publish updates to Directory Servers within a Replication Domain. ds-cfg-replication-server top Specifies the addresses of other to which this tries to connect at startup time. Addresses must be specified using the syntax: "hostname:port". If IPv6 addresses are used as the hostname, they must be specified using the syntax "[IPv6Address]:port". ^.+:[0-9]+$ HOST:PORT A host name followed by a ":" and a port number. ds-cfg-replication-server Specifies a unique identifier for the . Each must have a different server ID. ds-cfg-replication-server-id Specifies the window size that the uses when communicating with other . This option may be deprecated and removed in future releases. 100000 ds-cfg-window-size Specifies the number of changes that are kept in memory for each directory server in the Replication Domain. 10000 ds-cfg-queue-size The path where the stores all persistent information. changelogDb ds-cfg-replication-db-directory The database implementation that stores all persistent information. je Implementation based on Berkeley DB JE database. Implementation based on log file. ds-cfg-replication-db-implementation The time (in seconds) after which the erases all persistent information. 3 days ds-cfg-replication-purge-delay The port on which this waits for connections from other or Directory Servers. ds-cfg-replication-port The group id for the replication server. This value defines the group id of the replication server. The replication system of a LDAP server uses the group id of the replicated domain and tries to connect, if possible, to a replication with the same group id. 1 ds-cfg-group-id The timeout value when waiting for assured mode acknowledgments. Defines the number of milliseconds that the replication server will wait for assured acknowledgments (in either Safe Data or Safe Read assured sub modes) before forgetting them and answer to the entity that sent an update and is waiting for acknowledgment. 1000ms ds-cfg-assured-timeout The number of pending changes as threshold value for putting a directory server in degraded status. This value represents a number of pending changes a replication server has in queue for sending to a directory server. Once this value is crossed, the matching directory server goes in degraded status. When number of pending changes goes back under this value, the directory server is put back in normal status. 0 means status analyzer is disabled and directory servers are never put in degraded status. 5000 ds-cfg-degraded-status-threshold The weight of the replication server. The weight affected to the replication server. Each replication server of the topology has a weight. When combined together, the weights of the replication servers of a same group can be translated to a percentage that determines the quantity of directory servers of the topology that should be connected to a replication server. For instance imagine a topology with 3 replication servers (with the same group id) with the following weights: RS1=1, RS2=1, RS3=2. This means that RS1 should have 25% of the directory servers connected in the topology, RS2 25%, and RS3 50%. This may be useful if the replication servers of the topology have a different power and one wants to spread the load between the replication servers according to their power. 1 ds-cfg-weight The period between sending of monitoring messages. Defines the duration that the replication server will wait before sending new monitoring messages to its peers (replication servers and directory servers). Larger values increase the length of time it takes for a directory server to detect and switch to a more suitable replication server, whereas smaller values increase the amount of background network traffic. 60s ds-cfg-monitoring-period Whether the replication server will compute change numbers. This boolean tells the replication server to compute change numbers for each replicated change by maintaining a change number index database. Changenumbers are computed according to http://tools.ietf.org/html/draft-good-ldap-changelog-04. Note this functionality has an impact on CPU, disk accesses and storage. If changenumbers are not required, it is advisable to set this value to false. true ds-cfg-compute-change-number