---
title: Monitor appliance health
description: How to monitor the state of an Analytics Platform System appliance by using the Admin Console, or by directly querying the Parallel Data Warehouse dynamic management views. 
author: mzaman1 
ms.prod: sql
ms.technology: data-warehouse
ms.topic: conceptual
ms.date: 04/17/2018
ms.author: murshedz
ms.reviewer: martinle
ms.custom: seo-dt-2019
---

# Monitor Appliance Health State
This article explains how to monitor the state of an Analytics Platform System appliance by using the Admin Console, or by directly querying the Parallel Data Warehouse dynamic management views. 
  
## To Monitor the Appliance State  
A system administrator can use the Admin Console or the SQL Server PDW Dynamic Management Views (DMVs) to retrieve the full hierarchy of nodes, components, and software. The following diagram gives a high level understanding of the components that SQL Server PDW monitors.  
  
![Monitoring overview](./media/monitor-appliance-health-state/SQL_Server_PDW_Monitoring_Overview.png "SQL_Server_PDW_Monitoring_Overview")  
  
### Monitor Component Status By Using the Admin Console  
To retrieve component status by using the Admin Console:  
  
1.  Click on the **Appliance State** tab.  
  
2.  On the Appliance State page, click on a specific node to view the node details.  
  
    ![PDW Admin Console State](./media/monitor-appliance-health-state/SQL_Server_PDW_AdminConsol_State.png "SQL_Server_PDW_AdminConsol_State")  
  
### Monitor Component Status By Using System Views  
To retrieve component status by using system views, use [sys.dm_pdw_component_health_status](../relational-databases/system-dynamic-management-views/sys-dm-pdw-component-health-status-transact-sql.md). For example, the following query retrieves the status for all components.  
  
```sql  
SELECT   
   s.[pdw_node_id],  
   n.[name] as [node_name],  
   n.[address] ,  
   g.[group_id] ,  
   g.[group_name] ,  
   c.[component_id] ,  
   c.[component_name] ,  
   s.[component_instance_id] ,   
   p.[property_name] ,  
   s.[property_value] ,  
   s.[update_time]  
FROM [sys].[dm_pdw_component_health_status] AS s  
JOIN sys.dm_pdw_nodes AS n   
   ON s.[pdw_node_id] = n.[pdw_node_id]  
JOIN [sys].[pdw_health_components] AS c   
   ON s.[component_id] = c.[component_id]  
JOIN [sys].[pdw_health_component_groups] AS g   
   ON c.[group_id] = g.[group_id]  
JOIN [sys].[pdw_health_component_properties] AS p   
   ON s.[property_id] = p.[property_id] AND s.[component_id] = p.[component_id]  
WHERE p.property_name = 'Status'  
ORDER BY  
   s.[pdw_node_id],  
   g.[group_name] ,   
   s.[component_instance_id] ,  
   c.[component_name] ,   
   p.[property_name];  
```  
  
Possible values returned for the Status property are:  
  
-   Ok  
  
-   NonCritical  
  
-   Critical  
  
-   Unknown  
  
-   Unsupported  
  
-   Unreachable  
  
-   Unrecoverable  
  
To see all the properties for all components, remove the `WHERE  p.property_name = 'Status'` clause.  
  
The **[update_time]** column shows the last time the component was polled by the SQL Server PDW health agents.  
  
> [!CAUTION]  
> Be sure to investigate the issue when a component has not been polled for 5 minutes or longer; there could be an alert that indicates an issue with the software heartbeats.  
  
## See Also  
<!-- MISSING LINKS [Common Metadata Query Examples &#40;SQL Server PDW&#41;](../sqlpdw/common-metadata-query-examples-sql-server-pdw.md)  -->  
[Appliance Monitoring &#40;Analytics Platform System&#41;](appliance-monitoring.md)