Oracle International Corporation

Netapp, Inc.

Network Appliance, Inc.

Andrey Gusev

Jonathan Creighton

Juan Tellez

Sarat B Kakarla

Raj K Kammend

Ming Zhu

Tirthankar Lahiri

David Brower

Tak Fung Wang

Richard Meyer

Rohan Aranha

Ashwinee Khaladkar

Sourav Ghosh

Ashish Motivala

Charles Hoch

Ben Fulton

Kumar Gajjar

Chan Y Ng

Vinay Manivel

Yuri Khodosh

Rajesh Dasari

The fail-over computer cluster enables multiple computing devices to operate using adaptive quorum rules to dictate which nodes are in the fail-over cluster at any given time. The adaptive quorum rules provide requirements for communications between nodes and connections with voting file systems. The adaptive quorum rules include particular recovery rules for unplanned changes in node configuration, such as due to a disruptive event. Such recovery quorum rules enable the fail-over cluster to continuing to operate with various changed configurations of its node members as a result of the disruptive event. In the changed configuration, access to voting file systems may not be required for a majority-group subset of nodes. If no majority-group subset remains, nodes may need direct or indirect access to voting file systems.

Computer cluster using expiring recovery rules

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

Computer cluster with adaptive quorum rules

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.

Method and system for a speed-up cluster reconfiguration time via a generic fast self node death detection

A cluster manager manages copies of a mid-tier database as a mid-tier database cluster. The cluster manager may concurrently manage a backend database system. The cluster manager is configured to monitor for and react to failures of mid-tier database nodes. The cluster manager may react to a mid-tier database failure by, for example, assigning a new active node, creating a new standby node, creating new copies of the mid-tier databases, implementing new replication or backup schemes, reassigning the node's virtual address to another node, or relocating applications that were directly linked to the mid-tier database to another host. Each node or an associated agent may configure the cluster manager during initialization, based on common cluster configuration information. Each copy of the mid-tier database may be, for example, a memory resident database. Thus, a node must reload the entire database into memory to recover a copy of the database.

Automated integrated high availability of the in-memory database cache and the backend enterprise database

A method and apparatus for resource dispersion in a clusterware system is provided. The resource profiles for resources within a cluster of nodes are stored. Each resource profile includes attributes for a resource. One or more of the attributes in one of the resource profiles includes a dispersion dependency attribute that specifies that the resource is preferably not co-located with another resource. Resources are placed within the cluster based, at least in part, on the dispersion preference.

Dispersion dependency in oracle clusterware

A method and computer-readable storage representing resources in a cluster by a plurality of attribute-value pairs that together are part of a 'resource profile,' in which each attribute-value pair defines all, or a portion of, a management policy that applies to the resource. A clusterware manager configures a resource according to the resource profile, and follows a resource profile syntax that specifies a runtime value for the resource in which an actual value is substituted at runtime.

Special values in oracle clusterware resource profiles

A method for a self-testing clusterware agent is provided. A clusterware agent that includes clusterware-side components and application-side components is configured to interface between a cluster manager and an application. The application-side components are invoked by clusterware-side components via an application programming interface, or API that includes API functions that are invocable by a cluster manager. Without any cluster manager invoking the clusterware agent, one or more of the API functions are invoked.

Self-testable HA framework library infrastructure

A cluster management system dynamically assigns a category to a server node based on the server node's attributes. The system can detect a change in value of a server attribute of a first server node. Then, in response to detecting the change in value of the server attribute, the system identifies a plurality of server attributes for the first server node, and dynamically assigns to the first server node a category that matches the server attributes. Further, the system can add the server node to a server pool that has a matching category with the server node.

Assigning server categories to server nodes in a heterogeneous cluster

A method and computer-readable storage representing resources in a cluster by a plurality of attribute-value pairs that together are part of a 'resource profile,' in which each attribute-value pair defines all, or a portion of, a management policy that applies to the resource. Included is a built-in resource type called a 'Local Resource' type. Local resources that are attached to the server are instantiated with a combination of the resource profile and values of the Local Resource type.

'Local resource' type as a way to automate management of infrastructure resources in oracle clusterware

A cluster management system can facilitate determining a response to a hypothetical event by a computer cluster. During operation, the system can receive a request for a policy outcome from a user, such that the request can indicate a hypothetical event within a computer cluster. Next, the system determines a cluster state that includes state information from a plurality of server nodes in the cluster. The system then generates a response plan to the hypothetical event based on the cluster state, without executing the response plan in the cluster. The response plan indicates actions that are to be performed by server nodes in the cluster in response to the hypothetical event. The system then provides the response plan to the user.

System and method for generating a response plan for a hypothetical event

A clusterware manager configures a resource according to resource attributes values specified by a resource profile. The resource profile conforms to a resource profile syntax that the clusterware manager is configured to interpret pursuant to clusterware manager software. The resource profile syntax prescribes a start dependency syntax defining a dependency between a first resource and a second resource in which the second resource must be in an online state before the first resource is started. The resource profile syntax further prescribes a stop dependency syntax defining a dependency between a first resource and a second resource in which the first resource is brought in an off-line state when the second resource leaves an online state.

Unidirectional resource and type dependencies in oracle clusterware

A clusterware manager on a cluster of nodes interprets a resource profile. The resource profile defines resource profile attributes. The attributes include at least one attribute that defines a cluster dependency based on resource type. The attribute does not identify any particular resource of that resource type. Dependencies between resources are managed based on the attribute that specifies the cluster dependency.

Dependency on a resource type

A cluster manager is configured to manage a plurality of copies of a mid-tier database as a mid-tier database cluster. The cluster manager may concurrently manage a backend database system. The cluster manager is configured to monitor for and react to failures of mid-tier database nodes. The cluster manager may react to a mid-tier database failure by, for example, assigning a new active node, creating a new standby node, creating new copies of the mid-tier databases, implementing new replication or backup schemes, reassigning the node's virtual address to another node, or relocating applications that were directly linked to the mid-tier database to another host. Each node or an associated agent may configure the cluster manager to behave in this fashion during initialization, based on common cluster configuration information. Each copy of the mid-tier database may be, for example, a memory resident database. Thus, a node must reload the entire database into memory to recover a copy of the database.

Growing community of inventors

Dublin, CA, United States of America

Andrey Gusev