# Read the Connectors page

**Settings > Connectors** lists every connector in your tenant. For each one it
answers two questions: can Mesh reach the source, and is data arriving?

> **note**
This page describes the Connectors screen in the
[Generative UI](../user-guide/generative-ui/index.md) dashboard. In the
classic dashboard, connectors are under **System configurations > Connectors**.

## Where to find it

1. Select **Settings**.
2. Select **Connectors**.

You need the **Admin** role. See [Roles and access](./roles-and-access.md).

![The Connectors page: a table with Connector, Provider, Health, Sync status and Status columns, showing rows that are Unknown and Failed, Unhealthy and Failed, Healthy and Good, and Healthy and Stopped](/img/admin-guide/connectors-page/01-connectors-list.png)

## The table

| Column | What it shows |
| --- | --- |
| **Connector** | The name you gave the connector |
| **Provider** | The system it connects to. **offline** means a connector you feed by uploading files |
| **Health** | Whether Mesh can reach and sign in to the provider |
| **Sync status** | Whether data collection is working, with a time underneath |
| **Status** | Whether the connector is turned on |
| **⋯** | Actions for that connector |

- The page refreshes itself every minute. Select the refresh icon to update it
  now.
- Select **?** next to a column heading to see what its values mean.
- Select **ⓘ** next to a status to see the reason behind it.
- Select a row to open its [Collected data](#collected-data).
- Select **Add connector** to connect a new source.
- Times follow your tenant's
  [timestamp display preference](./audit-logging.md#how-timestamps-display).

## Health

Health is about the **connection**: can Mesh reach the provider and sign in?

| Health | Meaning | What to do |
| --- | --- | --- |
| **Healthy** | The provider connection is available | Nothing |
| **Unhealthy** | The provider rejected a request, or could not be reached | Select **ⓘ** for the reason, then see [Fix a connector](#fix-a-connector) |
| **Unknown** | Mesh cannot confirm the connection yet — common for a new connector, or when the latest sync failed for a reason other than the connection | Check **Sync status** and its reason |

> **note**
If **ⓘ** next to **Unhealthy** reads **Health check unavailable**, Mesh could
not read this connector's status. This is not a problem with your credentials
or the provider. The page tries again every minute. If it persists, or shows on
every row, contact Axiad support.

## Sync status

Sync status is about the **data**: is collection working?

| Sync status | Meaning | What to do |
| --- | --- | --- |
| **Starting** | The connector is running its first sync | Wait. A large source can take hours the first time |
| **Good** | The first sync is complete and the latest sync has no failure | Nothing |
| **Failed** | The latest sync failed | Select **ⓘ** for the reason, if shown, then open **Logs** |
| **Stopped** | The connector is turned off | Turn **Status** on to resume |
| **Unknown** | Mesh cannot tell the sync state, including a first sync that has made no progress for more than 24 hours | Open **Logs**, then contact Axiad support |

The time under the status:

- **Good** and **Failed** — when the page last refreshed its data, not when the
  connector last ran. For that, see **Last run** in
  [Collected data](#collected-data), or **Started** in **Logs**.
- **Failed** during a first sync — how long ago the last attempt was.
- **Starting** — **Since** *time* while the first sync runs normally, or
  **No successful sync — last attempt** *time* if a run has failed or partly
  failed.

## Health and Sync status together

The two are checked separately, so they can disagree. Read them as a pair:

| Health | Sync status | What it means |
| --- | --- | --- |
| Healthy | Good | Everything is working |
| Unknown | Good | Working, but the latest sync was incomplete. Check **Logs** |
| Healthy or Unknown | Starting | The first sync is running — wait |
| Healthy or Unknown | Failed | Collecting data failed for a reason other than the connection, often a system error. Check the reason and **Logs**, and contact Axiad support if it repeats |
| Healthy or Unknown | Unknown | Mesh cannot tell the sync state — often a first sync that stopped making progress. Open **Logs**, then contact Axiad support |
| Unhealthy | Failed | Mesh cannot reach or sign in to the provider. Fix the connection first. If the reason is **Health check unavailable**, see [Health](#health) |
| Any | Stopped | The connector is turned off |

## Status

The **Status** toggle turns a connector on (**Enabled**) or off (**Disabled**).
The change takes effect immediately; there is no confirmation. A disabled
connector collects nothing, and its Sync status shows **Stopped**. The data it
already collected is kept.

When you turn a connector on, Mesh may show a warning for a few seconds. It
means the connector's first sync had stopped making progress: the connector
runs on its normal schedule, but its quick automatic re-runs stay paused until
a run makes progress. Check **Logs** for the cause.

