Abstract: Methods, systems, and media for correlating error events of a cluster are disclosed. Embodiments may identify systems of a cluster potentially impacted by an error and identify one or more error events associated with those systems. Then, embodiments may select one of the identified error events based upon data associated with the identified error event, disregarding other identified error events generated for the same error or errors symptomatic of the error, to report the error to a maintenance provider via a single error event. Many embodiments may identify one or more error events potentially resulting from the same error by identifying error events within a specified time period of the event that triggered the correlation. Several embodiments correlate the error events in an environment that is substantially independent of the cluster. Further embodiments obtain data that describes system interconnections of the cluster and generate a topology based upon the data.

Abstract: Art apparatus, system, and storage medium that, in an embodiment, register each partition in a logically-partitioned computer to send local events to an aggregation component and register a first partition to send events associated with a group of partitions to the aggregation component. In various embodiments, the group may be all of the partitions or some of the partitions. If the connection to the first partition is unavailable, the first partition is registered to cancel sending events associated with the group, a second partition is selected, and the second partition is registered to send events associated with the group to the aggregation component. If the connection to the first partition is restored subsequent to being unavailable, a determination is made whether to change registration back to the first partition. In this way, in an embodiment, redundant reporting of events is supported while decreasing the number of events reported.

Abstract: A method, apparatus, system, and signal-bearing medium that, in an embodiment, register each partition in a logically-partitioned computer to send local events to an aggregation component and register a first partition to send events associated with a group of partitions to the aggregation component. In various embodiments, the group may be all of the partitions or some of the partitions. If the connection to the first partition is unavailable, the first partition is registered to cancel sending events associated with the group, a second partition is selected, and the second partition is registered to send events associated with the group to the aggregation component. If the connection to the first partition is restored subsequent to being unavailable, a determination is made whether to change registration back to the first partition. In this way, in an embodiment, redundant reporting of events is supported while decreasing the number of events reported.

Abstract: Methods, systems, and media for correlating error events of a cluster are disclosed. Embodiments may identify systems of a cluster potentially impacted by an error and identify one or more error events associated with those systems. Then, embodiments may select one of the identified error events based upon data associated with the identified error event, disregarding other identified error events generated for the same error or errors symptomatic of the error, to report the error to a maintenance provider via a single error event. Many embodiments may identify one or more error events potentially resulting from the same error by identifying error events within a specified time period of the event that triggered the correlation. Several embodiments correlate the error events in an environment that is substantially independent of the cluster. Further embodiments obtain data that describes system interconnections of the cluster and generate a topology based upon the data.

Abstract: An apparatus, system, and storage medium that, in an embodiment, register each partition in a logically-partitioned computer to send local events to an aggregation component and register a first partition to send events associated with a group of partitions to the aggregation component. In various embodiments, the group may be all of the partitions or some of the partitions. If the connection to the first partition is unavailable, the first partition is registered to cancel sending events associated with the group, a second partition is selected, and the second partition is registered to send events associated with the group to the aggregation component. If the connection to the first partition is restored subsequent to being unavailable, a determination is made whether to change registration back to the first partition. In this way, in an embodiment, redundant reporting of events is supported while decreasing the number of events reported.

Abstract: Methods, systems, and media for correlating error events of a cluster are disclosed. Embodiments may identify systems of a cluster potentially impacted by an error and identify one or more error events associated with those systems. Then, embodiments may select one of the identified error events based upon data associated with the identified error event, disregarding other identified error events generated for the same error or errors symptomatic of the error, to report the error to a maintenance provider via a single error event. Many embodiments may identify one or more error events potentially resulting from the same error by identifying error events within a specified time period of the event that triggered the correlation. Several embodiments correlate the error events in an environment that is substantially independent of the cluster. Further embodiments obtain data that describes system interconnections of the cluster and generate a topology based upon the data.

Abstract: Method, apparatus and article of manufacture for error notification in a dynamically partitioned environment in which logical partitions are added and/or removed during operation of a system. Logical partitions are notified of errors based on a predefined criterion(s). In one aspect, a criterion is applied to the maintenance of an error notifications list. In another aspect, a notification and/or acknowledgment criterion is applied in the determination of whether to notify a logical partition of an error notification that has not been discarded.

Abstract: Methods, systems, and media for correlating error events of a cluster are disclosed. Embodiments may identify systems of a cluster potentially impacted by an error and identify one or more error events associated with those systems. Then, embodiments may select one of the identified error events based upon data associated with the identified error event, disregarding other identified error events generated for the same error or errors symptomatic of the error, to report the error to a maintenance provider via a single error event. Many embodiments may identify one or more error events potentially resulting from the same error by identifying error events within a specified time period of the event that triggered the correlation. Several embodiments correlate the error events in an environment that is substantially independent of the cluster. Further embodiments obtain data that describes system interconnections of the cluster and generate a topology based upon the data.

Abstract: Method, apparatus and article of manufacture for error notification in a dynamically partitioned environment in which logical partitions are added and/or removed during operation of a system. Logical partitions are notified of errors based on a predefined criterion(s). In one aspect, a criterion is applied to the maintenance of an error notifications list. In another aspect, a notification and/or acknowledgment criterion is applied in the determination of whether to notify a logical partition of an error notification that has not been discarded.