Please take a moment to read the latest blog post written by my manager, Kelley Anders . Her blog explains internal management changes and uncovers division changes that impact the Rational product portfolio. In a nutshell, Rational is now a part of the Cloud division. Learn more below...
-results Optional. The name of the results file. The default result file is the schedule or test name with a time stamp appended.
-overwrite Optional. Determines whether a results file with the same name is overwritten. The default value, true, means that the results file is overwritten.
-exportlog Optional. The complete path to a file in which to store the exported HTTP test log.
-exportstats Optional. The complete path to a directory in which to store exported statistical report data. The statistical report data is stored in comma-separated values (CSV) format, with the file name derived from the report name. "If the -exportstatreportlist option is not specified, the reports specified on the Export Reports page of the Performance Test Report preferences are exported".It says : If the -exportstatreportlist option is not specified, the reports specified on the Export Reports page of the Performance Test Report preferences are exported.
-exportstatreportlist Optional. A comma-separated list of absolute paths to custom report format files (.view files) to use when exporting statistical report data with the -exportstats option. This parameter overrides workspace preferences. For example, -exportstatreportlist c:/customreport.view,c:/customreport2.view.
First step what you need to do is configure the application in the workbench under web application as shown in the picture.
Upon adding connect the mobile client which is running on the device to the workbench. In the below picture, you can see Arm Emulator is connected with the workbench
After connecting you would be able to see www.google.com application listed in your mobile client under Manage web application.
Initiate the recording from the mobile client and have the sample script recorded as shown in the picture.
I have recorded very small and simple script, which launches the google page, enters the user name as IBM and clicks on submit button.
Now we have test script ready which can be executed from the command line.
Firstly you should be keep the Mobile Client on Passive mode and close the running workbench.
Note : If you close the workbench before keeping the mobile client on passive mode, you would be able to set passive mode as the option disabled and grayed out as shown in the below picture.
To run a test from a command line interface, we need navigate to the directory that contains the cmdline.bat file. On Windows operating systems, this directory is typically product installation directory
After navigating the cmdline directory, execute the command in Command prompt
I have added only the mandatory parameters like -workspace,-eclipsehome,-plugins,-suite and also -result from the optional parameters to generate the test log, not included the other parameters. Based on your requirement you can add the parameters you wanted.
Command line execution log will be captured in CommandLineLog.txt, path of this file will be displayed in the command line interface.
Possible error 1:
When you execute the command, test may fail saying “Workspace in use”
1.Make sure you have closed running Workbench before executing the command.
2. Though you have closed the workbench, sometimes you might get this error as process might right in the background, so kill the workbench process from task manager
Possible error 2:
Execution of the test will not start and will not complain any substantial error.
This could be due to the synchronization problem as the emulator sometime takes little longer time to connect to the workbench process and leads to the failure.
The best way to address this issue would be by introducing the delay in the test.
1.Select the Test
2.Right click on the test
4.Introduce the delay as per your environment
Upon completion of all these you should be able to execute the mobile test script from the command line successfully.
During RFT installation, sometimes you might have encountered the below error in IBM Installation Manager.
Error during "post-install configure" phase:
Error executing the C:\ProgramFiles(x86)\IBM\RFT\FunctionalTester\installscripts\installInternal.bat command:status=255.
The below message could be found in the installation manager log file located in C:\ProgramData\IBM\Installation Manager\logs: C:\ProgramData\IBM\RFT\customization XCOPY is not recognized as an internal or external command, operable program or batch file.
cacls is not recognized as an internal or external command, operable program or batch file.
To resolve the post-install configure error during installation you can try following the below procedures one by one:
Launch Installation Manager with Run as Administrator privilege to install RFT:
Click on Start > All Programs > IBM Installation Manger.
Right click on 'IBM Installation Manger' and select 'Run as Administrator' option
Clean un-installation of RFT:
Uninstall Rational Functional Tester from IBM Installation Manager.
Open the Windows Registry editor. Go to HKEY_LOCAL_MACHINE\SOFTWARE\Rational Software\RationalTest and take a backup and delete the entire 8 Folder under it.
Uninstall IBM Installation Manager from control panel.
Remove all files and folders under Rational Functional Tester installation directory, for example: C:\Program Files (x86)\IBM\SDP
Navigate to C:\Users\IBM_ADMIN\AppData\Roaming\IBM and remove Rational Functional Tester folder.
