Dashboards

cancel
Showing results for 
Search instead for 
Did you mean: 

Trying to create an infrastructure disk utilization report for a Hadoop cluster

Explorer

Trying to create an infrastructure disk utilization report for a Hadoop cluster

I have been asked to create a weekly report with hourly observations for disk storage utilization within a Hadoop cluster.  The cluster has 6 data nodes and each node has 8 HDFS disk devices.  Each data node has two additional disk devices for o/s which I want to exclude.

Generating a Secheduled Report doesn't appear to have sufficient flexibility to exclude the o/s disk devices.  I have tried to use the On Demand reporting tool, but I have to laboriously add each disk one-at-a-time for each of the 6 data nodes.

After the first node, adding each subsequent disk takes more than a minute.  At about node three, the reporting tool chokes and dies.

There has to be a better way.  Is the AppD data repository schema externalized so I can write a query using a SQL statement or some other API to retrieve the data I want?

I appreciate the help.

Trying to create an infrastructure disk utilization report for a Hadoop cluster
2 REPLIES
Employee

Re: Trying to create an infrastructure disk utilization report for a Hadoop cluster

Hi,

 

Can you try creating a custom dashboard and configure it with the specific nodes as per your requirement. Once it is created, schedule a report for this dashboard and let us know if that helps.

 

For more information, please folllow 

https://docs.appdynamics.com/display/PRO44/Dashboards+and+Reports

 

Thanks,

Yogesh



Found something helpful? Click the Accept as Solution button to help others find answers faster.
Liked something? Click the Thumbs Up button.
Explorer

Re: Trying to create an infrastructure disk utilization report for a Hadoop cluster

Yogesh,

Thank you for the response.  I have tried to create a custom report.  The problem is that when using the AppD console UI, as I add each additional disk (8 disks per node, 22 data nodes in the largest cluster) the UI slows to the point that it's unusable.  I understand that there are a lot of data points (8 x 22 x 168 hourly observations for a week).  Is there a problem with memory or some other configurable parameter so the I can use the UI sucessfully?

 

I can set up a GoToMeeting session for you so I can demonstrate the problem.

 

Thanks,

Frank