Abstract: Mechanisms are provided for performing a write operation on a shared resource in a cluster of data processing systems. The mechanisms determine whether a locally cached copy of the shared resource is present in a local cache memory. The mechanisms, in response to a determination that a locally cached copy is present, determine whether the locally cached copy is a latest version of the shared resource. The mechanisms, in response to determining that that locally cached copy is a latest version: perform the write operation on the locally cached copy to generate an updated locally cached copy, and transmit a cluster event notification to other data processing systems of the cluster indicating the shared resource was written to and which data processing system performed the write operation. The cluster event notification is logged in at least one change log data structure associated with the other data processing systems.

Abstract: Mechanisms are provided for performing a write operation on a shared resource in a cluster of data processing systems. The mechanisms determine whether a locally cached copy of the shared resource is present in a local cache memory. The mechanisms, in response to a determination that a locally cached copy is present, determine whether the locally cached copy is a latest version of the shared resource. The mechanisms, in response to determining that that locally cached copy is a latest version: perform the write operation on the locally cached copy to generate an updated locally cached copy, and transmit a cluster event notification to other data processing systems of the cluster indicating the shared resource was written to and which data processing system performed the write operation. The cluster event notification is logged in at least one change log data structure associated with the other data processing systems.

Abstract: An event notification system for distributed processing systems provides for replacement of event identifiers associated with transient objects, such as processes within the processing system. An event management interface receives a notification from an event producer that an identifier of an event produced by the event producer is no longer valid. The event management interface blocks event notification for the event and dissociates an event registration entry associated with the event from the event identifier. A new event identifier is received from the event producer by the event management interface, which associates the registration entry with the new event identifier. Event notification for the event is then resumed.

Abstract: In response to a stimulus indicating configuration of a node into a cluster of a plurality of nodes including the node, the node determines whether or not the node has a universally unique identifier (UUID), and if not, the node provides its own persistent self-assigned UUID. The node searches a cluster configuration database for a temporary identifier associated with the node. In response to the node locating the temporary identifier of the node in the cluster configuration database, the node writes its self-assigned UUID into the cluster configuration database and joins the cluster.

Abstract: A mechanism is provided in a data processing system for unobtrusive failover in a clustered file system. A node in the clustered file system periodically monitors publicized resource events from a cluster manager in the clustered file system for a triggering event that triggers a grace period. Responsive to detecting a triggering event, the node enters a grace period without restarting. During the grace period, the node processes a state request from a first client.

Abstract: A mechanism is provided in a data processing system for protocol servicing in network-attached storage. A protocol service executing in the data processing system in a clustered file system establishes a high priority recovery thread. The high priority recovery thread monitors health counters that count total requests in and total requests out for a client accessing a network-attached storage device via the protocol service. The high priority recovery thread determines a category of health of the protocol service based on the health counters. The mechanism takes corrective action based on the category of health of the protocol service.

Abstract: According to one aspect of the present disclosure, a method and technique for product license management for a clustered environment having a plurality of nodes is disclosed. The method includes unlocking a product on a first node of the plurality of clustered nodes; responsive to unlocking the product on the first node, indicating an unlocked status of the product on a shared storage device accessible to the plurality of clustered nodes; and transmitting a self-unlock message from the first node to remaining nodes of the cluster to enable the remaining nodes of the cluster to self-unlock the product on the respective remaining nodes based on the status indication of the shared storage device.

Abstract: Communication between clusters of processing nodes is initiated by designating a link layer address and a target node name in a remote cluster. The link layer address is transmitted to the target node via a regular network transmission. The link layer address is stored locally in each cluster and the nodes in each cluster are signaled that a configuration change has been made, so that gateway nodes in the clusters can obtain the link layer address to use for subsequent communications with the other cluster, such as event notifications of node status changes.

Abstract: A gossiping scheme for sharing node status in a cluster of nodes provides a robust mechanism for determining node status within the cluster. Nodes transmit gossip messages to each other nodes, the gossip messages listing other nodes in the cluster that are operational. When a node does not receive a gossip message from a particular node within a predetermined time period, then the node transmits messages to the other nodes indicating that the particular node is down. However, if another node has received a packet from the particular node within the predetermined time period and receives the node down message, then the other node responds with a node alive message.

Abstract: Communication ability between nodes in a cluster-based computer system is tracked to inform applications executing on the nodes of the existence and quality of the endpoint-to-endpoint communications available between the nodes. Communications between a node and other nodes are tracked, and a database records the communication ability between the node and the other nodes for each link between the nodes. The tracking and recording are repeated at the other nodes. A registration by an application executing at a particular one of the nodes to receive notifications of changes in the communication ability with another node over a particular link (or in general) will cause notification of the application when the link status changes.

