LDAP-Based Monitoring

DS servers publish whether the server is alive and able to handle requests in the root DSE. They publish monitoring information over LDAP under the entry cn=monitor.

The following example reads all available monitoring entries:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN cn=monitor \
 "(&)"

The monitoring entries under cn=monitor reflect activity since the server started.

Many different types of metrics are exposed. For details, see LDAP Metrics Reference.

Monitor Privilege

The following example assigns the required privilege to Kirsten Vaughan’s entry to read monitoring data, and shows monitoring information for the backend holding Example.com data:

$ ldapmodify \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=admin \
 --bindPassword password << EOF
dn: uid=kvaughan,ou=People,dc=example,dc=com
changetype: modify
add: ds-privilege-name
ds-privilege-name: monitor-read
EOF

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=kvaughan,ou=People,dc=example,dc=com \
 --bindPassword bribery \
 --baseDN cn=monitor \
 "(ds-cfg-backend-id=dsEvaluation)"

dn: ds-cfg-backend-id=dsEvaluation,cn=backends,cn=monitor
ds-mon-backend-is-private: false
ds-mon-backend-entry-count: <count>
ds-mon-backend-writability-mode: enabled
ds-mon-backend-degraded-index-count: <count>
ds-mon-backend-ttl-is-running: <boolean>
ds-mon-backend-ttl-last-run-time: <timestamp>
ds-mon-backend-ttl-thread-count: <count>
ds-mon-backend-ttl-queue-size: <size>
ds-mon-backend-ttl-entries-deleted: <summary>
ds-mon-backend-filter-use-start-time: <timestamp>
ds-mon-backend-filter-use-indexed: <count>
ds-mon-backend-filter-use-unindexed: <count>
ds-mon-db-version: <version>
ds-mon-db-cache-evict-internal-nodes-count: <count>
ds-mon-db-cache-evict-leaf-nodes-count: <count>
ds-mon-db-cache-total-tries-internal-nodes: <count>
ds-mon-db-cache-total-tries-leaf-nodes: <count>
ds-mon-db-cache-misses-internal-nodes: <count>
ds-mon-db-cache-misses-leaf-nodes: <count>
ds-mon-db-cache-size-active: <size>
ds-mon-db-log-size-active: <size>
ds-mon-db-log-cleaner-file-deletion-count: <count>
ds-mon-db-log-utilization-min: <percentage>
ds-mon-db-log-utilization-max: <percentage>
ds-mon-db-log-size-total: <size>
ds-mon-db-log-files-open: <count>
ds-mon-db-log-files-opened: <count>
ds-mon-db-checkpoint-count: <count>
objectClass: top
objectClass: ds-monitor
objectClass: ds-monitor-backend
objectClass: ds-monitor-backend-pluggable
objectClass: ds-monitor-backend-db
ds-cfg-backend-id: dsEvaluation

Server Health (LDAP)

Anonymous clients can monitor the health status of the DS server by reading the alive attribute of the root DSE:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --baseDN "" \
 --searchScope base \
 "(&)" \
 alive

dn:
alive: true

When alive is true, the server’s internal tests have not found any errors requiring administrative action. When it is false, fix the errors and either restart or replace the server.

If the server returns false for this attribute, get error information, as described in Server Health Details (LDAP).

Server Health Details (LDAP)

The default monitor user can check whether the server is alive and able to handle requests on cn=health status,cn=monitor:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN "cn=health status,cn=monitor" \
 --searchScope base \
 "(&)"

dn: cn=health status,cn=monitor
ds-mon-alive: true
ds-mon-healthy: true
objectClass: top
objectClass: ds-monitor
objectClass: ds-monitor-health-status
cn: health status

When the server is either not alive or not able to handle requests, this entry includes error diagnostics as strings on the ds-mon-alive-errors and ds-mon-healthy-errors attributes.

Replication Delay (LDAP)

The following example uses the default monitor user account to check the delay in replication:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN cn=monitor \
 "(ds-mon-receive-delay=*)" \
 ds-mon-receive-delay

dn: ds-mon-domain-name=dc=example\,dc=com,cn=replicas,cn=replication,cn=monitor
ds-mon-receive-delay: <delay>

dn: ds-mon-server-id=<id>,cn=remote replicas,ds-mon-domain-name=dc=example\,dc=com,cn=replicas,cn=replication,cn=monitor
ds-mon-receive-delay: <delay>

DS replicas measure replication delay as the local delay when receiving and replaying changes. A replica calculates these local delays based on changes received from other replicas. Therefore, a replica can only calculate delays based on changes it has received. Network outages cause inaccuracy in delay metrics.

