Skip to Content

Dear Folks,

Just wanted to share some of the groundwork I did with BusinessObjects monitoring.BusinessObjects monitoring application helps administrators to identify if an application is functioning normally and response times are as expected.It also allows them to

  • Check the performance of each server
  • View the recent failures on the dashboard screen
  • Check system availability and response time
  • Analyze peak load and peak period for the CMS

Frequently used terms in monitoring

Dashboard


The Dashboard page provides a centralized view for the system administrator to monitor the performance of all servers. It provides real-time information on the system KPIs, recent alerts, watches, and corresponding graphs based on the watch state.

Metrics


Metrics are foundation of the monitoring engine. A metric in its simplest definition is something to be measured. There are bunches and bunches of them predefined in the system such as CPU’s, CPU Utilization for the last 15 minutes, Total Memory, Number of Deadlocked Jobs.

Probes


Probes monitor different services and simulate the different functionalities of SAP BusinessObjects Enterprise components. By scheduling probes to run at specified intervals, the system administrator can track the availability and performance of key services provided by SAP BusinessObjects BI platform

Watch


Watches provide real-time status and historical trends of servers and workflows within the SAP BusinessObjects BI platform environment. Users can associate thresholds and alerts to a watch. You can create a watch using data from probes, servers or Derived Metrics.

Alert


An alert is a notification generated by the monitoring application, when a user-defined threshold value set for different metrics applied to a watch is breached. You can choose to receive alerts either through Email or view on the “Dashboard” page.


Below are all some of the probes and metrics that are utilized for monitoring in our environment.

List of probes


Probe Name

Probe Description

Schedule plan

CMS DB Connection

This probe tests the availability of the repository database

Every 30 mins

CMS Ping

This probe sends an empty query to the CMS and check for its response.

Every 30 mins

CMS Logon Logoff

This probe tests the availability of the CMS and the ability of users to log on to the system through client applications.

Every 30 mins

Info View Probe

This probe will log in to the BI Launch Pad using an account and authentication type you specify.

Every 30 mins

Crystal Reports Service

This probe will run a Crystal report using page and Cache Servers or RAS and track how long it takes to refresh it.

Every 30 mins

Interactive Analysis

This probe will run a Web Intelligence report and track how long it takes to refresh it.

Every 30 mins

Start Stop Servers

It will stop and start all of your servers.

Weekly

List of Metrics (Shortlisted based on applicability)

Servers

Metrics

Core

Crystal

WebI

Dashboard

Schedule frequency

Central Management Server

Input File Repository

Output File Repository

Connection Server

Crystal Reports Cache Server

Crystal Reports Processing Server

Adaptive Job Server

Adaptive Processing Server

Web Intelligence Processing Server

Dashboards Cache Server

Dashboards Processing Server

Health State

X

X

X

X

X

X

X

X

X

X

30 mins

Server Running State

X

X

X

X

X

X

X

X

X

X

X

30 mins

Server Enabled State

X

X

X

X

X

X

X

X

X

X

X

30 mins

Busy Server Threads

X

X

X

X

X

X

X

X

X

X

30 mins

Disk Size

X

X

X

X

X

X

X

X

X

X

30 mins

CPU Usage Percentage last 15 minutes

X

30 mins

CPU Usage Percentage last 5 minutes

X

30 mins

Total Disk Space in Root Directory (GB)

X

X

30 mins

Free Disk Space in Root Directory (GB)

X

X

30 mins

CPU Usage (%)

X

30 mins

Virtual memory size (MB)

X

30 mins

                                                                           X – represents the applicability.

     We will see more in detail about scheduling of probes and metrics, creating watches and alerting in my upcoming blogs.

     Looking for your valuable feedback.  Thanks for reading.

SAP BusinessObjects 4.0 Monitoring Blogs

SAP BusinessObjects 4.0 Monitoring Configuration – Part(1)

SAP BusinessObjects 4.0 Monitoring Configuration – Part(2)

SAP BusinessObjects 4.0 Monitoring Configuration – Part(3)

To report this post you need to login first.