Abstract: A gossiping scheme for sharing node status in a cluster of nodes provides a robust mechanism for determining node status within the cluster. Nodes transmit gossip messages to each other nodes, the gossip messages listing other nodes in the cluster that are operational. When a node does not receive a gossip message from a particular node within a predetermined time period, then the node transmits messages to the other nodes indicating that the particular node is down. However, if another node has received a packet from the particular node within the predetermined time period and receives the node down message, then the other node responds with a node alive message.

Abstract: Provided are techniques related generally to computer communication and, more specifically, to techniques that provide legacy applications with access to a cluster infrastructure. A pseudo interface layer provides applications with internet protocol (IP) family sockets access to cluster awareness protocols in a manner that may be transparent to both the application and the cluster awareness layer of a kernel.

Abstract: A second node receives a message from a first node in a cluster environment. The message includes a unique identifier of a shared data storage device including a cluster configuration database that defines membership of nodes in a cluster. In response to receiving the message, the second node attempts to find the shared data storage device. In response to finding the shared data storage device, the second node locates and reads the cluster configuration database on the shared data storage device. The second node then assimilates a cluster configuration update indicated by the cluster configuration database.

Abstract: An event notification method for distributed processing systems provides remote and local node event notification in systems that require local registration of an event consumer in order produce an event notifications. To provide notification of an event occurring on a remote node, either event consumers on all nodes in the cluster register locally to receive event notifications and specify that the event is a cluster event, in which case the nodes send notification of their locally-occurring events to all nodes, or remote registrations are accepted at nodes and if a local consumer for the event is not present, a listener thread registers as an event consumer. The listener thread sends the event notifications to the remote nodes registered as consumers for the event by observing communication between the event producer and the local consumer, or receiving the event notifications directly if there is no local consumer.

Abstract: Provided are techniques related generally to computer communication and, more specifically, to techniques that provide legacy applications with access to a cluster infrastructure. A pseudo interface layer provides applications with internet protocol (IP) family sockets access to cluster awareness protocols in a manner that may be transparent to both the application and the cluster awareness layer of a kernel.

Abstract: An event notification system for distributed processing systems provides remote and local node event notification in systems that require local registration of an event consumer in order produce an event notifications. To provide notification of an event occurring on a remote node, either event consumers on all nodes in the cluster register locally to receive event notifications and specify that the event is a cluster event, in which case the nodes send notification of their locally-occurring events to all nodes, or remote registrations are accepted at nodes and if a local consumer for the event is not present, a listener thread registers as an event consumer. The listener thread sends the event notifications to the remote nodes registered as consumers for the event by observing communication between the event producer and the local consumer, or receiving the event notifications directly if there is no local consumer.

Abstract: A scheme for monitoring node operational status according to communications transmits messages periodically according to a heartbeat rate among the nodes. The messages may be gossip messages containing the status of the other nodes in the pairs, are received at the nodes and indications of the communications delays of the received messages are stored, which are used to compute statistics of the stored communications delays. Parameters of the node status monitoring, which are used for determining operational status of the nodes, are adjusted according to the statistics, which may include adjusting the heartbeat rate, the maximum wait time before a message is considered missed, and/or the maximum number of missed messages, e.g., the sequence number deviation, before the node is considered non-operational (down).

Abstract: Communication ability between nodes in a cluster-based computer system is tracked to inform applications executing on the nodes of the existence and quality of the endpoint-to-endpoint communications available between the nodes. Communications between a node and other nodes are tracked, and a database records the communication ability between the node and the other nodes for each link between the nodes. The tracking and recording are repeated at the other nodes. A registration by an application executing at a particular one of the nodes to receive notifications of changes in the communication ability with another node over a particular link (or in general) will cause notification of the application when the link status changes.

Abstract: Communication between clusters of processing nodes is initiated by designating a link layer address and a target node name in a remote cluster. The link layer address is transmitted to the target node via a regular network transmission. The link layer address is stored locally in each cluster and the nodes in each cluster are signaled that a configuration change has been made, so that gateway nodes in the clusters can obtain the link layer address to use for subsequent communications with the other cluster, such as event notifications of node status changes.

Abstract: A gossiping scheme for sharing node status in a cluster of nodes provides a robust mechanism for determining node status within the cluster. Nodes transmit gossip messages to each other nodes, the gossip messages listing other nodes in the cluster that are operational. When a node does not receive a gossip message from a particular node within a predetermined time period, then the node transmits messages to the other nodes indicating that the particular node is down. However, if another node has received a packet from the particular node within the predetermined time period and receives the node down message, then the other node responds with a node alive message.