Navigate to C:\ProgramData\IBM\RFT and remove customization and configuration folder.
Restart the computer.
Try to install RFT to a path without spaces and special characters:
During installation, change RFT installation directory to a path without spaces and special characters (For Example. D:\IBM).
Use a new User credentials:
Create a new User with administrator privileges and login to the machine with this new user.
So many factors will affect the automation script play back when you are interacting with application GUI. Once such instance is, where play back fails when RFT tries to perform click action on hyper link control (could be any other control). Though this issue does not consistently happen it might happen intermittently due to application loading speed, RFT play back speed, Application Program Interface (API) used or some other factor as well.
RFT provides API such as sleep(), watiforexistance(),IsVisible().... using which you can handle the play back failure due to above said behavior, however if the issue is not consistent and occurrence of the issue is minimal, introducing sleep() API would consume additional time which could be avoided.
Test Script consist of steps which performs inserting text into number of text fields and then clicking on button, which takes the application to the next page, where RFT needs to perform click action on hyperlink upon loading .
Though the application page has not loaded completely (half page visible to the user), the expected hyperlink control is visible on the screen and RFT tries to perform click action on the control. The moment RFT is about to click, the pages loads completely that makes hyperlink control to move elsewhere and that makes RFT to click on unexpected control which leads to the failure.
Textfiled1.setText(" Name" ) ;
Textfiled2.setText("Address" ) ;
Textfiled3.setText("City" ) ;
Execution of above code snippet might replicate the scenario that we are discussing.
In order to overcome this issue, before performing click action, use getScreenRectangle() API to find the middle point of the control co-ordinate.
By doing this, play back failure could be avoided as RFT relies on co-ordinates of the test object ( not the screen co-ordinates so this script must run in any resolution or screen size )
java.awt.Rectangle r = Hyberlink_AddnewEntry.getScreenRectangle();
java.awt.Point pt = new java.awt.Point(r.x+ r.width/2,r.y+ r.height/2);
During RFT installation, you might encounter the below error
rtivregister.exe - Common Language Runtime Debugging Services
Application has generated an exception that could not be handled.
Process id = XXXX, Thread id = XXXXX
Click OK to terminate the application.
Click Cancel to debug the application.
After clicking OK , RFT installation would be carried out; however you will not be able launch RFT on the machine.
In Windows Registry Editor, under the HKEY_LOCAL_MACHINE\SOFTWARE\Rational Software\Rational Test\8 path, Rational FT Net Assembly Name key would not have created.
To resolve the error during installation you can try following the below procedures
1. Contact IBM Support for obtaining a new rtivregister.exe file. You can raise PMR through SR tool.
2. Install base version of RFT ( RFT 8.6 , 184.108.40.206 or 220.127.116.11. Ignore the error popup message during installation.
3. Take a backup of the rtivregister.exe file present in RFT bin directory located in C:\Program Files (x86)\IBM\SDP\FunctionalTester\bin directory file and replace it with the one obtained from IBM
4. Launch command prompt with 'Run as Administrator' privilege and navigate to the bin directory.
5. Run rtivregister.exe and confirm that Rational FT Net Assembly Name key is getting generated.
For upgrading RFT, you can replace the rtivregister.exe with the original file and perform the upgrade process by ignoring the error.
After upgrade, continue the steps 3 to 5 to complete the installation.
8. As a finishing touch, modify the Start Programs short cut for Library Manager to be able to run as non admin by specifying the target as "C:\Users\perfadmin\IBM\RationalIntegrationTester\eclipse.exe --launcher.ini LibConfig.ini"
If you have configured your license server to deny the anonymous user an Insight license, Insight will not start up correctly and users will not be able to log into Insight. In new Insight installations, this problem may not happen until the trial license expires.
When you have this problem, you will see errors like this one in your WAS SystemOut.log file:
00000060 licensing E CRREL1033E: User 'anonymous' cannot log on to the Rational Insight report server. The user is not authorized to access the report server web application.
User/host not on INCLUDE list for feature.
License path: <port>@<license_server_host_name>:
FLEXnet Licensing error:-39,147
For further information, refer to the FLEXnet Licensing documentation,
available at "www.flexerasoftware.com".
at com.ibm.rcl.ibmratl.FLEXLicenseControl.checkoutName(Unknown Source)
at com.ibm.rcl.ibmratl.FLEXLicense.Out_Name(Unknown Source)
The errors correspond to the time when Rational Insight starts up.
