The following table provides a description of the attributes in the cn=Replication Summary monitor entry.

dn: cn=Replication Summary <baseDN>,cn=monitor
Monitor Attribute Description
base-dn:<baseDN> Base DN summary.
replica: replica-id, ldap-server connected-to, generation-id, replication-backlog, recent-update-rate, peak-update-rate, age-of-oldest- backlog-change Summary information for each replica in the topology. This entry appears for each replica in the topology with its own respective properties.
replication-server: server-id, server, generation-id, status, last-connected, last-failed, failed-attempts, attributes. Summary information for each remote replication server only in the topology. This entry appears for each Replication Server in the topology with its own respective serverID and server properties.
update-queue: id, max-count, current-count, max-size, current-size, polling-source, polling-source-changed Summary information for each update queue on a server:
  • id. The ID of the receiving replica or replication server
  • max-count. The maximum number of update messages that the sending replication server will keep in memory for the receiving replica or replication server. If the receiver cannot accept messages fast enough for any reason (high load, network latency, etc), then this queue will fill up. When that happens, the sending replication server will read update messages from the changelog backend. This slows down the update processing considerably.
  • current-count. The number of update messages currently on the queue that have not been sent to the receiving replica or replication server. Every time the sending replication server sends an update to the receiving replica or replication server, this counter is decremented.
  • max-size. The maximum total size (in bytes) of update messages that may be in the queue. This queue is capped by both the maximum count (max-count) and the max-size setting, whichever is reached first.
  • current-size. The total size of update messages currently on the queue that have not been sent to the receiving replica or replication server. Every time the sending replication server sends an update to the receiving replica or replication server, this value is decremented by the size of the published update message.
  • polling-source. Either 'memory' or 'db'. If set to 'memory', the sending replication server relies only on the in-memory queue to push update messages to the receiving replica or replication server. If set to 'db', update messages are read and sorted from the changelog database, which is significantly slower than publishing updates from the in-memory queue.
  • polling-source-changed. The total number of times the polling-source attribute has changed value (either from 'db' to 'memory' or from 'memory' to 'db'). If this value changes very frequently, then the queue size setting is probably too low.