Wednesday, 15 June 2016

Validating the health of your EPM install



Well, I have spent the last couple of days exploring EPM by breaking it into bits and pieces… And by breaking into bits and pieces means, I have crashed my environment and finally my system could not take any more abuse and I had to delete the whole installation and start over… But, learned some really cool things… 

So, first off, let us start with how to validate the health of the EPM installation that you have. Now, it is very common that you may want to check the health of your EPM infrastructure. EPM is now a complex web of components, held together by baling wire and it is quite possible that you may have some components that fail… (In terms of probability, if you have n components, the probability that at least one component may fail is equal to 1 – [probability that none of the components fail] ….so call me superstitious, but it pays to know which utility to run in case I need to have a quick summary of what is messing up my environment)

Now, the utility that is used to validate my EPM installation is called, drumrolls please, validate.bat (no surprises that it is called validate…It would have been a different story if it had a cool name though, like Xerxes or Aggamennon)

The path of this file is basically <EPM SYSTEM Home>\user_projects\epmsystem?\bin\ as shown in the below snapshot. This is the same path where the stop and start batches used to startup and shutdown the EPM environment is present.






On running this utility, it will check all the components that are installed on the particular machine… So it will check the database connectivity, Oracle HTTP server and so on… 

Snapshots of running the utility is as shown in the next couple of snapshots.


In the below snapshot, the utility is checking the connectivity to the RDBMS schemas
 


Checking connection to 19000 port fails with a connect error since I have messed up the Weblogic configuration.
 


Once the utility finishes running, it will generate the status report as an HTML file and open it in the browser for you. 

The summary report is as shown in the below snapshot.
By default this report is stored at <EPM Middleware home>\user_projects\epmsystem?\diagnostics\reports\ folder and will have the timestamp information in the file name.

Information includes the timestamp when the diagnostics were run, version of EPM, server name where validation was run and other information.


Now, in the above snapshot, port 19000 is closed, so all the weblogic connect statements are failing…

Below snapshot has some components that are validating successfully, like my Essbase environment. So now, if I was in firefighting mode in this server, I would know that something is causing my port number 19000 to not be open and all my issues (for want of a better word) can be tracked back to this port… Valuable information (actionable insights is how I would like to describe the reports) since it tells us exactly which component is failing.
 

No comments:

Post a Comment