MDM Monitoring Metrics
The following article provides a comprehensive list of available monitoring metrics for MDM, organized by functional categories.
System and infrastructure metrics
gRPC server metrics
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The duration of incoming gRPC calls, grouped by |
|
Summary |
The duration of incoming gRPC calls, grouped by |
|
Counter |
The number of received gRPC streaming messages on the server side, grouped by |
|
Counter |
Total number of exceptions on gRPC request processing. |
gRPC client metrics
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The number of active clients in the gRPC pool. |
|
Gauge |
The duration of outgoing gRPC calls, grouped by |
|
Counter |
The number of outgoing gRPC calls, grouped by |
|
Gauge |
The duration of outgoing gRPC calls, grouped by |
|
Counter |
The number of sent gRPC streaming messages on the client side, grouped by |
HTTP server metrics
Metric | Metric type | Description |
---|---|---|
|
Counter |
Total number of exceptions on HTTP request processing. |
|
Summary |
The duration of incoming HTTP calls, grouped by |
|
Counter |
Total number of incoming http calls, grouped by |
|
Gauge |
Total duration of incoming HTTP calls, grouped by |
|
Gauge |
Maximum time spent in incoming HTTP call processing, grouped by |
JVM and process metrics
Metric | Metric type | Description |
---|---|---|
|
Gauge |
An estimate of the total capacity of the buffers in this pool. |
|
Gauge |
Size of long-lived heap memory pool after reclamation. |
|
Summary |
Time spent in GC pause. |
|
Gauge |
Maximum time spent in GC pause. |
|
Gauge |
The amount of memory used. |
|
Counter |
Incremented for an increase in the size of the (young) heap memory pool after one GC until the next. |
|
Gauge |
The uptime of the Java Virtual Machine (JVM). |
|
Gauge |
The recent CPU usage for the JVM process. |
|
Gauge |
The number of processors available to the JVM. |
Tomcat metrics
Metric | Metric type | Description |
---|---|---|
|
Gauge |
Maximum duration of an alive Tomcat session. |
|
Gauge |
Current number of active Tomcat sessions. |
MDA service metrics
Metric | Metric type | Description |
---|---|---|
|
Counter |
Total number of calls to the MDM Web App layer, grouped by |
|
Gauge |
Total duration of calls to the MDM Web App layer, grouped by |
|
Gauge |
Maximum time spent in calls to the MDM Web App layer, grouped by |
MDM business metrics
Data access operations
Data access metrics track record processing across four categories: batch operations, publish actions from the web application, online services (REST, SOAP), and streaming platforms (such as Kafka or SQS). Each category contains metrics for input records and final changes categorized by operation type (Insert, Update, Delete).
Batch operations
Metric | Metric type | Description |
---|---|---|
|
Counter |
The number of records that were ever inputted to any batch load operation. |
|
Counter |
The number of records that were changed by a batch load operation.
The |
Publish operations
This category covers publish operations that a user initiated from the web application.
Metric | Metric type | Description |
---|---|---|
|
Counter |
The number of records inputted to publish actions from the web application. |
|
Counter |
The number of changed records by publish actions from the web application.
The |
Service (REST, SOAP) operations
Metric | Metric type | Description |
---|---|---|
|
Counter |
The number of records on input for publish actions from online services (for example, REST, SOAP). |
|
Counter |
The number of changed records by publish actions from online services.
The |
Streaming operations
This category covers record processing executed on streaming platforms like Kafka or SQS.
Metric | Metric type | Description |
---|---|---|
|
Counter |
The number of records that were ever inputted to any streaming operation. |
|
Counter |
The number of records changed by streaming operations.
The |
User activity and configuration
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The number of configured MDM Web App users, categorized by permission type.
The |
|
Counter, Gauge |
The total number of users with activity in MDM Web App.
The |
|
Counter, Gauge |
The number of users active in the last hour in MDM Web App.
The |
|
Counter, Gauge |
The number of users active in the last 24 hours in MDM Web App.
The |
Data model structure
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The total number of records in the system.
The |
|
Gauge |
The number of records in the largest entity.
The |
|
Gauge |
The total number of entities in the data model.
The |
|
Gauge |
The total number of relationships in the data model. |
|
Gauge |
The number of columns in the logical model for instance entities.
The |
|
Gauge |
The number of columns in the logical model for master entities.
The |
|
Gauge |
The number of history collectors used in instance entities. |
|
Gauge |
The number of instance entities linked to multiple master entities. |
|
Gauge |
The number of instance entities with multiple matching columns. |
|
Gauge |
The number of master entities configured with matching criteria. |
Task and draft management
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The total number of drafts and historical drafts in MDM Web App.
The |
|
Gauge |
The total number of tasks in MDM Web App.
The |
|
Gauge |
The daily number of tasks created, resolved, or closed in MDM Web App.
The |
Performance and monitoring
Startup performance
Startup metrics measure the duration of various application startup phases.
The ataccama_one_mdm_startup_total_seconds_sum
metric represents the total startup time from JVM initiation to the application context being fully started.
All startup metrics are Prometheus summary types, providing _count
, _sum
(total time in seconds), and _max
(maximum time in seconds) values.
They all include a startup_ts="<TIMESTAMP>"
label, which records the JVM startup timestamp.
Metric | Metric type | Description |
---|---|---|
|
Summary |
Measures the total startup time from the JVM startup until the application context is started.
Labels: |
|
Summary |
Measures the time taken for the application context to become active during startup.
Labels: |
|
Summary |
Measures the time taken for the application context startup phase.
This includes initialization of beans and other context-related setup.
Labels: |
|
Summary |
Measures the JVM startup time, from the beginning of JVM initialization to the start of the root context.
Labels: |
|
Summary |
Measures the time taken for the management context to become active during startup.
Labels: |
|
Summary |
Measures the time taken for the management context startup phase.
Labels: |
|
Summary |
Measures the startup time of the root context.
Labels: |
Data Quality engine threadpool monitoring
Metrics for monitoring the Data Quality engine thread pool. These allow tracking of thread creation and completion, enabling calculation of currently active threads.
The difference between created
and finished
counts indicates the number of currently executing threads.
In versions prior to 16.1.0, this metric was available as dqc_threadpool_threads_total and dqc_threadpool_threads_finished_total .
|
Metric | Metric type | Description |
---|---|---|
|
Counter |
The total number of threads in the DQ engine, distinguished by their lifecycle stage.
The |
Event handling
Metric | Metric type | Description |
---|---|---|
|
Gauge |
The total number of event handlers in the MDM system. |
|
Gauge |
The number of event handlers that function as publishers in the system. |
System configuration
Metric | Metric type | Description |
---|---|---|
|
Gauge |
Indicates whether VLDB persistence is enabled for a specific instance. |
Was this page useful?