Patent EP0887731A1 successfully refreshed from Google Patents.
EP0887731A1
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Membership in an unreliable distributed computer system
Assignee: Sun Microsystems Inc
•
Filing: 1998-06-26
•
Granted: —
•
Expires: 2018-06-26
Abstract
Abstract Multiple nodes can concurrently gain membership in a cluster of nodes of a distributed computer system by broadcasting reconfiguration messages to all nodes of the distributed computer system. In response to a reconfiguration request resulting from a node petitioning to join a cluster or a node leaving the cluster, each node determines to which nodes of the distributed computer system the node is connected, i.e., which are sending reconfiguration messages which the node receives. In addition, if multiple nodes fail substantially simultaneously, each node which continues to operate does not receive a reconfiguration message from each of the failed nodes and the failed nodes are omitted from the proposed new cluster. Thus, multiple simultaneous failures are processed in a single reconfiguration. Each of the member nodes of the proposed cluster determine the membership of the proposed cluster and broadcast a reconfiguration message to all proposed member nodes and collects similar messages. If all reconfiguration messages agree, the proposed cluster is accepted. In the case in which one or more nodes leave the cluster, quorurn is established in the new cluster relative to the old cluster.
Inventors
- 👤 Hossein Moiin
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-01 12:37
CPC Classifications
G; G06; G06F; G06F11/00; G06F11/07; G06F11/14; G06F11/1402; G06F11/1415; G06F11/142; G06F11/1425
Claims Analysis
Total Claims: 21
Claim 1
Independent Claim
126 words
A method implemented by a subject node computer of a distributed computer system for adding at least one new node to a cluster which includes at least one member node and which includes at least the subject node computer, the method comprising: (a) determining which connected other nodes of the distributed computer system are operative and in communication with the subject node computer; (b) creating a reconfigure message which specifies a proposed new cluster which includes the connected nodes; (c) transmitting the reconfigure message to each of the other nodes; (d) receiving a responsive reconfiguration message from each of the other nodes; and (e) determining that each responsive reconfiguration message specifies equivalent membership of the proposed new cluster to the membership specified by the reconfigure message.
Claim 2
Dependent Claim
69 words
The method of Claim 1 wherein step (a) comprises: waiting to receive messages from the at least one member node of the distributed computer system for a predetermined period of time; and determining that each of the at least one member node of the distributed computer system from which a message is received in the predetermined period of time is operative and in communication with the subject node computer.
Claim 3
Independent Claim
101 words
A method implemented by a subject node computer of a distributed computer system for removing at least one leaving node from an old cluster which includes at least one member node and which includes at least the subject node computer and the leaving node, the method comprising: determining which connected other nodes of the at least one member node of the distributed computer system are operative and in communication with the subject node computer; configuring a prospective new cluster which includes the connected other nodes; and attempting to establish a quorum for the proposed new cluster relative to the old cluster.
Claim 4
Dependent Claim
71 words
The method of Claim 3 wherein the step of determining comprises: waiting to receive messages from the at least one member node of the distributed computer system for a predetermined period of time; and determining that each of the at least one member node of the distributed computer system from which a message is received in the predetermined period of time is operative and in communication with the subject node computer.
Claim 5
Dependent Claim
78 words
The method of Claim 3 wherein the step of attempting to establish a quorum comprises: determining a number of member nodes of the old cluster; attempting to establish the quorum using a face mechanism upon a condition in which the number of member nodes of the old cluster is two; and attempting to establish the quorum using a vote mechanism upon a condition in which the number of member nodes of the old cluster is greater than two.
Claim 6
Dependent Claim
61 words
The method of Claim 5 wherein the step of attempting to establish the quorum using a race mechanism comprises: determining whether the leaving node and the subject node computer share a quorum device; and attempting to establish the quorum using an alternative mechanism upon a condition in which the leaving node and subject node computer do not share a quorum device.
Claim 7
Dependent Claim
40 words
The method of Claim 6 wherein the alternative mechanism includes performance of the steps of: prompting a human operator to select a winner node from a group consisting of the leaving node and the subject node computer using user-interface techniques.
Claim 8
Dependent Claim
162 words
A computer readable medium useful in association with a subject node computer which includes at least one processor and a memory, the computer readable medium including computer instructions which are configured to cause the subject node computer to add at least one new node to a cluster which includes at least one member node and which includes at least the subject node computer by performing the steps of: (a) determining which connected other nodes of the distributed computer system are operative and in communication with the subject node computer regardless of membership of the connected nodes in the cluster; (b) creating a reconfiguration message which specifies a proposed new cluster which includes the connected nodes; (c) transmitting the reconfiguration message to each of the other nodes; (d) receiving a responsive reconfiguration message from each of the other nodes; and (e) determining that each responsive reconfiguration message specifies equivalent membership of the proposed new cluster to the membership specified by the reconfiguration message.
