Webi XI 3.1 SP2 LOAD TESTING ISSUE WITH LOAD RUNNER

Hello,
We are launching a test campaign in order to test the Webi XI 3.1 SP2 (FP2.2) platform to validate it is capable to host 150 concurents users.
The test is pretty simple:
Logon to Infoview
Navigate to the Document list page
Make a search to find the adequate report
Open this document
Fill the prompts (doc is auto refresh)
Wait for the results
Logoff InfoView
The problem is related to the tool with which we are scripting that is not able to run this script (Tool is Load Runner (ex Mercury)). In fact, we are able to capture the script correctly but when it is replayed, it crashes when displaying the document list page as it can't find the frame with the list of documents and the tree view...
Do you know whether there are some specific settings to make it working correctely?
For information, the Protocol used for our testing are:
Web Click-&-Script
HTTP-HTML
AJAX

I've used LoadRunner for this purpose before. There is a tutorial that comes with the product. My suggestion is to go through that and from there you should be able to get it working with enterprise. I feel that's the best way to get up and running quickly.

Similar Messages

Ajax Testing issue with e-Tester

I'm new to using Empirix and we are using it to work with a new site that uses Ajax. I have recorded the script fine but there is an issue with the Link that needs to be clicked on that it will either always change or the index will change. I've tried to use a wild card after the number in the link that always stays the same but when I run the tests it always shows "Used secondary attribute 'index' ". Is there a way to get this to stop showing every time or get it to always use this attribute?
Thanks,

Hi,
<br>There are two different issues here:
<ol style="margin-top:0px;">
<li> The link that will be clicked will always change or the index will change.
<li> e-Tester is always using the secondary attribute "index"
</ol>
Let me issue #2 first since its easier to explain.
<h2 style="color:navy"> e-Tester is always using the secondary attribute "index"</h2>
<h2 style="color:navy">* For scripts that are not created yet:</h2>
In order to change the order that e-Tester records the use of attributes you can go to:
<br> <font style='color:navy'>e-Tester » Options » Current Script</font>
<br> Then on the left side of the Current Script Options window go to:
<br> <font style='color:navy'>Recording » Object Identification</font>
<br>The following image presents how the screen looks like:
<br><img src="http://qazone.empirix.com/servlet/JiveServlet/download/25-518-1621-527/upanddown.jpg">
<br>
Notice the bottons with arrows that are on the right side of the window. Use those to determine which attribute should be used first when recording the script.
<h2 style="color:navy">* For existing scripts:</h2>
Go to the page where resides the action of clicking the script, expand the page, <font style='color:navy'>right click on the Address Node</font> and select properties(a new window will appear), double click on the action that is making the click to the link (a new window called Event Properties will appear). In this new window click on the Attributes tab. The following image presents how the Event Properties window looks like:
<br><img src="http://qazone.empirix.com/servlet/JiveServlet/download/25-518-1621-528/upanddownattributestab.jpg">
<br>Notice the bottons with arrows that are on the right side of the window. Use those to determine which attribute should be used first when playing back the script.
<br><br>I will create another post in order to answer the other issue. I need to organize my ideas.
<br><br>Regards,
<br>Zuriel

Wls 7.0 sp2 proxy plugin issue with iplanet6.x and wls 6.1 sp3 app server

Hi,
I've a setup with iplanet 6.x webserver using the wls 7.0 sp2 proxy plugin to
route requests to the clustered app server instances (2 of them) that are in wls
6.1 sp3.
The issue is, I'm noticing that the session is not sticky and is getting routed
onto the other app server instance. The error that I see in the wlproxy.log is
as follows:
*******Exception type [PROTOCOL_ERROR] raised
at line 654 of URL.cpp
Thu Nov 13 11:30:08 2003 failure on sendRequest() w/ recycled connection
to Instance1:7001, numfailures=1
Thu Nov 13 11:30:08 2003 Marking Instance1:7001 as bad
Thu Nov 13 11:30:08 2003 got exception in sendRequest phase:
PROTOCOL_ERROR [line 654 of URL.cpp]: unexpected EOF
reading HTTP status at line 1010
Thu Nov 13 11:30:08 2003 Failing over after sendRequest exception
Thu Nov 13 11:30:08 2003 attempt #1 out of a max of 5
Has anyone seen this issue? Can anyone explain why this issue is occuring.?
Thanks.

