MONITORING STORAGE CLUSTER ELEMENTS
First Claim
1. A method comprising:
- determining topology of a backend of a first storage cluster and a backend of a second storage cluster associated with the first storage cluster;
periodically obtaining first operational information from network elements of the first storage cluster backend and from network elements of the second storage cluster backend;
after obtaining the first operational information from any one of the network elements,updating storage cluster information that is communicated among the first storage cluster based, at least in part, on the operational information;
determining whether the operational information violates a first set of one or more rules;
determining whether a notification should be generated in response to a determination that the operational information violates the first set of rules;
generating the notification in response to determining that the notification should be generated;
obtaining second operational information from a local subsystem;
determining whether the second operational information in combination with the first operational information violates a second set of one or more rules; and
generating a notification if the combination of the first operational information and the second operational information violates the second set of rules.
1 Assignment
0 Petitions
Accused Products
Abstract
Monitoring health of associated, but separated storage clusters can be done at both a node scope and a cluster scope. Monitoring the storage clusters at the cluster scope includes monitoring the network elements that support the storage clusters and connect the storage clusters. Initially, a fabric monitor in each cluster discovers cluster topology. This cluster topology is communicated and maintained throughout the managing storage elements of the storage clusters. After the storage cluster topologies have been discovered, the fabric monitors of each cluster can periodically determine status of network elements of the storage clusters. This allows the storage clusters to maintain awareness of interconnect status, and react to changes in status. In addition, each managing storage element monitors its own health. This information is aggregated to determine when to trigger corrective actions, alerts, and/or storage features in accordance with rules defined at the managing storage elements.
94 Citations
20 Claims
-
1. A method comprising:
-
determining topology of a backend of a first storage cluster and a backend of a second storage cluster associated with the first storage cluster; periodically obtaining first operational information from network elements of the first storage cluster backend and from network elements of the second storage cluster backend; after obtaining the first operational information from any one of the network elements, updating storage cluster information that is communicated among the first storage cluster based, at least in part, on the operational information; determining whether the operational information violates a first set of one or more rules; determining whether a notification should be generated in response to a determination that the operational information violates the first set of rules; generating the notification in response to determining that the notification should be generated; obtaining second operational information from a local subsystem; determining whether the second operational information in combination with the first operational information violates a second set of one or more rules; and generating a notification if the combination of the first operational information and the second operational information violates the second set of rules. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. One or more non-transitory machine-readable media having program code stored therein, the program code to:
-
determine topology of a backend of a first storage cluster and a backend of a second storage cluster associated with the first storage cluster; periodically obtain operational information from network elements of the first storage cluster backend and from network elements of the second storage cluster backend; in response to receipt of operational information from a network element, update storage cluster information that is communicated among the first storage cluster based, at least in part, on the operational information from the network element; determine whether the operational information violates a first set of one or more rules; determine whether a notification should be generated in response to a determination that the operational information violates the first set of rules; generate the notification in response to determining that the notification should be generated; obtain operational information from a local subsystem; determine whether the operational information from the local subsystem in combination with the periodically obtained operational information violates a second set of one or more rules; and generate a notification if the combination of operational information violates the second set of rules. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. An apparatus comprising:
-
a processor; and a machine-readable storage medium having program code stored therein, the program code executable by the processor to cause the apparatus to, determine topology of a backend of a first storage cluster and a backend of a second storage cluster associated with the first storage cluster; periodically obtain operational information from network elements of the first storage cluster backend and from network elements of the second storage cluster backend; in response to receipt of operational information from a network element, update storage cluster information that is communicated among the first storage cluster based, at least in part, on the operational information from the network element; determine whether the operational information violates a first set of one or more rules; determine whether a notification should be generated in response to a determination that the operational information violates the first set of rules; generate the notification in response to determining that the notification should be generated; obtain operational information from a local subsystem; determine whether the operational information from the local subsystem in combination with the periodically obtained operational information violates a second set of one or more rules; and generate a notification if the combination of operational information violates the second set of rules. - View Dependent Claims (18, 19, 20)
-
Specification