Cluster properties
The file cluster.properties stores several settings related to clustering.
You should edit these settings via the user interface. Please note that some of these properties may be stored encrypted.
| Property | Default value | Encrypted | Explanation |
|---|---|---|---|
| cluster.group.name | XFC-CLUSTER | no | An identifier for the cluster network. |
| cluster.node.name | (generated from local inet address, time etc.) | no | An identifier for the master server. |
| cluster.node.type | MASTER | no | Whether this server is a master server or a backup (failover) server. Allowed values are MASTER and FAILOVER. |
| cluster.active | false | no | Activates or deactivates the cluster mode. |
| cluster.protocol.file | (empty string) | no | Path to the file on the file system with a JGroups XML configuration. |
| cluster.protocol.type | UDP | no | The network protocol, either UDP or TCP. |
| cluster.config.extended | false | no | Whether to apply the advanced settings (type of protocol and its configuration, eg. IP addresses and ports). |
| cluster.udp.bind.adr | (local host address) | no | Bind address when using UDP. |
| cluster.udp.bind.port | 7600 | no | Bind port when using UDP. |
| cluster.udp.external.use | false | no | Whether an external address and port is used for UDP. |
| cluster.udp.external.adr | (same as cluster.udp.bind.adr) | no | The external address when using UDP. |
| cluster.udp.external.port | (same as cluster.udp.bind.port) | no | The external port when using UDP. |
| cluster.udp.mcast.adr | 228.8.8.8 | no | Multicast address when using UDP. |
| cluster.udp.mcast.port | 7600 | no | Multicast port when using UDP. |
| cluster.tcp.bind.adr | (local host address) | no | Bind address when using TCP. |
| cluster.tcp.bind.port | 7600 | no | Bind port when using TCP. |
| cluster.tcp.external.use | false | no | Whether an external address and port is used for TCP. |
| cluster.tcp.external.adr | (same as cluster.tcp.bind.adr) | no | External address when using TCP. |
| cluster.tcp.external.port | (same as cluster.tcp.bind.port) | no | External port when using TCP. |
| cluster.tcp.initial.hosts | (empty list) | no | For discovering other cluster servers in the network. Required only when the protocol is set to TCP. A list of all expected cluster servers that are to be a part of the network. Multiple values are specified by using the same key on several lines. |
| cluster.discovery.timeout | 0 | no | The number of milliseconds to wait for the connection to another server of the cluster to be established successfully. When set to 0, there is no timeout limit. |
| cluster.tcp.merge3.minInterval | 1000 | no | For TCP connections: minimum time in milliseconds before sending an info message to other member in the cluster. A cluster server periodically broadcasts its address and name to so that each server can discover the other servers in the cluster group. Must not be a negative number. |
| cluster.tcp.merge3.maxInterval | 30000 | no | For TCP connections: maximum time in milliseconds before sending an info message to other member in the cluster. AA cluster server periodically broadcasts its address and name to so that each server can discover the other servers in the cluster group. Must be a number greater than 0. |
| cluster.tcp.fdall.interval | 3000 | no | For TCP connections: Interval in milliseconds at which a HEARTBEAT is sent to other servers in the cluster. The cluster uses failure detection based on simple heartbeat protocol. Must be a number greater than 0. |
| cluster.tcp.fdall.timeout | 5000 | no | For TCP connections: Timeout in milliseconds after which a cluster server is suspected to be dead or not working anymore. Must not be a negative number. |
| cluster.tcp.gms.attempts | 6 | no | For TCP connections: When a server wishes to join a cluster, the number of join attempts before we give up and become a singleton. 0 means never give up. |
| cluster.tcp.gms.timeout | 10000 | no | For TCP connections: When a server wishes to join a cluster, the join timeout in milliseconds for a single join attempt. Must not be a negative number. |
| cluster.udp.merge3.minInterval | 1000 | no | For UDP connections: minimum time in milliseconds before sending an info message to other member in the cluster. A cluster server periodically broadcasts its address and name to so that each server can discover the other servers in the cluster group. Must not be a negative number. |
| cluster.udp.merge3.maxInterval | 30000 | no | For UDP connections: maximum time in milliseconds before sending an info message to other member in the cluster. A cluster server periodically broadcasts its address and name to so that each server can discover the other servers in the cluster group. Must be a number greater than 0. |
| cluster.udp.fdall.interval | 3000 | no | For UDP connections: Interval in milliseconds at which a HEARTBEAT is sent to other servers in the cluster. The cluster uses failure detection based on simple heartbeat protocol. Must be a number greater than 0. |
| cluster.udp.fdall.timeout | 5000 | no | For UDP connections: Timeout in milliseconds after which a cluster server is suspected to be dead or not working anymore. Must not be a negative number. |
| cluster.udp.gms.attempts | 6 | no | For UDP connections: When a server wishes to join a cluster, the number of join attempts before we give up and become a singleton. 0 means never give up. |
| cluster.udp.gms.timeout | 10000 | no | For TCP connections: When a server wishes to join a cluster, the join timeout in milliseconds for a single join attempt. Must not be a negative number. |