Did you find a solution for this problem?
I have somthing very similar going on where I am currently working.
Thanks!
Andy.
"Simple Guy" <[email protected]> wrote:
>
Hi,
I've a setup with iplanet 6.x webserver using the wls 7.0 sp2 proxy plugin
to
route requests to the clustered app server instances (2 of them) that
are in wls
6.1 sp3.
The issue is, I'm noticing that the session is not sticky and is getting
routed
onto the other app server instance. The error that I see in the wlproxy.log
is
as follows:
*******Exception type [PROTOCOL_ERROR] raised
at line 654 of URL.cpp
Thu Nov 13 11:30:08 2003 failure on sendRequest() w/ recycled connection
to Instance1:7001, numfailures=1
Thu Nov 13 11:30:08 2003 Marking Instance1:7001 as bad
Thu Nov 13 11:30:08 2003 got exception in sendRequest phase:
PROTOCOL_ERROR [line 654 of URL.cpp]: unexpected EOF
reading HTTP status at line 1010
Thu Nov 13 11:30:08 2003 Failing over after sendRequest exception
Thu Nov 13 11:30:08 2003 attempt #1 out of a max of 5
Has anyone seen this issue? Can anyone explain why this issue is occuring.?
Thanks.

Connection Test issue with newly created INTEGRATION_DIRECTORY_HMI in PI7.0

Hi Pals,
Due to change in password for user "PIISUSER" in our system, i changed the login credentials in sm59 -> INTEGRATION_DIRECTORY_HMI.
But the test connection failed, popping up the username/password login dialog again and again.
So i deleted the RFC Connection "INTEGRATION_DIRECTORY_HMI" and recreated it with all the care mentioned in the XI installation guide, manually, step by step.
But again the test connection failed, popping up the username/password login dialog again and again.
I tried this exercise for almost 15 times but without success.
NOTE: I am able to login with "PIISUSER" from SAP system Login successfully.
Looking forward for your kind suggestions.
Thank you!
Saravanan

hi
could be the problem with password..we also faced the same problem while using an alphanumeric one.
after changing to a completely numeric one it started working fine
rgds
Arun

How to test issue with accessing tables over a DB link?

Hey all,
Using 3.1.2 on XE, I have a little app. The database schema for this app only contains views to the actual tables, which happen to reside over a database link in a 10.1.0.5.0 DB.
I ran across an issue where a filter I made on a form refused to work (see [this ApEx thread| http://forums.oracle.com/forums/message.jspa?messageID=3178959] ). I verified that the issue only happens when the view points to a table across a DB link by recreating the table in the local DB and pointing the view to it. When I do this, the filter works fine. When I change the view back to use the remote table, it fails. And it only fails in the filter -- every other report and every other tool accessing the remote table via the view works fine.
Anyone know how I can troubleshoot this? For kicks, I also tried using a 10.2.0.3.0 DB for the remote link, but with the same results.
TIA,
Rich
Edited by: socpres on Mar 2, 2009 3:44 PM
Accidental save...

ittichai wrote:
Rich,
I searched metalink for your issue. This may be a bug in 3.1 which will be fixed in 4.0. Please see Doc ID 740581.1 Database Link Tables Do NoT Show Up In Table Drop Down List In Apex. There is a workaround mentioned in the document.
I'm not sure why I never thought of searching MetaLink, but thanks for the pointer! It doesn't match my circumstances, however. The Bug smells like a view not being queried in the APEX development tool itself -- i.e. the IDE's coding needs changing, not necessarily those apps created with the IDE.
I'm working on getting you access to my hosted app...
Thanks,
Rich

