title: “Executions”
description: “Learn what an execution is and how Foveus uses executions to explain application behavior.”
Executions
An execution is one run of work in your application.
For an API service, an execution is usually one HTTP request.
For a worker, an execution may eventually represent one job, message, scheduled task, or background operation. Worker support is evolving, so HTTP request capture is currently the most complete execution boundary.
Foveus uses executions to show what happened, when it happened, how long it took, and what evidence was captured.
Why executions matter
Logs tell you what was written.
Metrics tell you what changed.
Traces show request flow.
Executions show the actual unit of work you need to inspect.
An execution can include:
* service name
* endpoint or operation
* request context
* response context
* duration
* outcome
* logs
* failure details
* linked issue, if Foveus grouped it into one
Execution examples
An HTTP API request:
Each one can be represented as an execution model, but support varies by source.
HTTP execution capture is currently the most complete execution boundary. Worker services can send logs, exceptions, outbound HTTP telemetry, metrics, and trace context today, but full automatic per-job execution capture for arbitrary worker loops is still evolving.
What Foveus captures
The exact data depends on your SDK configuration and service type.
A typical HTTP execution can include:
| Field | Description |
|---|
service | The service that produced the execution. |
environment | The deployment environment, such as development, staging, or production. |
mode | The Foveus data mode: test or live. |
endpoint | The HTTP path or operation name. |
method | The HTTP method, such as GET or POST. |
duration | How long the execution took. |
outcome | Whether the execution succeeded, failed, or is unknown. |
context | Safe request, response, query, or custom context fields. |
logs | Logs correlated to the execution, if configured. |
issue | The issue linked to the execution, if Foveus grouped it. |
Service name
Foveus uses the service name to group executions by logical service.
If you do not set ServiceName, Foveus uses your project assembly name.
You can override it:
builder.Services.AddFoveus(options =>
{
options.ApiKey = builder.Configuration["Foveus:ApiKey"];
options.ServiceName = "orders-api";
});
orders-api
payments-api
identity-api
billing-worker
Environment
Use Environment to describe where your service is running.
Examples:
development
test
staging
production
service:orders-api environment:production
Mode
Foveus has two modes:
| Mode | Use for |
|---|
test | Sandbox, development, QA, and onboarding |
live | Production telemetry |
test.
Set Mode only when sending live telemetry.
{
"Foveus": {
"ApiKey": "fov_live_...",
"ServiceName": "orders-api",
"Environment": "production",
"Mode": "live"
}
}
Outcomes
An execution can have an outcome such as:
| Outcome | Meaning |
|---|
success | The execution completed successfully. |
failed | The execution failed or returned a failure response. |
unknown | Foveus could not confidently classify the outcome. |
Context
Execution context is structured data captured from the request, response, query string, or custom context.
For example:
{
"customerId": "cus_12345",
"orderStatus": {
"value": 1,
"label": "Confirmed"
}
}
service:orders-api context:customerId="cus_12345"
service:orders-api context:orderStatus.value=1
Execution Context Search uses indexed execution context. It does not scan arbitrary raw request bodies.
Linked logs
If logging is configured, Foveus can show logs related to an execution.
This helps you inspect the log lines that occurred during the same unit of work instead of searching through unrelated log streams.
For worker services, logs are currently one of the most useful telemetry sources.
Worker telemetry
Foveus can be used in .NET Worker Service and BackgroundService apps.
Worker services use:
builder.Services.AddFoveus(builder.Configuration);
ILogger logs
* logged exceptions
* outbound HTTP telemetry through IHttpClientFactory
* trace and execution header propagation on outbound HTTP
* SDK shutdown flushing
* source classification as worker
* supported MassTransit instrumentation hooks
Full automatic per-job execution capture for arbitrary worker loops is not yet implemented.
See Worker services.
Linked issues
Foveus can group failed executions into issues.
An issue is a repeated or meaningful failure pattern.
For example, several failed executions on the same endpoint may become one issue.
Open the issue to see:
* failure pattern
* representative execution
* linked executions
* recent activity
* lifecycle
* comments and activity history
Search executions
The Executions page is query-first because execution data can be high-volume.
Common searches:
service:orders-api method:POST endpoint:/orders
service:orders-api status:failed
service:orders-api context:customerId="cus_12345"
service:orders-api context:orderStatus.value=1
When to use executions
Use executions when you need to:
* investigate a support complaint
* find a request by customer ID or request ID
* inspect request and response context
* understand why a request failed
* move from an issue to the exact failed execution
* correlate logs with a unit of work
For worker services, use Foveus today to inspect logs, exceptions, outbound HTTP activity, and supported messaging telemetry while full per-job execution capture evolves.
What to do next
* View your first execution: View your first execution
* Search by context: Execution Context Search
* Learn about issues: Issues
* Set up worker services: Worker services
* Investigate a user complaint: Investigate a user complaint