18 Comments

You must be Logged on to comment or reply to a post.

  1. Sirisha Mantha

    Hi Mani,

    By  the metric ‘Health State’ , do you mean to say that whether each of the servers are responsive or not? How exactly is it captured?

     

    Can you also please clarify the metrics on “CPU Usage” apply to which of the core components?

    (0) 
    1. Manikandan Elumalai Post author

      Hello Sirisha,

       

      To Answer your question

       

      Health State

      Highlights the current status of the service:

      • “0” – Indicates that the health of the metric is bad. 
      • “1” – Indicates that the health of the metric is deteriorating and needs immediate attention. 
      • “2” – Indicates that the health of the metric is good.

      CPU Usage Percentage last 15 minutes

      Adaptive Processing Server

       

      Web Application Server

      The percentage of total CPU time used by the server during the previous 15 minutes. For example, if a single thread fully utilizes one CPU of a four-CPU system, the utilization is 25%. All processors allocated to the JVM are considered. A value of greater than 70% may indicate a bottleneck.

      CPU Usage Percentage last 5 minutes

      Adaptive Processing Server

       

      Web Application Server

      The percentage of total CPU time used by the server during the previous five minutes. For example, if a single thread fully utilizes one CPU of a four-CPU system, the utilization is 25%. All processors allocated to the JVM are considered. A value of greater than 80% may indicate a CPU bottleneck.

       

      Regards,

      Mani

      (0) 
      1. Carter Russell

        I have values up to 4.7 on a quad core. I would imagine 100% utilization of all four cores would post a value of 4 to Monitoring. How can I have a 4.7 value from CPU USAGE Percentage last 5 minutes?

         

        Where did you find the above screenshots? I couldn’t find this in the Admin document.

         

        Thanks!

        (0) 
  2. karthik rajagopal

    Hi Mani,

     

    We have our audit DB and database DB at the same server and they vary by means of schema name alone. I created a probe for CMS DB connection. Normally we have a down time of whole server(where audit and database DB resides). During that time CMS becomes unreachable and the monitoring service is also down. Hence the Probe is not working as expected. Do you have any work around for this. Thanks!

    (0) 
    1. Abhijit Das

      Hi Bala,

          yes, that’s possible but we implemented through JAVA SDK programming. We wrote an java process which reads the the server status and if the status is not in started mode. it sends email to Admin group. We faced issue with server restart which triggered us to implement the java process for notification

      (0) 
  3. Jennifer Szulinski

    Hi Mani,

     

    We are using BOBJ 4.1 How can I track historical concurrent usage?  Can I do it using the Monitoring module in the CMC.  If so, what is the syntax for the derived metric?

    (0) 
  4. Shinde Amit

    Hi,

     

    I have scheduled the probes but all checks are pending. Could you guide me which type of server will pick this pending recurrences?

    (0) 
      1. Shinde Amit

        Thanks Henry. It worked for me.

         

        Now I am working on configuring email notifications. Added watch and its showing status as “Danger” in CMC. However I am not receiving email notifications.

         

        I have added the correct User under “Notification Setting”. Do I need to add SMTP hostname in any Job Server?

        (0) 
  5. Sachin Joshi

     

    Hi Mani,

     

    With respect to the various metrics available, can you provide me some unit test cases on how to check the various core component servers.

    Also would like to know what possible scenarios would be applicable in such analysis.

    (0) 
    1. Manikandan Elumalai Post author

      I have written this almost 5 years back. Please  refer the section List of metrics based on applicability(Though there is no test cases with captured test results).

      Thanks

      Mani

       

      (0) 
  6. Sachin Joshi

     

    Mani,

    Thanks for your prompt reply.

    Could you please let me know what would be the possible test scenarios. This is a part of upgrade study which we are performing for our client.

    Could you please refer any other blogs as we are already stuck on this activity ?

    Any pointers would definitely go a long way. Appreciate your help.

    (0) 

Leave a Reply