| 摘要 |
The present application is directed towards invalidating (also referred to as poisoning) ASDR table entries that are determined to be inaccurate because of changes to a multi-node system. For example, when a node leaves or enters a multi-node system, the ownership of the entries in the ASDR table can change thus invalidating cached and replica entries. More specifically, the system and methods disclosed herein include searching an ASDR table for cached entries responsive to the system determining the multi-node system has changed. After finding a cached entry, the system may determine if the entry should be poisoned. The decision to poison the entry may be responsive to the creation time of the entry, the time when the change to the multi-node system occurred, and in the case of a replica, the owner of the replica's position in a replication chain relative to source of the replica. |
| 主权项 |
1. A method for invalidating a cached entry in an application-state, distributed replication table to track resources of applications across a cluster of nodes, the method comprising:
maintaining, across each node in a cluster of nodes an application-state, distributed replication table comprising entries to track resources of applications across the cluster of nodes, each of the nodes in the cluster of nodes owning one or more partitions of the application-state, distributed replication table; determining, by a first node in the cluster of nodes, that a second node exited the cluster; identifying, by the first node in the cluster of nodes, a plurality of entries assigned to the second node and stored in cache by the first node; marking, by the first node in the cluster of nodes, the identified entries as belonging to a third node in the cluster of nodes; and obtaining, by the first node in the cluster of nodes, each of the marked entries from the third node. |