In addition, you will see errors like this in the license server log:
(ibmratl) DENIED: "insight_user" anonymous@<insight_server_host_name>[#u#anonymous#u##s#<insight_server_host_name>#s#] (User/host not on INCLUDE list for feature. (-39,349))
You may also see errors like these:
00000060 licensing E CRREL1005E: The trial period for the report server has expired. Users cannot log on to the report server until this issue is resolved.
00000060 licensing E
Insight requires an anonymous license during startup. To resolve this problem, you need to reconfigure your license server's options file so that the anonymous user can check out an Insight license. The license is checked out and immediately returned, so it will not impact users' ability to connect to Insight. You will need to restart your license server after updating the options file, and the change will take 12 hours to take effect.
The above applies to token licenses. If you are using named-user floating licenses (also known as authorized user licenses) and you follow the above steps, it will work but the license server won't let you list more users in the options file than the number of licenses you have -- so if you have X licenses and include anonymous in the options file, you can only list X-1 users, which does impact your users (unlike what happens with token licenses). This is due to a FlexLM limitation specific to authorized user licenses. With Insight, the anonymous user needs to be able to check out a license in order to start the report server, the anonymous user must be included in the options file. Thus, Sales needs to provide enough licenses to cover all the users purchased plus an additional license for the anonymous user. If you are running into this issue, please reach out to your sales rep.
You might notice during import of test data using RQM Excel import utility that 'keywords' included in the RQM excel import excel sheet overwrites existing keywords stored in a RQM Project Area.This happens due to not meeting the required conditions while importing keywords as part of your test data in the XLS sheet.
Here is a sample to begin with using keyword import available along with the RQM Excel Import Utility Installation
Now lets take a look at what you could perform to over come this:
You can note the following points to avoid overwriting existing keywords during your XLS import :
1.The following parameter from the CFG file is critical while you export the 2nd time from the same sheet (with different Test script and Keywords).
This Worksheet name from the XLS has to be different each time. I started with "keyword1" and went on upto "keyword13" for various iterations to confirm this. I am unsure what is the significance of this Worksheet name causing the dependency. Checking with my developer.
2.Secondly the use of the "manual script" column in the keyword worksheet is critical as well ,and each time you have a new set of the Test script and Keywords, please have new set of Identifiers for the same.
The TS Column cause the creation and linking of the scripts to the respective keywords. I notice that re-use of the TS data , will not create the script and tags the existing scripts (created with previous iteration getting tagged with the new keywords.
3.Use of new worksheet name for "keywords" if re-using the same sheet with new data.
Every iteration of import with new test scripts , you need to have different Test script name, different Keyword Worksheet name and the TS ID's in Keyword Worksheet for next iteration.
You might encounter a situation where RFT is unable to recognize a vb.label control in a Visual Basic (VB) application using descriptive code.
RFT is able to find the control using the recorded script. However, the regular descriptive find() code ( shown below) is unable to recognize a vb.label control.
TestObject findlabel = (findwindow1.find(atDescendant(".class","Label", "Name","Label1" )));
Properties of the vb.label control in object map:
The below shown modified find() code snippet helps to recognize the vb.label control using descriptive code:
public void testMain(Object args)
RootTestObject root = getRootTestObject();
ALL SOURCE CODE AND/OR BINARIES ATTACHED TO THIS DOCUMENT ARE PROVIDED "AS IS," WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT OF PATENTS, COPYRIGHTS OR OTHER PROPRIETARY RIGHTS OF OTHERS. NEITHER THE AUTHOR NOR IBM WARRANT THAT THE FUNCTIONS, ROUTINES, AND DATA CONTAINED IN, OR GENERATED AS A RESULT OF THE DOWNLOAD(S), WILL MEET YOUR REQUIREMENTS OR BE ERROR-FREE. The entire risk related to the quality and performance of the download(s) is with you. In the event that there is any defect, you assume the entire cost of all necessary services, repair or correction. IN NO EVENT WILL THE AUTHOR OR IBM BE LIABLE TO YOU OR TO ANY THIRD PARTY FOR ANY DIRECT OR INDIRECT DAMAGES (INCLUDING, WITHOUT LIMITATION, LOST PROFITS, LOST SAVINGS OR OTHER INCIDENTAL OR CONSEQUENTIAL DAMAGES) ARISING OUT OF THE USE OR INABILITY TO USE THE DOWNLOAD(S), EVEN IF THE AUTHOR OR IBM HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Some jurisdictions do not allow the exclusion of implied warranties or the limitation or exclusion of liability for incidental or consequential damages, so some of the above may not apply to you."