A replica calculates delay metrics based on times reflecting the following events:

t₀: the remote replica records the change in its data
t₁: the remote replica sends the change to a replica server
t₂: the local replica receives the change from a replica server
t₃: the local replica applies the change to its data

This figure illustrates when these events occur:

Replication keeps track of changes using change sequence numbers (CSNs), opaque and unique identifiers for each change that indicate when and where each change first occurred. The t_n values are CSNs.

When the CSNs for the last change received and the last change replayed are identical, the replica has applied all the changes it has received. In this case, there is no known delay. The receive and replay delay metrics are set to 0 (zero).

When the last received and last replayed CSNs differ:

Receive delay is set to the time t₂ - t₀ for the last change received.

Another name for receive delay is current delay.
Replay delay is approximately t₃ - t₂ for the last change replayed. In other words, it is an approximation of how long it took the last change to be replayed.

As long as replication delay tends toward zero regularly and over the long term, temporary spikes and increases in delay measurements are normal. When all replicas remain connected and yet replication delay remains high and increases over the long term, the high replication delay indicates a problem. Steadily high and increasing replication delay shows that replication is not converging, and the service is failing to achieve eventual consistency.

For a current snapshot of replication delays, you can also use the dsrepl status command. For details, see Replication Status.

Replication Status (LDAP)

The following example uses the default monitor user account to check the replication status of the local replica:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN cn=monitor \
 "(ds-mon-status=*)" \
 ds-mon-status

dn: ds-mon-domain-name=dc=example\,dc=com,cn=replicas,cn=replication,cn=monitor
ds-mon-status: Normal

If the status is not Normal, how you react depends on the value of the ds-mon-status attribute:

Status Explanation Actions to Take

Status	Explanation	Actions to Take
`Bad generation id`	Replication is broken. Internally, DS replicas store a shorthand form of the initial state called a generation ID. The generation ID is a hash of the first 1000 entries in a backend. If the replicas' generation IDs match, the servers can replicate data without user intervention. If the replicas' generation IDs do not match for a given backend, you must manually initialize replication between them to force the same initial state on all replicas. This status arises for one of the following reasons: The replica and the replication server have different generation IDs for the data because the replica was not initialized with the same data as its peer replicas. The replica has fallen further behind the replication server than allowed by the replication-purge-delay. In other words, the replica is missing too many changes, and lacks the historical information required to synchronize with peer replicas. The fractional replication configuration for this replica does not match the backend data. For example, you reconfigured fractional replication to include or exclude different attributes, or you configured fractional replication in an incompatible way on different peer replicas.	Whenever you see this status: If fractional replication is configured, make sure the configuration is compatible on all peer replicas. For details, see Fractional Replication. Reinitialize replication to fix the bad generation IDs. For details, see Manual Initialization.
`Degraded`	Unless this status is persistent, replication is operating normally. The replica has fallen further behind peer replicas than the degraded-status-threshold. By default, the threshold is 5000, meaning this state is triggered if the replica falls 5000 or more changes behind. Additionally, the number of pending changes to apply is an approximation calculated internally using change sequence numbers that are not necessarily sequential. This status can arise periodically during normal operation when, for example, replication absorbs a burst of updates. In a directory service that sustains 5000 updates a second, a temporary `Degraded` status can represent a one-second delay.	If the `Degraded` status persists: Make sure peer replica systems are sized appropriately. If some replicas are on more powerful systems with faster I/O than others, the replicas on the smaller systems can fall behind as load increases. Consider raising the `degraded-status-threshold` setting.
`Full update`	Replication is operating normally. You have chosen to initialize replication over the network. The time to complete the operation depends on the network bandwidth and volume of data to synchronize.	Monitor the server output and wait for initialization to complete.
`Invalid`	This status arises for one of the following reasons: The replica has encountered a replication protocol error. This status can arise due to faulty network communication between the replica and the replication server. The replica has just started, and is initializing.	If this status happens during normal operation: Review the replica and replication server error logs, described in About Logs, for network-related replication error messages. Independently verify network communication between the replica and the replication server systems.
`Normal`	Replication is operating normally.	Nothing to do.
`Not connected`	This status arises for one of the following reasons: The replica has just started and is not yet connected to the replication server. The replica cannot connect to a replication server.	If this status happens during normal operation: Review the replica and replication server error logs for network-related replication error messages. Independently verify network communication between the replica and the replication server systems.

Bad generation id

Replication is broken.

