Skip to content

Latest commit

 

History

History
165 lines (125 loc) · 7.75 KB

query-log.md

File metadata and controls

165 lines (125 loc) · 7.75 KB
title description menu weight influxdb3/cloud-dedicated/tags related
View the query log
View and analyze queries executed on your cluster using the Admin UI Query History or by querying the _internal database with influxctl.
influxdb3_cloud_dedicated
name parent
View the query log
Troubleshoot and optimize queries
351
query
observability
admin
/influxdb3/cloud-dedicated/reference/cli/influxctl/query/
/influxdb3/cloud-dedicated/reference/admin-ui/
/influxdb3/cloud-dedicated/query-data/troubleshoot-and-optimize/system-information/

The query log records queries executed on your {{% product-name %}} cluster. Use it to monitor query performance, find slow-running queries, and troubleshoot failed executions.

Note

Query logging is not enabled by default

The query log is not enabled by default on all clusters as it can create a negligible ingest load on your cluster. To enable it for your cluster, contact InfluxData support.

Use the Admin UI or the influxctl query command to view the query log.

{{< tabs-wrapper >}} {{% tabs %}} Admin UI influxctl {{% /tabs %}} {{% tab-content %}}

Access the Cloud Dedicated Admin UI

  1. To access the {{% product-name %}} Admin UI, visit the following url in your browser:

    https://console.influxdata.com

  2. Use the credentials provided by InfluxData to log into the Admin UI. If you don't have login credentials, contact InfluxData support.

  3. Open the cluster you want to inspect and go to Query History.

If query logging is enabled for your cluster, any admin user can access the query log in the Admin UI automatically; no database token is required.

In Query History you can:

  • Search by Database Token ID to see queries run with a specific token.
  • Filter by:
    • Status (for example, success, failure)
    • Database
    • Query type (for example, SQL, InfluxQL)
    • Source (for example, User Queries, UI)
    • Time range (for example, last 24 hours)

The table lists each query with its status, query text, database, query type, duration, and timestamp. You can use the column headers to sort (for example by duration or time).

{{< img-hd src="/img/influxdb3/cloud-dedicated-admin-ui-query-log-list-view.png" alt="Query History list view in the Admin UI with search, filters, and table" />}}

You can also expand a row to see more details about that execution.

{{< img-hd src="/img/influxdb3/cloud-dedicated-admin-ui-query-log-detail.png" alt="Query History detail view in the Admin UI" />}}

{{% /tab-content %}} {{% tab-content %}}

Use the influxctl query command to run SQL against the _internal database and query_log table. Query log entries are stored in the _internal database.

  1. If you haven't already, download and install the influxctl CLI, and then configure an influxctl connection profile for your cluster.
  2. Create a database token that has read access to the _internal database. Replace {{% code-placeholder-key %}}DATABASE_TOKEN{{% /code-placeholder-key %}} in the examples below with your token.
  3. Run the query subcommand with --database and --language (and optionally --config). Global flags such as --config must come before the command; query flags such as --database, --language, and --token must come after query.

{{% code-placeholders "DATABASE_TOKEN" %}}

List recent successful queries with compute duration above a threshold (for example, 0.6 ms):

influxctl query \
  --token DATABASE_TOKEN \
  --database _internal \
  --language sql \
  'SELECT * FROM query_log WHERE success = '\''true'\'' AND compute_duration_ns > 600000 LIMIT 10'

Filter by namespace (database) and time range:

influxctl query \
  --token DATABASE_TOKEN \
  --database _internal \
  --language sql \
  'SELECT * FROM query_log WHERE namespace_name = '\''my_database'\'' AND time >= now() - INTERVAL '\''1 day'\'' LIMIT 50'

{{% /code-placeholders %}}

Example output:

| auth_id        | compute_duration_ns | phase   | query_type | query_text                                              | success | time                    |
|----------------|---------------------|---------|------------|---------------------------------------------------------|---------|--------------------------|
| token-id-xxxx  |             2314333 | success | sql        | SELECT * FROM query_log WHERE success = 'true' AND ...  | true    | 2026-02-25T00:30:30Z     |
| token-id-yyyy  |          3673637621 | success | sql        | SELECT * FROM my_measurement WHERE time > now() - ...  | true    | 2026-02-25T00:28:57Z     |
| token-id-yyyy  |          1443145654 | success | sql        | SELECT COUNT(*) FROM query_log WHERE ...                | true    | 2026-02-25T00:29:02Z     |
+----------------+---------------------+---------+------------+---------------------------------------------------------+---------+--------------------------+

{{% /tab-content %}} {{< /tabs-wrapper >}}

Query log data and columns

The query_log table in _internal includes the following columns:

Column Data type Description
time timestamp Timestamp when the query log entry was recorded
id string Unique identifier for the query
namespace_id string Internal identifier for the database
namespace_name string Name of the database where the query was executed
query_type string Type of query syntax used (sql, influxql)
query_text string The actual query statement text
query_params string Query parameters (if applicable)
auth_id string Database token ID used to authenticate the query
trace_id string Trace ID for debugging and monitoring
success string Query execution status ('true' or 'false' as string)
running string Indicates if query is currently running ('true' or 'false' as string)
phase string Current query phase (for example, received, planned, permit, success, fail, cancel)
query_issue_time_ns int64 Time when the query was issued (nanoseconds)
permit_duration_ns int64 Time spent waiting for query permit (nanoseconds)
plan_duration_ns int64 Time spent planning the query (nanoseconds)
execute_duration_ns int64 Time spent executing the query (nanoseconds)
end_to_end_duration_ns int64 Total end-to-end query duration (nanoseconds)
compute_duration_ns int64 Compute time for the query (nanoseconds)
partition_count int64 Number of partitions accessed
parquet_file_count int64 Number of Parquet files read
max_memory_bytes int64 Maximum memory used during query execution (bytes)

Note

Use string literals for status columns

In the query_log table, success and running are stored as strings ('true' or 'false'), not booleans. In SQL predicates, use string comparison—for example, success = 'true' or running = 'false'—to avoid type coercion errors.