While executing Rational Functional Tester (RFT) scripts, “Java version out of Date” Pop Up could come up. This could halt the the proceedings.
This is a Java security protocol to help user to remind and upgrade to the latest JravaRuntime (JRE). This cannot be disabled or cannot be handles as unwanted windows.
Here's a workaround that might help!
You can run the Java Control Panel as administrator, and then make changes to the Java update settings. Schedule the update
notification to the desired time and avoid untimely notification. Note: Running the Java Control Panel as administrator is necessary even if you are a user with administrative privilege with User Account Control (UAC)..
Follow these steps to run Java Control Panel as administrator.
Click Start. In the Start Search box, type command.
A list of matches will appear. Right-click Command Prompt in the Programs list.
Click Run as administrator
4. To open the Java Control Panel, in the Command Prompt window type: "c:\Program Files (x86)\Java\jre6\bin\javacpl.exe"
5.In the Java Control Panel, click on the Update tab.
6.Check the checked box for “Check for Updates Automatically” and click on Advanced Tab.
7.Change the update notification settings by selecting option to when you want your system to ask for java version out of date popup for example weekly, daily or monthly and at time for the same is also configurable.
Note: If you are not logged in to system as administrator then options to change Java update notifications will be disabled and you will not be able to make changes.
In my experience, there are a couple of ways you can try to resolve this. During playback, you can try following the below procedures in an RFT script. An alternative is to use a test machine to increase heap size, find those
Functional Test Script:
1. Unregister the test objects used in the scripts, once the scope of the test object ends.
This can be achieved by calling unregisterAll() API at the end of each test scripts.
Refer the techhote 1507963 which explains about the usage of unregisterAll() method in RFT
2. Use the cleanup() API in each script to clear the Garbage collector data. This API also can be called at the end of each script.
On a Test Machine:
1. Close RFT and make sure there are no Javaw.exe processes running in the task manager.
2. Go to Start > Run > %temp% and clear all the temp contents.
3. Go to the Workspace directory and take a backup of the .metadata folder (if it exists) and delete it.
4. Increase Spy Heap Memory in registry editor:
Add a DWORD in registry to increase the SpyHeapSize. (Refer technote 1318359)
5. Increase JVM heap size in the eclipse.ini file present in RFT installaton directory located in C:\Program Files (x86)\IBM\SDP\ folder.
(for example increase it to 2044) -Xmx2044m
(Default value would be -Xmx1024m, make sure that your machine has enough memory to increase the JVM heap size)
6. Set a clean initialization of RFT by adding the -clean switch in the shortcut Properties of Eclipse Scripting.
Go to Start > All Programs> IBM Software Delivery Platform > IBM Rational Functional Tester > Right Click on Eclipse Scripting> Properties, append the target with -clean as below:
"C:\Program Files (x86)\IBM\SDP\eclipse.exe" -product com.ibm.rational.rft.product.ide -clean
7. Restart the machine and trigger the playback again
IBM Rational Performance Tester facilitates the script execution to be performed from the user interface section or alternatively execute the scripts via command line editor.
For more information on how do you specify the command line parameters for script execution, you may refer this link.
Sometimes, it could also be a case that you specify each and every command line parameters, but notice a memory leak problem taking place. And, the same when performed from the RPT web UI section, the memory leak problem does not surface.
So the question arises as to why would the memory leak problems only be observed when you execute the scripts via command line editor and NOT when executed from RPT's web UI??
Generally, The RPT "workbench" process (javaw.exe) for a command line playback will use a default maximum heap size unless you specifically specify a size on the command line. So if you are using the -Xmx option in the eclipse.ini to specify a maximum heap size for the GUI-driven playback, then you should specify the same size for the command line playback using the -vmargs option.
For example, if the eclipse.ini specifies -Xmx4096m, then add "-vmargs -Xmx4096m" to the command line as shown below:
Having said all these fundamentals, you might also assume that The 2 eclipse.ini files used by RPT have the same heapsize value set. So command line and RPT workbench should be using the same values.
