Disk volume Health Rule best practices

[Rugved.Sarode]

Learn best practices for Health Rules on different disk volumes

This article is about creating health rules on volumes and disks, including volume average, individual partitions, and all selected partitions.

In this article...

• Health Rule alerting on average of entire volume
• Health Rule alerting on individual volume partition
• Health Rule alerting on all the partition volumes from the selected server list.
• Edge Case

Health Rule alerting on average of the entire volume

1. Create Health Rule from Alert and Respond. 
2. Affected Entities: 

   Health Rule type: Servers (under APM application)
   Health Rule type: Server Health (under Servers application)
   Affected Entity All Servers of Specific Servers

   
   
3. Critical criteria- Select a metric: Absolution path: Hardware Resources|Volumes|Used (%)
   

   Set the Value > 90

   
   
4. Warning criteria- Select a metric: Absolution path: Hardware Resources|Volumes|Used (%)
   Set the Value  > 75
   
   Sample screenshot of absolute path metric selection

From the metric browser, this monitors the metric:
Application Infrastructure Performance|<TierName>|Individual Nodes|<Node Name>|Hardware Resources|Volumes|Used (%)

Back to Table of Contents

Health Rule alerting on individual volume

This process is the same as in “Health Rule alerting on average of the entire volume” above, but in the Critical/Warning section, use a relative path.

1. Create Health Rule from Alert and Respond. 
2. Affected Entities: 

   Health Rule type: Servers (under APM application)
   Health Rule type: Server Health (under Servers application)
   Affected Entity All Servers of Specific Servers

   
   
3. Critical criteria- Select a metric: Absolution path: Hardware Resources|<PartitionName>|Used (%)
   

   Set the Value  > 90

   
   
4. Warning criteria- Select a metric: Absolution path: Hardware Resources|<PartitionName>|Used (%)
   Set the Value  > 75
   
   Example: Hardware Resources|Volumes|/opt|Used (%)
From the metric browser, this monitors the exact metric:
   Application Infrastructure Performance|<TierName>|Individual Nodes|<NodeName>|Hardware Resources|Volumes|/opt|Used (%)</'/code>
   Metric Selection screenshot

Back to Table of Contents

Health Rule alerting on all the partition volumes from the selected server list.

This process is the same as in “Health Rule alerting on average of the entire volume” above, but in the Critical/Warning section use a relative path with a wildcard.

1. Create Health Rule from Alert and Respond. 
2. Affected Entities: 

   Health Rule type: Servers (under APM application)
   Health Rule type: Server Health (under Servers application)
   Affected Entity All Servers of Specific Servers

   
   
3. Critical criteria- Select a metric: Absolution path:Hardware Resources|Volumes|*|Used (%)
   

   Set the Value  > 90

   
   
4. Warning criteria- Select a metric: Absolution path: Hardware Resources|Volumes|*|Used (%)
   Set the Value  > 75
   
   From the metric browser, this monitors the exact metric:
   Application Infrastructure Performance|<TierName>|Individual Nodes|<NodeName>|Hardware Resources|Volumes|<All volumes>|Used (%)
   Tiers and Nodes within the metric path depends on Affected Entities selection of the Health Rule Screenshot
   

Back to Table of Contents

Edge Case

With the use of a wildcard, if there are 4 disks in a node, A,B,C,D, if metrics for  C disk is not reporting, the entire Affected Entity (Node / Tier / Server) will be unknown/gray color under Health Rule -> Current Evaluation Status tab. But rest assured, the health rule will trigger at the respective configured values for A,B,D, and for  C when data is available and triggering conditions are breached.

Back to Table of Contents

[Balakrishna.Maganti]

Thank you this worked for me

[Claudia.Landivar]

Hey, @Balakrishna.Maganti , so glad this article was useful!
I have a tip that will help other Community members. Kudoing content has the effect of bumping up findability for other Community members.

How do you do it? Click the "thumbs up" button to the far left of the article's Comment button.