## Actions

Select **⋯** at the end of a row. Which actions you see depends on the
connector.

| Action | What it does |
| --- | --- |
| **View collections** | Opens [Collected data](#collected-data) |
| **Logs** | This connector's recent runs — **Status**, **Sync**, **Started**, **Trigger**, **Items** and **Took**. Expand a row for details. If the connector has a Sync status reason, it is repeated at the top |
| **Edit** | Change the connector's settings, such as credentials or schedule. Some settings, such as an Okta org URL, cannot be changed |
| **Relink** | HR connectors only: reconnect to your HR provider. HR connectors have no **Edit** |
| **Re-sync** | Kubernetes crypto collectors only: clear the last run's status, reconnect and collect everything again now. Use it if the connector looks stuck or its status looks wrong. The row returns to **Starting** and can take a minute to update. The connector must be enabled |
| **Disable** / **Enable** | Same as the **Status** toggle, on connectors that offer it here |

For run history across every connector, see
[Data source syncs](./data-source-syncs/index.md).

## Collected data

Select a row, or **View collections**, to see what that connector has brought
into Mesh.

![The Collected data dialog: Schedule with Last run, Next run and Sync stage Synchronizing; Entities with counts of Users, Credentials, Machine identities and Applications; and the sign-in activity note](/img/admin-guide/connectors-page/02-collected-data.png)

### Schedule

| Field | What it shows |
| --- | --- |
| **Last run** | When the connector last collected. **Never** if it has not run yet |
| **Next run** | When it runs next. **Not scheduled** if no run is planned |
| **Sync stage** | Which stage the connector is in — see below |

| Sync stage | Meaning |
| --- | --- |
| **Initializing** | The first sync is running |
| **Initializing (stalled)** | The first sync stopped making progress, so Mesh paused its quick automatic re-runs. The connector keeps running on its normal schedule and returns to normal once a run makes progress. A stall usually follows failed runs — check **Logs**. If Sync status shows **Unknown**, contact Axiad support |
| **Synchronizing** | The first sync is complete; the connector is collecting changes |
| **Stopped** | The connector is turned off |

Sync stage says *where* the connector is; Sync status on the table says *how it
is going*. For example, a first sync can be at stage **Initializing** while its
Sync status is **Failed**.

### Entities

How many records of each type Mesh holds from this connector — for example
**Users**, **Credentials**, **Machine identities**, **Groups**,
**Applications**, **Machines** or **Software**. Only types with at least one
record are listed.

- These are totals, not the result of one run. For per-run counts, see
  [Data source syncs](./data-source-syncs/index.md).
- **No entity data yet** means nothing has arrived. If the first sync has
  finished, check the connector's Sync status reason.
- A message in place of the counts — such as **Waiting for the first sync to
  complete** or **Failed to fetch count** — means the counts are not available
  right now. This is common just after a connector is created. Try again later,
  and contact Axiad support if it persists.
- Crypto collectors list what their last run found, by category. Hover a
  category for what it includes.

> **note**
This note appears on every connector. Sign-in activity is not included in these
counts. It does not mean something is wrong with this connector.

## Fix a connector

When Health is **Unhealthy**, or Sync status is **Failed** or **Unknown**,
select **ⓘ** next to the status and match what it says:

| ⓘ says | What to do |
| --- | --- |
| **Authentication failed**, **Credentials expired**, or *could not authenticate* | Update the credentials with **Edit** (**Relink** for HR connectors). See [Credential management](./credential-management.md) |
| **Authorization denied**, or *was denied by the provider* | Grant the missing permissions in the source system |
| **Connection or network error**, *could not reach*, or *could not resolve the connector endpoint DNS* | Check the connector URL and that the source is reachable from Mesh |
| *was blocked before the request could be sent* | Check the connector URL. If it is correct, contact Axiad support |
| *found a configuration error* | Review the connector settings with **Edit** |
| *is blocked from making progress* | Check the connector's collection limits and endpoint settings. If you have not changed them, contact Axiad support |
| **Health check unavailable**, or *The sync status is temporarily unavailable* | Nothing on your side — see [Health](#health) |
| *The Azure authentication event stream cursor expired* | Nothing — your credentials are fine and the stream retries on its own. If it is still shown after a day, contact Axiad support |
| *encountered a system error* | Open **Logs**. If it repeats on the next run, contact Axiad support |
| *failed before completion* | Open **Logs** for the details and match them to this table. If nothing there helps, contact Axiad support |
| *has not made progress for over 24 hours* | Open **Logs** and contact Axiad support. Kubernetes crypto collectors can try **Re-sync** first |
| *The first sync has not produced a checkpoint yet* | Wait for the first run. If it is still shown after the next scheduled run, contact Axiad support |
| *The initial sync is still running* | Wait. If Sync status is **Failed**, open **Logs** and check the partial runs |
| *The connector is disabled* | Turn **Status** on |

After a fix, the status updates on the connector's next run.

If you contact Axiad support, include the details in
[What to include in a support request](./troubleshooting-and-support.md#what-to-include-in-a-support-request),
plus the connector name, its Health and Sync status, both **ⓘ** texts exactly
as shown, the time under Sync status, and **Last run** from Collected data.