Get occurrences metrics over a span of time by filtering, grouping, and aggregating. The endpoint follows search/query semantics as a POST request using a Project or Account Access Token with read scope.

The account must have access to the Metrics API, which is part of the Advanced package. Other accounts receive a 403.

Projects queried

• Project Access Token: the query always runs against the token's own project. Any project_id filter in the request is ignored.
• Account Access Token: without a project_id filter, every project the token can read is queried. With one, projects the token can't read are dropped from the filter, and if none are left the response is a 403.

Time range and timepoints

start_time and end_time are optional. A missing start_time, or one earlier than the account's data retention, starts the query at the retention boundary. A missing end_time leaves the range open-ended.

With a granularity, results are bucketed into one timepoint per period. Without one, all rows are returned in a single timepoint whose timestamp is the time of the latest occurrence.

Columns

These columns can be used in filters, group_by, aggregates, and sort:

• project_id
• item_id
• environment
• browser_family
• browser_version
• os_family
• os_version
• device_brand
• device_model
• ip_address
• item_status
• item_level
• item_group_item_id
• item_title
• item_counter
• person_username
• person_email
• person_id
• code_version
• count
• occurrence_id
• uuid
• context
• platform
• framework
• platform_canonical
• framework_canonical
• language
• language_name
• notifier_name
• notifier_version
• occurrence_count
• message_body
• timestamp
• fingerprint
• server_host
• server_root
• server_pid
• server_cpu
• scm_branch
• request_url
• request_method
• request_query_string
• request_body
• js_client_settings
• exception_class
• exception_message
• exception_description
• traceframe_stack_level
• traceframe_index
• traceframe_reverse_index
• traceframe_lineno
• traceframe_colno
• traceframe_filename
• traceframe_method
• traceframe_class_name
• traceframe_in_project
• telemetry_level
• telemetry_type
• telemetry_source
• telemetry_timestamp
• telemetry_body_subtype
• telemetry_body_method
• telemetry_body_url
• telemetry_body_message
• telemetry_body_element
• telemetry_body_from
• telemetry_body_to
• telemetry_body_status_code
• telemetry_body_start_timestamp
• telemetry_body_end_timestamp
• replay_id

Key/value columns

These columns hold lists of name/value pairs:

• request_headers
• request_params
• request_gets
• request_posts
• customs
• traceframe_locals
• attributes

In filters and aggregates, set key to the name of the entry to use, for example {"field": "request_headers", "key": "User-Agent", "operator": "eq", "values": ["curl/8.0"]}. When the entry's value is JSON, text after the first unescaped . in key is read as a JSONPath into it (for example payload.user.id); escape a literal dot in the name as \.. Each of these columns also has _name and _value variants (for example request_headers_name) that address every name or every value in the list.

Filter operators

• eq and ne match any of the listed values and accept more than one value.
• gt, gte, lt, and lte take one value.
• like and ilike (case-insensitive) take one SQL LIKE pattern; use % as the wildcard.
• is_null and not_null take no values.

item_status and item_level are filtered by name, for example active or error.

A filter whose field is an aggregate alias is applied to the aggregated result, for example {"field": "person_count", "operator": "gt", "values": [10]}.