Claim 9
Dependent Claim
71 words
The computer readable medium of Claim 8 wherein step (a) comprises: waiting to receive messages from the at least one member node of the distributed computer system for a predetermined period of time; and determining that each of the at least one member node of the distributed computer system from which a message is received in the predetermined period of time is operative and in communication with the subject node computer.
Claim 10
Dependent Claim
132 words
A computer readable medium useful in association with a subject node computer which includes at least one processor and a memory, the computer readable medium including computer instructions which are configured to cause the subject node computer to remove at least one leaving node from an old cluster which includes at least one member node and which includes at least the subject node computer and the leaving node by performing the steps of: determining which connected other nodes of the distributed computer system are operative and in communication with the subject node computer regardless of membership of the connected other nodes in the old cluster; configuring a prospective new cluster which includes the connected other nodes; and attempting to establish a quorum for the proposed new cluster relative to the old cluster.
Claim 11
Dependent Claim
73 words
The computer readable medium of Claim 10 wherein the step of determining comprises: waiting to receive messages from the at least one member node of the distributed computer system for a predetermined period of time; and determining that each of the at least one member node of the distributed computer system from which a message is received in the predetermined period of time is operative and in communication with the subject node computer.
Claim 12
Dependent Claim
80 words
The computer readable medium of Claim 10 wherein the step of attempting to establish a quorum comprises: determining a number of member nodes of the old cluster; attempting to establish the quorum using a race mechanism upon a condition in which the number of member nodes of the old cluster is two; and attempting to establish the quorum using a vote mechanism upon a condition in which the number of member nodes of the old cluster is greater than two.
Claim 13
Dependent Claim
63 words
The computer readable medium of Claim 12 wherein the step of attempting to establish the quorum using a race mechanism comprises: determining whether the leaving node and the subject node computer share a quorum device; and attempting to establish the quorum using an alternative mechanism upon a condition in which the leaving node and subject node computer do not share a quorum device.
Claim 14
Dependent Claim
43 words
The computer readable medium of Claim 13 wherein the alternative mechanism includes performance of the steps of : prompting a human operator to select a winner node from a group consisting of the leaving node and the subject node computer using user-interface techniques.
Claim 15
Dependent Claim
160 words
A subject node computer system comprising: at least one processor; a memory operatively coupled to the at least one processor; and a cluster membership module configured to cause the subject node computer system to add at least one new node to a cluster which includes at least one member node and which includes at least the subject node computer system by performing the steps of: (a) determining which connected other nodes of the distributed computer system are operative and in communication with the subject node computer system regardless of membership of the connected nodes in the cluster; (b) creating a reconfigure message which specifies a proposed new cluster which includes the connected nodes; (c) transmitting the reconfigure message to each of the other nodes; (d) receiving a responsive reconfiguration message from each of the other nodes; and (e) determining that each responsive reconfiguration message specifies equivalent membership of the proposed new cluster to the membership specified by the reconfigure message.
Claim 16
Dependent Claim
52 words
The subject node computer system of Claim 15 wherein step (a) comprises: waiting to receive reconfiguration messages for a predetermined period of time; and determining that each node from which a reconfiguration message is received in the predetermined period of time is operative and in communication with the subject node computer system.
Claim 17
Dependent Claim
129 words
A subject node computer system comprising: at least one processor; a memory operatively coupled to the at least one processor; and a cluster membership module configured to cause the subject node computer system to remove at least one leaving node from an old cluster which includes at least one member node and which includes at least the subject node computer system and the leaving node by performing the steps of: determining which connected other nodes of the distributed computer system are operative and in communication with the subject node computer regardless of membership of the connected other nodes in the old cluster; configuring a prospective new cluster which includes the connected other nodes; and attempting to establish a quorum for the proposed new cluster relative to the old cluster.
Claim 18
Dependent Claim
54 words
The subject node computer system of Claim 17 wherein the step of determining comprises: waiting to receive reconfiguration messages for a predetermined period of time; and determining that each node from which a reconfiguration message is received in the predetermined period of time is operative and in communication with the subject node computer system.
Claim 19
Dependent Claim
81 words
The subject node computer system of Claim 17 wherein the step of attempting to establish a quorum comprises: determining a number of member nodes of the old cluster; attempting to establish the quorum using a race mechanism upon a condition in which the number of member nodes of the old cluster is two; and attempting to establish the quorum using a vote mechanism upon a condition in which the number of member nodes of the old cluster is greater than two.
Claim 20
Dependent Claim
66 words
The subject node computer system of Claim 19 wherein the step of attempting to establish the quorum using a race mechanism comprises: determining whether the leaving node and the subject node computer system share a quorum device; and attempting to establish the quorum using an alternative mechanism upon a condition in which the leaving node and subject node computer system do not share a quorum device.
Claim 21
Dependent Claim
44 words
The subject node computer system of Claim 20 wherein the alternative mechanism includes performance of the steps of: prompting a human operator to select a winner node from a group consisting of the leaving node and the subject node computer system using user-interface techniques.
References Cited (Backward Citations)
EP0481231A2
Cited
EP0772126A2
Cited
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.