---
url: https://simpledatalogs.com/docs/guide/observability.md
description: >-
  Requests, jobs, scheduled tasks, slow queries, outgoing calls, traces and
  grouped exceptions for your app, next to its logs and events.
---

# App and Issues

Once your app sends telemetry, two tabs appear on its environment: **App** and **Issues**.
With the [Laravel SDK](/sdks/laravel#telemetry) that happens as soon as a key is set; other stacks can send [telemetry objects](/api/telemetry) directly.

## App

| Section | What it shows |
|---|---|
| Overview | Requests, p95, error rate, exceptions, failed jobs and slow queries, each with its change; throughput and latency (p50, p95, p99) against the previous period; slowest routes, top issues, busiest queries, failing jobs, outgoing hosts |
| Requests | Every route pattern with count, p95, error rate and trend; a route opens its own page |
| Jobs · Schedule · Commands | Runs, failures and durations per job class, task and command; Schedule marks a task **Late** when its last run is older than one and a half intervals |
| Queries | Normalized SQL with count, total and slowest time |
| Outgoing | Calls by host, then by `METHOD host` |
| Traces | Search by route or job, status, duration, errors, user or trace id; a row opens the waterfall |

Sections appear only when your app sends that kind of work.
p95 and the other percentiles are within about 1 %; counts are exact.

## Traces

A trace shows one request or job as a waterfall: every query, outgoing call, cache step and mail in time order, with repeated calls folded as `×37`.
A job dispatched by a request shows in the same trace, so you can follow a checkout from the click to the invoice email.
**Open logs** lists the log lines written during it.

## Issues

Exceptions with the same cause group into one issue. They come from [telemetry objects](/api/telemetry#exceptions), and from errors sent through [Sentry SDKs](/integrations/sentry) or OpenTelemetry logs with an `exception.type`.

* Filter by Open, Resolved, Ignored, Regressions and Assigned to me; search and sort by last seen, occurrences or users.
* **Assign**, **Resolve** (with a note, or "in next release"), **Ignore** and **Reopen**, one at a time or in bulk. Viewers can read but not change issues.
* A resolved issue that happens again in a newer release reopens as a **regression**.
* The issue page shows the stack trace with your own frames in bold, occurrences over time, breakdowns by route, release, OS, browser and country, and each occurrence's trace.

## What it costs

A request with everything inside it is usually one event. See [What counts toward your plan](/api/telemetry#what-counts-toward-your-plan).