The "full eclipse" eclipse.ini is used for some settings for the command line playback. However, the maximum heap size (-Xmx) for the command line playback is not picked up from the eclipse.ini.
You can readily demonstrate this. . .....
Modify the -Xmx value in the eclipse.ini so that it is invalid (-Xmx25m for example).
Try to start the RPT workbench; it will fail with the error "Failed to create the Java Virtual Machine".
Now try a command line playback; it will start successfully.
As noted earlier, specify the maximum heap for the command line playback using the -vmargs option on the command line. For example, to specify a 4096m maximum heap size, add "-vmargs -Xmx4096m" to the command line:
In today's world, mobile technologies and applications developments are growing bigger and faster.
It is always a challenge to develop,test and release the mobile app in no time without compromising its quality to meet the market demand.
Having said that, It is always expected that, mobile automation tool support for cross platform app testing, which plays major role in reducing the time consumption of the tester during the test cycle.
IBM Rational Test Workbench supports cross platform in mobile automation testing for iOS and Android. Meaning, you can record ATU (Application Under Test) in Android and play it back on iOS and vice versa.
This significantly reduces so much effort and time for the tester.
In such scenario you might end up facing synchronization and object identification issue.
To overcome or identify the cause, you need to verify the following things
1. AUT should be a hybrid application.
2. AUT should render with same layout and property values in both platforms (iOS and Android)
If your AUT renders with different layout and property values on different platforms, then it won't qualify for the cross platform app testing. This is the reason cross platform testing is not possible on native mobile application
Assume that your AUT is satisfying the above 2 condition and you have recorded test script on Android device.
After recording, you are initiating the playing back on iOS device, it complains about the Synchronization policy as shown in the below picture.
In such scenario you have to set the right synchronization policy . For details more details on Synchronization policy Click here
Set the Synchronization policy to None and initiate the play back
Steps to set the synchronization policy::
1. Open the Test
2. Click on Launch Application
3. Set the Synchronization policy to None
Hopefully you will be able to run your test scripts on iOS and Android.
method as being unable to create the Work Item of type review the case.
The likely cause in this case is that the Work Item type is null which means the process template the your teams using might not contain the Work item "Review" type defined.
You will need to check if you have the work item of Review type defined.
You may click on Manage Project Area > Work Items > Types and Attributes to check if the Work Item Review has been defined.
If the work item “Review type’ is not found , you will need to define the WI Review Type for all affected project areas by following the below steps:
1. Open an RTC client. Connect to the QM server by going to the Team Artifacts tab, then right-click on Repository Connections > New... Jazz Repository Connection, and then fill the requested information (Note: The URL should go as https://<hostname>:<port>/<qm-context>/). Click Ok.
2. Right-click on the newly created connection > Manage Connected Project Areas... and check the project area where the issue occurs.
3. Now the project area is shown. Right-click on it and click on "Open" or just double-click on it. The project area information tab will be opened. Click on the "Process Configuration" tab (shown at the bottom).
4. A tree showing all configuration options will be displayed at the left. Expand the Work Items item and click on Types and Attributes. A panel will appear showing all current work item types defined for this
QM project area.
5. Click on the "Add..." button. A dialog will appear showing 2 fields. Please fill those fields according to the following information:
The performance testing tool that IBM offers is Rational Performance Tester. A new feature in the version 9 that came out March 2016 is the Starter Edition. What's special about the Starter Edition is that you don't need to buy a license. This Starter Edition is the special offer: performance testing without license.
Besides the HTTP protocol, Rational Performance Tester also supports other network protocols. You can do performance testing on network traffic from Citrix, SAP, mainframes (TN3270), for example.
Show me your license
With the starter edition, you can develop your performance tests anytime, anywhere, anybody without buying a license. On top of that fact, much of the documentation about the licensed edition also applies to the Starter Edition when we're talking about the features that are available in both editions.
If we're talking about performance testing, we often mean load tests. However, there are also other kinds of tests in the space of performance testing. One tool is better suitable than another tool for the kind of performance testing that you want to do. Here is an incomplete list of performance testing types.
You do load testing when you want to know what your application does under a specific load. For example, how would your server respond when 200 users send requests to your application at the same time?
If you want to know under what load your application just stops, stress performance testing is the thing for you to do.
Soak or endurance testing
You want to see what happens when your application suffers a certain load for a long time. A great way to see memory leaks.
