# Issues

Track recurring system anomalies (directory-sync, provisioning, and metadata failures), grouped and worked to resolution.

An **issue** is a recurring *system* anomaly the platform raised on its own: an operational failure that happens in the background and tends to repeat. Where the [audit log](https://dts-identity.com/docs/monitoring/audit-logs.md) records individual events, issues roll many occurrences of the same underlying problem into a single entry you can track to resolution, so a directory that fails to sync every hour appears as **one** issue with a rising occurrence count rather than dozens of scattered log lines.

You find it under **Audit Trail → Issues** in the Admin Console.

Think of Issues as your **triage queue for things the platform noticed on its own** — sync, provisioning, metadata, and capacity failures and similar repeating patterns surface here automatically, already grouped, so you work the *problem* rather than chase individual log lines. It pairs with the [audit log](https://dts-identity.com/docs/monitoring/audit-logs.md): the audit log is what you *search* when you have a specific question ("what happened to user X?"); Issues is what you *review* to catch problems that are quietly recurring in the background.

## What becomes an issue

Issues are raised for background operations that fail, typically:

- **Directory synchronization** — a sync run that failed, an upstream user-sync error, a password-sync failure, or a per-user sync problem (such as a duplicate email already in use by another user).
- **Provisioning (SCIM)** — failures creating, updating, deactivating, or deleting users or groups in a connected downstream application, or a connection that could not be reached.
- **SAML metadata** — a metadata URL that could not be refreshed, or a request that could not be parsed.
- **Capacity** — a device enrollment rejected because the device seat limit was reached.

Each issue carries a **severity** of *Warning* or *Error*, and is linked to the related object (application, directory, device, or user) where one applies.

## How grouping works

Repeated occurrences of the same root cause are folded into a single issue. Each issue tracks:

- an **occurrence count** — how many times the underlying event has been seen;
- a **first-seen** and **last-seen** timestamp;
- a histogram of occurrences over time, which you can view over the **last 24 hours, 30 days, or 90 days** — so you can tell a one-off blip from an ongoing problem at the granularity you need.

## Issue states

An issue moves through these states:

| State | Meaning |
|---|---|
| **Unresolved** | Open and needing attention — the default for a newly raised issue. |
| **Resolved** | You have addressed the underlying cause and marked it done. |
| **Regressed** | A previously resolved issue that has started occurring again. |
| **Archived** | Set aside and no longer shown in the active lists. |

The view opens on the **Unresolved** tab and also offers **Resolved**, **Regressed**, and **Archived** tabs, so you can move between open work, things you have handled, problems that came back, and items you have set aside.

[Image: The Issues list with recurring provisioning failures and their occurrence counts]

## Working an issue

1. Open **Audit Trail → Issues**. The **Unresolved** tab lists open issues, most-recently-seen first.
2. Select an issue to open its detail view. You see its type, severity, related object, occurrence count, first/last-seen times, a per-issue histogram, and the full event context.
3. Investigate the root cause using the context — for example, fix the directory credentials or the downstream provisioning mapping.
4. **Change the state** — mark the issue *Resolved* once you have addressed it. If the same problem recurs later, it returns as *Regressed* so it does not get lost. A resolved or archived issue can be **reopened** (set back to *Unresolved*) from its detail page.
5. **Archive** an issue you want out of the active lists but kept on record — use **Actions → Archive** on the detail page. An archived issue is hidden from the Unresolved/Resolved tabs and lives on the **Archived** tab; the issue and its history are retained, and you can still reopen it. Archiving is the reversible "I have triaged this, set it aside" action.
6. **Delete** an issue only when you truly do not need it on record. Deleting removes the issue permanently — unlike archiving, there is no undo and no kept history.

## Related

- [Audit logs](https://dts-identity.com/docs/monitoring/audit-logs.md) — the per-event record that issues are derived from.
- [Plans & add-ons](https://dts-identity.com/docs/getting-started/plans-and-addons.md) — what your tier includes.
