGNITE-2450 > > >>> > > >>> Could someone experienced in marshalling take a look at this? Seems > > that > > >>> the original proxy related issue wasn't fully fixed. > > >>> > > >>> -- > > >>> Denis > > >>> > > >>> > > >>> > > >>> > > > > > >
