JBoss 5.0.1 GA
Does anyone have a working configuration ( That is, what changes did you make in jboss-cache-manager-jboss-beans.xml ?? ) to implement the cache for SFSB so that all nodes in a cluster share the same store, as per this documentation:
What I am trying to achieve is:
1) The cache for SFSB to use INVALIDATION_ASYNC instead of REPL_ASYNC
2) But the only way to use INVALIDATION_ASYNC is if each node in the cluster share the same store
... as per the following documentation:
If a cache is configured for invalidation rather than replication, every time data is changed in a cache other caches in the cluster receive a message informing them that their data is now stale and should be evicted from memory. Invalidation, when used with a shared cache loader (see chapter on cache loaders) would cause remote caches to refer to the shared cache loader to retrieve modified data. The benefit of this is twofold: network traffic is minimized as invalidation messages are very small compared to replicating updated data, and also that other caches in the cluster look up modified data in a lazy manner, only when needed.
I am thinking of starting off with FileCacheLoaderConfig, but each node pointing to the same location / directory ( e.g. directory on a SAN, NFS, etc. ) then get to use JDBCCacheLoader ... except that our backend is using Oracle10g, and there is an existing bug in JBossCache shipped with JBoss 5.0.1 GA that throws the exception :
ORA-01461: can bind a LONG value only for insert into a LONG column
I forgot to mention that we have sticky-SFSB, so that the same node ina a cluster is used to serve requests for a SFSB, until such time that the node fails or is shutdown. Thus, when using the same store across nodes in a cluster for the SFSB cache, only one node will write to the store for a specific SFSB session.