Activity Explorer
Activity Explorer shows runtime database activity — sessions, connections, waits, blocking, and transactions. It is built for operators who need to understand what is happening inside a database while a workload is running.
What it answers
- Which databases currently have active workload?
- Which sessions or connections are consuming time?
- Are users waiting on locks, I/O, client activity, or other engine wait classes?
- Is one session blocking other sessions, and did deadlocks or long waits increase compared with the previous period?
Scope
Detailed activity views cover PostgreSQL, SQL Server, and Oracle. MongoDB, Redis, and Valkey are covered by their monitoring, alerting, and explorer workflows instead. Query-level performance history belongs in Query Explorer.
Reading it
Every metric uses the selected time window and compares against the previous period of the same length. An upward trend is not always bad — rising throughput can simply mean more workload. Rising deadlocks, wait time, blocked sessions, or blocking duration usually deserves review.
Supported Engines
Activity Explorer currently provides detailed activity views for:
| Engine | Activity scope |
|---|---|
| PostgreSQL | Sessions, connection states, wait events, blocking chains, transactions, deadlocks, query duration, and resource context. |
| Microsoft SQL Server | Sessions, connection states, waits, blocking chains, transactions, deadlocks, CPU/read/write session metadata, and resource context. |
| Oracle | Sessions, wait events, blocking chains, SQL execution rate, Oracle session serial, SQL identifiers, wait classes, and resource context. |
MongoDB, Redis, and Valkey continue to be covered by their supported monitoring, alerting, health, and explorer workflows, but they are not part of the current Activity Explorer tab model.
Activity Database List
Every database that can be reviewed for activity, with its hosting instance, engine, state, and the session pressure it is under in the selected time window.
Columns
- Database / Instance — The database and the instance hosting it. Opening a row leads to the detail tabs: Overview, Connections, Wait Events, and Blocking Chains.
- Engine — PostgreSQL, SQL Server, or Oracle for detailed activity views.
- Status — Active means the target is currently reporting; inactive targets stay listed with their last known information.
- Sessions — Total sessions observed in the window.
- Active — Sessions actively executing work in the engine.
- Blocked — Sessions waiting on locks. The strongest column to sort by during an incident.
- Alerts — Critical and warning alert counts for the database.
Reading it
Sort by Blocked when triaging an incident, then narrow with the engine, instance, or status filters. The search box matches database and instance names.
If a database shows no data
- No activity in the selected time window.
- The monitoring user lacks access to the engine's activity views.
- High-frequency or workload collectors are disabled.
- The database target is inactive.
Detail Tabs
Selecting a database opens the Activity Explorer detail view. The detail view uses five tabs; the last one appears only for Oracle:
| Tab | Purpose |
|---|---|
| Overview | High-level activity metrics, recent activity rows, and resource utilization. |
| Connections | Current sessions or connections, connection state distribution, and connection-level details. |
| Wait Events | Current wait events, wait type distribution, and wait details. |
| Blocking Chains | Root blockers, blocked sessions, chain depth, wait duration, and deadlock-related context. |
| Wait Activity (Oracle only) | ASH-driven wait analysis: active session time, CPU vs wait split, top wait classes, events, and SQL. |
Each tab uses the selected time window to calculate trends and compare current values with the previous comparable period.
Overview
The Overview tab summarizes activity across the selected database: active sessions, transaction rate (PostgreSQL and SQL Server), deadlocks, average query time, the server's resource utilization, and a recent-activities list.
Metric definitions and how each value is calculated live in the Overview reference.
Connections
The Connections tab shows the sessions connected to the database: totals split by state, a connection-state history, and a per-connection table with an in-place detail view.
Metric definitions, the shared state model, and how each value is calculated live in the Connections reference.
Engine specifics: Oracle connection details can include session serial, SQL identifier, and last-call timing. Microsoft SQL Server can include host name, CPU time, reads, writes, and open transaction count. PostgreSQL rows represent query-bearing client backends.
Wait Events
The Wait Events tab focuses on sessions waiting for database resources: wait totals by category, a category history, and a per-wait table with an in-place detail view.
Metric definitions, the per-engine category mapping, and how each value is calculated live in the Wait Events reference.
Engine specifics: Oracle also exposes its native wait class. Microsoft SQL Server can expose wait resource and host name. PostgreSQL exposes its native wait event type and wait event name.
Blocking Chains
The Blocking Chains tab shows lock chains where one session blocks one or more other sessions: the root blocker, everything queued behind it, chain depth, durations, and deadlock context.
Metric definitions, worked examples, and how each value is calculated live in the Blocking Chains reference.
Engine specifics: Oracle blocking details can include lock type, lock mode, object owner, object name, and session serial. Microsoft SQL Server can include host name and wait resource. PostgreSQL uses PostgreSQL session, lock, and wait metadata.
Wait Activity (Oracle)
The Wait Activity tab appears only for Oracle databases. It analyzes Oracle's Active Session History — active session time, the CPU vs wait split, and the top wait classes, wait events, and SQL statements for the selected window. ASH requires the Oracle Diagnostics Pack.
Metric definitions and how each value is derived from ASH samples live in the Wait Activity reference.
Filtering, Sorting, And Pagination
Activity Explorer supports pagination and sorting across detail tables.
Connections can be filtered by connection state and sorted by duration, session identifier, user, or database.
Wait events can be filtered by wait event type and minimum duration, then sorted by duration, session identifier, event type, or event name.
Blocking chains can be filtered by minimum blocked session count and minimum duration, then sorted by duration, chain depth, blocked count, or root blocker session.
Status And Trends
Activity Explorer metrics can include:
- Current value.
- Change percentage.
- Trend direction.
- Status band.
Status bands are used to highlight operational pressure. They are based on runtime thresholds for the specific metric, such as active sessions, transaction or execution rate, deadlocks, wait counts, blocked processes, chain depth, and wait duration.
An upward trend is not always bad. For throughput metrics, an increase can indicate higher workload. For deadlocks, wait time, blocked sessions, or blocking duration, an increase usually deserves review.
Data Freshness
Activity Explorer uses collected runtime signals for the selected time window. Activity data is most useful when high-frequency and workload collectors are enabled and the monitoring user has access to the required engine activity views.
If activity data is unavailable, the database can still appear in the list with a no-data reason. Common causes include no activity in the selected time window, missing permissions, unsupported engine activity surfaces, disabled collectors, or an inactive database target.
Data Boundaries
Activity Explorer displays operational metadata. This can include query text or snippets, user names, application names, client addresses, session identifiers, wait resources, lock metadata, object names, and timing data.
Logstag does not copy application table rows through Activity Explorer. However, runtime metadata can still reveal sensitive workload behavior. Activity Explorer is intended for users who are allowed to inspect live database activity and operational metadata.