Internally, DS replicas store a shorthand form of the initial state called a generation ID. The generation ID is a hash of the first 1000 entries in a backend. If the replicas' generation IDs match, the servers can replicate data without user intervention. If the replicas' generation IDs do not match for a given backend, you must manually initialize replication between them to force the same initial state on all replicas.

This status arises for one of the following reasons:

The replica and the replication server have different generation IDs for the data because the replica was not initialized with the same data as its peer replicas.
The replica has fallen further behind the replication server than allowed by the replication-purge-delay. In other words, the replica is missing too many changes, and lacks the historical information required to synchronize with peer replicas.
The fractional replication configuration for this replica does not match the backend data. For example, you reconfigured fractional replication to include or exclude different attributes, or you configured fractional replication in an incompatible way on different peer replicas.

Whenever you see this status:

If fractional replication is configured, make sure the configuration is compatible on all peer replicas.

For details, see Fractional Replication.
Reinitialize replication to fix the bad generation IDs.

For details, see Manual Initialization.

Degraded

Unless this status is persistent, replication is operating normally.

The replica has fallen further behind peer replicas than the degraded-status-threshold. By default, the threshold is 5000, meaning this state is triggered if the replica falls 5000 or more changes behind. Additionally, the number of pending changes to apply is an approximation calculated internally using change sequence numbers that are not necessarily sequential.

This status can arise periodically during normal operation when, for example, replication absorbs a burst of updates. In a directory service that sustains 5000 updates a second, a temporary Degraded status can represent a one-second delay.

If the Degraded status persists:

Make sure peer replica systems are sized appropriately. If some replicas are on more powerful systems with faster I/O than others, the replicas on the smaller systems can fall behind as load increases.
Consider raising the degraded-status-threshold setting.

Full update

Replication is operating normally.

You have chosen to initialize replication over the network.

The time to complete the operation depends on the network bandwidth and volume of data to synchronize.

Monitor the server output and wait for initialization to complete.

Invalid

This status arises for one of the following reasons:

The replica has encountered a replication protocol error. This status can arise due to faulty network communication between the replica and the replication server.
The replica has just started, and is initializing.

If this status happens during normal operation:

Review the replica and replication server error logs, described in About Logs, for network-related replication error messages.
Independently verify network communication between the replica and the replication server systems.

Normal

Replication is operating normally.

Nothing to do.

Not connected

This status arises for one of the following reasons:

The replica has just started and is not yet connected to the replication server.
The replica cannot connect to a replication server.

If this status happens during normal operation:

Review the replica and replication server error logs for network-related replication error messages.
Independently verify network communication between the replica and the replication server systems.

Request Statistics (LDAP)

DS server connection handlers respond to client requests. The following example uses the default monitor user account to read statistics about client operations on each of the available connection handlers:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN "cn=connection handlers,cn=monitor" \
 "(&)"

For details about the content of metrics returned, see Metric Types Reference.

Work Queue (LDAP)

DS servers have a work queue to track request processing by worker threads, and whether the server has rejected any requests due to a full queue. If enough worker threads are available, then no requests are rejected. The following example uses the default monitor user account to read statistics about the work queue:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN "cn=work queue,cn=monitor" \
 "(&)"

For details about the content of metrics returned, see Metric Types Reference. To adjust the number of worker threads, see the settings for Traditional Work Queue.

Database Size (LDAP)

DS servers maintain counts of the number of entries in each backend and under each base DN. The following example uses the default monitor user account to read the counts:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN cn=monitor \
 "(|(ds-mon-backend-entry-count=*)(ds-mon-base-dn-entry-count=*))" \
 ds-mon-backend-entry-count ds-mon-base-dn-entry-count

Active Users (LDAP)

DS server connection handlers respond to client requests. The following example uses the default monitor user account to read the metrics about active connections on each connection handler:

$ ldapsearch \
 --hostname localhost \
 --port 1636 \
 --useSsl \
 --usePkcs12TrustStore /path/to/opendj/config/keystore \
 --trustStorePassword:file /path/to/opendj/config/keystore.pin \
 --bindDN uid=monitor \
 --bindPassword password \
 --baseDN cn=monitor \
 "(objectClass=ds-monitor-connection*)" \
 ds-mon-active-connections-count ds-mon-active-persistent-searches ds-mon-connection ds-mon-listen-address

For details about the content of metrics returned, see Metric Types Reference.

DS 7.1.8

LDAP-Based Monitoring

Monitor Privilege

Server Health (LDAP)

Server Health Details (LDAP)

Replication Delay (LDAP)

Replication Status (LDAP)

Request Statistics (LDAP)

Work Queue (LDAP)

Database Size (LDAP)

Active Users (LDAP)