Suddenly a huge number of users access your application. Then, what happens?
You create tests to see what the influence is of different configurations on your application performance.
From this overview of performance testing types, you can see that you might consider to use the Starter Edition for load test and configuration tests when you want to involve just a few virtual users. If you want to put more load on your application, you can benefit from the cloud capabilities of Rational Performance Tester through the SoftLayer® cloud. The cloud might work for you in terms of timely and easy performance testing.
Starter Edition: try out and find out
The best way to find out whether you can do performance testing with the Starter Edition, is to install the Starter Edition and test. If you are familiar with the installation procedure of IBM products, you might want to skip this chapter. The installation procedure is much the same.
You might encounter an issue where RFT 8.6 fails to recognize the Screenpoints of Html.INPUT.text control launched in Mozilla Firefox web browser
RFT is able to record the actions from the Html.INPUT.text control. Even Test Object Inspector is able to retrieve the properties correctly.
However,during playback, RFT fails to retrieve the screen coordinates of the control and the playback fails with 'No screen point found' error.
Even, Insert Test Object wizard highlights the control in the left hand side top most corner of the page.
We've found that this issue is reproducible only with Mozilla Firefox web browser; but oddly not with Internet Explorer 11. Additionally, we've found this problem to occur while running
RFT 8.6 with Mozilla Firefox 38.
After testing and reproducing this behavior in-house, we've found that to resolve the above issue, you can upgrade RFT to RFT 18.104.22.168 or the higher version.
The recent changes that happened to the server as pointed to us , was that the Database server was accidentally shutdown the day before and post that ,bringing up the Database server and QM application displays this error from RQM Diagnostics.
Note: If the topology is distributed and servers are located in varied timezone than the RQM Web UI user, you should look at the time where the RQM Server and other components are installed to analyse the logs.
IHS logs are huge in size and cannot be opened when IHS is up and running, hence you can copy the logs from IHS install directory and paste it else where to open in a enhanced text editor (Windos notepad has a limitation to open such huge files).
Monitor the server performance and if still continues to display the error, kindly report the error to IBM Middleware Client Support Team with the relevant log entries from each of above log files.
Installing IBM Rational Performance Tester on any Windows flavor is pretty straight forward and it is always suggested that you configure RPT via the IBM Installation Manager. There could be different scenarios under which you in stall the product. We'll talk about a couple of scenarios below:
Installing RPT on a fresh instance of operating system
Upgrading the existing version of RPT
The upgrade part is a simple task as you have to specifically make a pointer to the existing reference of the already installed product files and perform the "Update" operation via IBM Installation Manager.
The fresh installation of the product can be performed in couple of ways.
Install IBM Installation Manager and then configure the RPT application repositories
If the IBM Installation Manager (IM) installer files does not exist then you may manually click on the "launchpad.exe" (RIght click >> Run as Administrator).
Performing point # 2 will first install IM and then account the RPT setup files for installation. There could also be a case where even performing the point # 2 task might result in several errors. Lets discuss a few of them here.
Scenario # 1:
Sometimes a fresh installation of RPT on Windows machine may result in the following error.
The relevant install log may indicate the following symptoms as well.
Unable to install RPT v871 via IBM IM resulting in error..
!ENTRY org.eclipse.osgi 4 0 2016-03-09 05:01:19.752
!MESSAGE Application error
To tackle this error, check if there exists any HP applications (like QTP) installed on this machine?
If yes, then probably the JAVA_TOOL_OPTIONS: -agentlib:jvmhook might be one such probable cause restricting the IM to install the package files.
It could also be a case where despite working on the suggestions mentioned in Scenario # 1, the problem still persists and might also result in additional errors such as:
In such situation, you may either choose one of the below suggestion to get around this problem.
Have you seen the latest video on Rational Performance Tester? The video demonstrates how you can performance test a web application accessed via the Opera Mobile Emulator using IBM Rational Performance Tester. It would also cover the basic configuration required to be set on the Opera Mobile Browser Emulator from IBM Rational Performance Tester perspective. Check it out below!
Three node(triad/redundant) RLKS Licensing server is a fail-over configuration that provide maximum uptime and availability of network licenses. This configuration need three separate machines/nodes located in the same data-center(recommended), running same Operating system(recommended) and same version of License server.
