Abstract: Utilizing several passes of a token among the operating system level virtualization applications (OSLVAs) running on each of an ordered list of nodes selectively upgrade user applications in specific containers after determining that each OSVLA hosting a container to be upgraded is live. A failed liveness check can cause a revision to the list of nodes. Post-upgrade health checks of each application, and a list-wide notification of a completed upgrade of the application instances, can follow a successful upgrade. A roll back of the upgrade across all nodes can follow an unsuccessful upgrade.

Abstract: Utilizing several passes of a token among the operating system level virtualization applications (OSLVAs) running on each of an ordered list of nodes selectively upgrade user applications in specific containers after determining that each OSVLA hosting a container to be upgraded is live. A failed liveness check can cause a revision to the list of nodes. Post-upgrade health checks of each application, and a list-wide notification of a completed upgrade of the application instances, can follow a successful upgrade. A roll back of the upgrade across all nodes can follow an unsuccessful upgrade.

Abstract: Utilizing several passes of a token among the operating system level virtualization applications (OSLVAs) running on each of an ordered list of nodes selectively upgrade user applications in specific containers after determining that each OSVLA hosting a container to be upgraded is live. A failed liveness check can cause a revision to the list of nodes. Post-upgrade health checks of each application, and a list-wide notification of a completed upgrade of the application instances, can follow a successful upgrade. A roll back of the upgrade across all nodes can follow an unsuccessful upgrade.

Abstract: Utilizing several passes of a token among the operating system level virtualization applications (OSLVAs) running on each of an ordered list of nodes selectively upgrade user applications in specific containers after determining that each OSVLA hosting a container to be upgraded is live. A failed liveness check can cause a revision to the list of nodes. Post-upgrade health checks of each application, and a list-wide notification of a completed upgrade of the application instances, can follow a successful upgrade. A roll back of the upgrade across all nodes can follow an unsuccessful upgrade.

Abstract: Methods and systems for escalating component failures in a cloud are provided. A cloud controller of a cloud receives an indication that a collection of virtual machines of the first cloud has failed based on a collection of virtual machines escalation policy. The cloud controller initiates relocating the collection of virtual machines to a second cloud.

Abstract: Methods and systems for offloading health-checking policy in a distributed management environment are provided. A failure policy is received at a node of a cloud from a cloud health monitor. The node transmits a notification to a health monitor of the node that the node has failed when the failure policy is satisfied. The node reports at least one fault based on the satisfied failure policy to the cloud health monitor.

Abstract: Methods and systems for automatically recovering from a failure event in one or more nodes of a redundant ring network are provided. An indication is received at a first ring of nodes that an associated second ring of nodes has failed. A probing token is circulated around the second ring of nodes while a probing timer is started. The injecting a probing token and starting a probing timer are repeated if the probing timer expires before the probing token circulates completely around the second ring of nodes. A recovery token is circulated around the second ring of nodes to enable each node of the second ring of nodes and to disable the probing timer if the probing token circulates completely around the second ring of nodes.

Abstract: Methods and systems for automatically recovering from a failure event in one or more nodes of a redundant ring network are provided. An indication is received at a first ring of nodes that an associated second ring of nodes has failed. A probing token is circulated around the second ring of nodes while a probing timer is started. The injecting a probing token and starting a probing timer are repeated if the probing timer expires before the probing token circulates completely around the second ring of nodes. A recovery token is circulated around the second ring of nodes to enable each node of the second ring of nodes and to disable the probing timer if the probing token circulates completely around the second ring of nodes.

Abstract: Methods and systems for offloading health-checking policy in a distributed management environment are provided. A failure policy is received at a node of a cloud from a cloud health monitor. The node transmits a notification to a health monitor of the node that the node has failed when the failure policy is satisfied. The node reports at least one fault based on the satisfied failure policy to the cloud health monitor.

Abstract: Methods and systems for escalating component failures in a cloud are provided. A cloud controller of a cloud receives an indication that a collection of virtual machines of the first cloud has failed based on a collection of virtual machines escalation policy. The cloud controller initiates relocating the collection of virtual machines to a second cloud.