EPMA - Clustering - Load balance

How can we do EPMA Clustering / Load balance??

How to load balance EPMA server? Is it possible?

No, We can't load balance EPMA Server, you can configure EPMA dimension server and EPMA (other service name) in second node. 

EPMA Server must be in first server along with EPMA Data sync and EPMA Web Tier, in next server do not install EPMA Server, only select EPMA dimension server and EPMA Web Tier.

If we install itself, by default with configuration EPMA all 3 services will get created. So, needs to uncheck this while doing installation itself.

In case, if we install and configured all 3 services in both servers. Then stop or disable EPMA Server in second server.

Thanks,

Mady

How to change Essbase Arbor Path

 If you want to edit Arbor path in Configuration console, in case of wrongly updated or moving new drive, need to follow below steps to edit Essbase Arbor Path in Configuration console.

First execute below command which will enable config task for configuration again.

Go to this path using cmd Oracle/Middleware/user_projects/epmsystem1/bin

resetConfigTask.bat -product EssbaseServer -task RegisterEssbaseAsMultipleAgent

Then Launch ConfigTool again and check Essbase Server for configuration. Essbase Arbor path will be in edit mode, where you can change or update to new path. Execute configuration and check out Essbase path.

Copy applications from old path to New Arbor Path. Try to access those applications using EAS Console.

 Wrong ARBORPATH on Essbase Instance (Doc ID 1910743.1)

Thanks,

Mady

Hyperion SSL Implementation

SSL is important for security and sometimes needs to be implemented in Hyperion environment depending on an organization’s security requirements.  This can be expensive and time-consuming or simple depending on the implementation.

Hyperion has three ways of implementing SSL which are:

Full Implementation – implementing SSL in WebLogic, Oracle HTTP server, and IIS (if used).

OHS Implementation – implementing SSL in Oracle HTTP Server only.

Network Implementation – implementing SSL at the network level on a device such as a load balancer.

Full SSL implementation is the most complex setup and costlier since multiple certificates are needed.  Keystores need to be generated and managed, the OHS wallet needs to be managed, and a lot more configuration, testing, and troubleshooting are needed.  The benefit of a full implementation is that this is the most secure method.

OHS SSL implementation is simpler and requires minimal configuration.  The SSL certificate is managed in the OHS wallet and EPM configuration is relatively simple.  This implementation is less secure than a full implementation, but can still be secure if the Hyperion servers are isolated on their own subnet and access is only given to the OHS server.  Only one SSL certificate is required per OHS server so cost is minimal.

Network SSL implementation is the simplest method and good for security if the network is setup correctly.  Two SSL certificates are added to an external device such as an SSL Off loader or load balancer.  One certificate is needed for external communication between users and the device, and a certificate is needed for internal communication among Hyperion applications.  Hyperion is configured normally, and the load balancer forwards all requests from OHS.

Performance can and should be a factor when considering SSL implementations.  A full SSL implementation can cause Hyperion to run slower.  For example, a client wanted a full SSL implementation in their dev environment.  SSL certificates were obtained for every server and software piece that required them.  After configuration was done and tested, the environment was released to users who reported that Hyperion was running slower than normal.  It would take a minute or two just to get Workspace to load up. After this report, SSL was only implemented in OHS in another environment, and the performance was a lot better.  After seeing this difference, SSL was redone in dev for OHS only, and performance increased.

Reference: https://www.excelglobalpartners.com/considerations-for-ssl-implementation-in-hyperion-environment/

Thanks

Mady

Planning - Essbase -Location Alias Issue

 Hi Everyone, Today we found one issue with Hyperion Planning application, mainly in business rules.

While Launching/Executing Business Rules from Planning, We found error below.

Initially, We tried to check error in below ways:

1. We tried refreshing database from Planning to Essbase to check connect.
2. Even We tried checking Data Source of Planning to check Essbase Connection before first step.
3. Essbase side - Application is fine, no issues there.

We tried to check Location Alias for Cube, there we got same error. As we found one article from Oracle Support which is pointing same.

So, We verified Location Alias, it is giving error. We have updated Cluster name instead of Server name in Essbase Node section, It is verified successfully.

Now, Business Rules from Planning and executed successfully.

Oracle support: Doc ID: 1922470.1

Thanks,

Mady

In weblogic production mode if Admin Server abruptly shuts down do other managed servers continue to work???

 In WebLogic production mode if Admin Server abruptly shuts down do other managed servers continue to work??

Yes, Other Managed servers will not dependent on Admin server. Admin server is used for connecting Weblogic URL, where we can check all managed servers status and we can set all settings for all managed servers.

We can check below in Admin Server:

1. All Managed servers deployed in to that domain.
2. All Settings related to Managed server in that domain.
3. Settings, like Stuck Thread values, Time out settings, port number, Host names etc.

Thanks,

Madhusudhan

How to trace Essbase errors and check logs - Part2

Essbase logs, we need to check for any issues.

ESSBASE.LOG                                        —    Essbase Server activities and errors

ESSBASE_ODL.log                                 —    Essbase Server activities and errors

dataload_ODL.err                                     —    Data load and dimension build errors

log0000x.xcp                                            —    Errors that result when Essbase Server stops abnormally

leasemanager_server_HOSTNAME.log  —    Essbase Server Lease Manager information

leasemanager_essbase_HOSTNAME.log—    Essbase Agent Lease Manager information

log00001.xcp                                            —    Errors that result when the agent stops unexpectedly

easserver.log                                             —    Administration Services Server activity

EssbaseAdminServices0.log                    —    Administration Services Java web application activity

opmn.log                                                  —    Information about when Essbase starts, stops, and how                                                                                 many stop and start retry attempts are made

console~ESSBASE_CLUSTER_NAME~ESSBASE_PROCESS_TYPE~AGENT~1.LOG — All console messages are directed to a file that is called the "console" output file for a managed process, in this case, Essbase.

Note: ESSBASE.LOG and ESSBASE_ODL.log contain the same information in different formats.

Thanks,

Mady

How to trace Essbase errors and check logs - Part1

     While working with Essbase, we may face lot of issues in our day to day activities. Here, I am going to update how to check errors and validate connections of Essbase. Why because, Essbase runs with OPMN in most of the deployments.

Frist of all, Essbase is an multi-dimensional database, so we will have all data saved into the Essbase Arbor path only. Most of the issues, will be with data only. Example as blow:

1. Error while loading data

2. Error while Retrieving data

3. Error while updating or loading Metadata.

4. smartview data validations etc.

Along with those, lot of admin issues will be happening in our daily activities. Like, connectivity, user access, space getting full, compression not happening, cache setting errors, Restructuring errors etc.

So, whatever issue if our work will get impact, then only It is issue. 😄

You will find errors in below logs:

Fist of all, Essbase log. Navigate to the Essbase.log Path, to find the Essbase.log.

Along with Essbase.log, in the same folder, we can find Essbase_ODL.log files, which log will be rotating and will be created as multiple files.

You can search with error, login name, process key words.

If you find xcp file in application folder, then application may corrupted/recovered from corruption.

If you find xcp file in Database folder, Then Database may corrupted either data or entire database.

(you can try renaming .apb as .app for applications or .dbb as .db for database as initial step. If itwon't work, try to contact with Infra team to restore database/application. You can contact Oracle also to find RootCause for generaing xcp files.)

Will continue in next part about analysis and tracing logs...!! 👍

How to trace Essbase errors and check logs - Part2

Thanks,

Mady