Story #3474
Updated by Skye Roseboom over 11 years ago
Story The current merge policy in our Hazelcast clusters will add all entries to capture work needed ensure data consistency the cluster that creates a superset of entries found across CN datastores.
Due to partition tolerance of members. This may cause conflicts, especially when the distributed cache (hazelcast), it is possible same object has different content (like a system metadata record with differing replica entries and statuses). Create a new merge policy class that compares content for the storage datastore to become inconsistent across CN when create/update/deletes are performed entries in one cluster partition hzSystemMetadata, and not synchronized re-sync entries across nodes, prioritizing entries where the cluster when dateSysMetadataModified date is the partitions in newest and serialVersion is the cluster are resolved/joined.
highest.
Due to partition tolerance of members. This may cause conflicts, especially when the distributed cache (hazelcast), it is possible same object has different content (like a system metadata record with differing replica entries and statuses). Create a new merge policy class that compares content for the storage datastore to become inconsistent across CN when create/update/deletes are performed entries in one cluster partition hzSystemMetadata, and not synchronized re-sync entries across nodes, prioritizing entries where the cluster when dateSysMetadataModified date is the partitions in newest and serialVersion is the cluster are resolved/joined.
highest.