Having above said environment in place, License Administrator need to 'login' in to “Rational License key center” website to generate and download a license file for high availability RLKS environment. Apparently, PRIMARY, SECONDARY and TERTIARY node are decided in the order in which you enroll hostname and host-id data when you generate the license file. Subsequently, System Administrator need to download the license file and deploy(import) the SAME license file on ALL three Licensing server nodes.
Further going ahead, System Administrator has to start ‘License server service’ on ALL three nodes which eventually initiate heartbeat connection between 3-nodes to establish a quorum. what is quorum in Licensing server triad environment?
At least TWO License servers need to establish a connection out of three nodes in order to serve licenses to end-users/clients. Two License servers that form a quorum first, one out of them ACT as a MASTER License server which apparently serves the license keys. In case the MASTER License server goes down, the other two nodes try to establish a quorum between them to serve licenses to end-users/clients.
Now we have a good high level understanding of redundant RLKS License server deployment. Let me give you an overview of “IBM Rational Admin. and Reporting tool” high-level architecture from TRIAD/REDUNDANT RLKS perspective.
In a typical ART deployment scenario, you would have “Rational License key server Administration Agent” component on systems that have License server(RLKS) installed and on another node, an instance of “Rational License key server Administration and Reporting Server” component installed.
When all components of RLKS ART(License servers, ART Admin Agent, ART Admin Server) are installed on respective servers, you would have to login into ART Admin. server web console[https://Hostname:4743/rcladmin/Main.jsp] and add these servers as “redundant” cluster type and pass test connection between Admin and Agent nodes successfully.
Next action is “Start reporting” against each redundant license server node and this step in the background locate Master License server (out of three nodes)and create new log files( lmgrd.log.backup & ibmratl_timestamp.log) only on it.
Let me elaborate this point through an example, lets imagine that PRIMARY has formed a quorum with TERTIARY License server node, because TERTIARY can never be a MASTER License server, PRIMARY License server becomes the MASTER and serves licenses to users/client. When “Start Reporting” feature is started, reporting service create these files(below) ONLY on MASTER License server (here PRIMARY) -
In case of disaster/fail-over (PRIMARY goes down), SECONDARY becomes the MASTER (same set of files created on it). These log files data are crucial to do reporting because ART AGENT reporting service convert these logs data into a RDF data-set that is consumed by ART ADMIN server to get the desired reports.
Here is a unique use case where IBM Rational Performance Tester (RPT) is used to performance test web service calls communicated via MQ (Message Queue) Tunnel connection. In a general scenario, the MQ client runtime makes an HTTP connection to an MQ tunnel servlet at the specified URL (The broker must be configured to access the HTTP tunnel servlet). It can also be a case where the tunneling is established over SSH connection as well.
Overall, it is also recommended that you perform prior checks like:
Can you telnet to the MQ host and listener port?
Check access to Queue Manager
a. Use the MQ or JMS API (make sure you capture and display the linked exception when present in the JMSException.
b. Check that the firewall is letting the MQ protocol through
c. Check that communications are open bi-directional
d. Check doing a ping channel
Let's comeback to this specific use case where the topology for the use case was configured in the below fashion:
- MQ was on a remote server and you are not allowed to connect to it directly
- You have to initiate Putty and load the user credentials and connect through Tunnel
- The authentication (RSA) is done through Putty and then the MQ is configured in such a way that the connection establishes once the authentication is done
- The password is valid for that particular session only based on the key file which was generated.
At the initial stage, you could feel that the only blocker that you see with this kind of setup is using a username/password to connect to the MQ manager (this is the password requested by the MQ explorer ). However, from an RPT perspective this is not implemented at the moment into the product ( No basic authentication available ). So, this leaves you with an option of removing this blocker for testing purpose.
However, a MQ connection authentication is required to connect to the queue manager, and this should be removed at the queue manager side (the RPT product does not support this).
Having said that, if you test a local MQ sample with a Putty connection and RPT configured (Remember the connection is established between RPT machine and MQ) RPT should be able to capture the traffic.
All RAD versions come with an IBM Eclipse SDK (IES) and IBM JDK and you do not need to get or use any open source Eclipse download (which would also require its own JDK). It is not recommended to extend an existing Eclipse. Only problems reported with the IES and IBM JDK in provided by default with a RAD installation, are supported.
As of RAD v9.5, extending an existing Eclipse IDE is no longer supported!