Oracle International Corporation

Ming Zhu

Andrey Gusev

Mehul Dilip Bastawala

Mahesh Kallanagoudar

Cheng-Lu Hsu

Rajesh Dasari

Harsha Kancharthi

Described is a system, method, and computer program product for performing elections in a database cluster, where system resource statistics information is used to predict a cluster node failure. Resource statistics data is classified and used to identify anomalies. The anomalies can be used to determine the probability of a cluster node failure and to then elect a new master node and/or surviving sub-cluster.

Method and system to implement cluster failure prediction to facilitate split brain resolution

A computer program product, system, and computer implemented method comprises a multi-layered approach to virtual IP address assignment, where a managing computing node may control the generation of virtual IP addresses and assignment thereof to respective computing nodes, and where each respective computing node can control the allocation and binding of those virtual IP addresses to applications for the virtual IP addresses assigned to that computing node. Furthermore, in some embodiments, the approach includes a process to re-allocate virtual IP addresses to rebalance resources already allocated to a computing node and to address changing conditions.

Highly available virtual internet protocol addresses as a configurable service in a cluster

The embodiments disclosed herein relate to using SNMP functionality for fast and accurate detection of node failure in a computer cluster. When cluster nodes are connected via a switch that provides SNMP capability, a trap mechanism can be used to immediately and proactively notify cluster nodes that a particular cluster node is disconnected from the network, potentially even before the remaining cluster nodes have noticed a lack of response. In addition, when a cluster node notices a lack of response from a particular cluster node, the switch can be queried for link status and traffic information that can be used for fast determination of node status.

Determining the health of other nodes in a same cluster based on physical link information

Described is an improved approach to detect conflicts for the assignment of addresses in a computing system. The improved approach implements identification of address conflicts across multiple broadcast domains in a computing system.

Rapid duplicate IP address detection for floating IP address crossing multiple cluster broadcast domains

Described is a system, method, and computer program product to handle unresponsive node communications between two nodes of a database cluster. A high availability monitoring module is provided to address unresponsive node communications before having to automatically evict nodes from the cluster simply for exceeding a communications timeout period threshold.

Method and system for a high availability IP monitored by both OS/network and database instances

A software change system and method implements changes of a current active software version hosted one or more testing nodes of a cluster(s) to a different software version, while other nodes within the cluster(s) continue to employ the current active software version. To determine and resolve errors and incompatibilities introduced by the different software version, the at least one testing node is upgraded to the different software version and employed to query resources and data associated with the current active software version. Resolution data pertaining to the different software version is generated reflecting changes to the data employed by the different software version, or changes to the different software version, to resolve the errors and incompatibilities. The resolution data may be deployed to other nodes of the cluster(s) before, during, or after the nodes are upgraded on a rolling or other basis to the different software version.

Method and system for implementing software changes using rolling upgrades with reduced downtime

Described is a system, method, and computer program product for performing elections in a database cluster, where cluster topology representations are distributed across the cluster nodes using a prioritized messaging protocol. The cluster topology representations may be implemented as connectivity bit-vector that provide information that identifies which nodes in the cluster are able to communicatively connect with other nodes.

Method and system to implement messaging for cluster elections

The embodiments disclosed herein select and assign a floating IP address (FIP) to an application based on characteristics of the application. Floating IP addresses are assigned a classification, and one or more network adapters are mapped to a FIP according to the FIPs assigned classification. The endpoint of a connection for an application initiating a network connection with a target application may be assigned a floating IP address according to attributes of the application being consistent with the classification of a FIP. Applications may be grouped according to their respective classification, and the group of applications may share a corresponding FIP. FIPs may be dynamically configured for an application with mapped network adapters selected based on the attributes of the application. An application's characteristics may change during executions, and the network adapters mapped to the FIP assigned to the application's connection may be reassigned.

Client connection failover

Described is a system, method, and computer program product for identifying communication failures between nodes of database cluster system where each member node within the cluster constantly monitors itself. Upon detection of a node communication failure by a particular node, the particular node notifies other member nodes of the cluster that the particular node is having communication problems with other nodes of the database cluster by communicating through a different communication channel with the other nodes so that the particular node may be evicted from the cluster efficiently.

Growing community of inventors

Redwood City, CA, United States of America

Ming Zhu