Web ADI on 11.5.10.2 Issue with migrating templates desktop ADI to WebADI

Version 11.5.10.2
Applied patches:
7375915
7350414
3036401
5853005
11793768
12700905
Migrating FSG templates from desktop ADI to WebADI. When we open an existing template, we put the template name in the file and then upload we have the following issues:
- The template does not have headers.
- If we change the template at all, the columns get left justified instead of right justified.
- numeric value is showing for date and time instead of standard format for date and time.
- Column widths are not holding.
- Cannot get column headings to show up.

Have you applied all pre-req. patches?
Please see the following docs and make sure you have all the patches applied.
How To Determine Which FSG Patches Have Been Applied In Release 11i, 12.0 and 12.1 [ID 549289.1
Unable To Publish Report In Excel After Applying Patches 7375915/7350414 [ID 789725.1]
UNABLE TO VIEW OUTPUT WHEN PUBLISHING REPORT SET WITH CONTENT SET [ID 827461.1]
Unable To Publish FSG Reports When Content Sets Are Used [ID 787289.1]
How To Use Content Sets (Parallel Types) in Report Manager to Display FSGs in Multiple Workbooks [ID 881532.1]
FSG: Error : Content Set ID Is Not Empty And It Is An FSG Submission When Using Content Set [ID 1294149.1]
What Are the Patch Numbers for Prerequisites for Patch 5377946? [ID 945032.1]
Fsg Column Headers Are Offset, Date Format is incorrect after applying FRM patches [ID 834860.1]
Thanks,
Hussein

DHCP Renew Issue with Road Runner

I have had Road Runner for 6 weeks, the first four weeks everything was running smoothly. However, for the last two weeks the connection has been sporadic and I am constantly losing my connection.
My computer is connected directly to the modem with an ethernet cable; there is no router, there are no other computers. I'm simply trying to stay online with my little computer.
My network settings are for built-in ethernet, using DHCP, with the addresses provided by my ISP (Road Runner) and I've added two DNS server addresses for Road Runner (because several other discussions, posts, etc. recommended using those to help stay connected). I've tried turning my IPv6 off, I've tried manually configuring my DHCP connection, I've scoured the internet for help. All to no avail. Most topics deal with networking with PC's or airports and I have neither attached to this computer. I've created new locations to try and "erase" any default settings and I've kicked the modem a few times (accidentally I assure you ).
From what I can tell, (keep in mind I sell trees and know nothing about this) the DHCP lease does not renew when it's supposed to. I am disconnected every half hour or so, or I have to press "renew" to be able to stay online (highly annoying).
When I first subscribed to Road Runner the connection was fine and I was able to stay online indefinitely. Two weeks ago that changed, even though I did nothing (intentionally) to change network settings. This means I downloaded nothing that would affect my network configuration, I didn't poke around in network configurations, I did not change any settings. I've spoken with their online "tech" support four times. I've spoken to three people on the phone and I've had a tech come out to replace my modem and ethernet cable. When the problem persisted I contacted them again and was told the problem was with Apple <gasp>.
I believe I need help in finding a way to have my mac recognize changes in DHCP and to renew the lease automatically, for all I know it could be that the moon isn't aligned with Saturn <shrug>...preferably in vernacular I can understand. Thank you in advance for any help!
Lori

Hi Lori,
You're not really in the right forum for your problem (this area is for Apple-server) but I'll give it a go...
First, short explanation on DHCP. This process is pretty straightforward for the client's function so I'm suspicious of the "blame the mac" conclusion from Road Runner (although does not exclude possibility). When the mac starts up, it issues a 'Discovery' broadcast, looking for a DHCP server. The server responds with an 'Offer' of an IP address. The client then responds with a 'Request' to the server for the same IP address and the server then sends an 'Acknowledgement' that it may use this address, along with other information including the 'Lease Time' - how long the client may continue to use that info.
The lease time is important. After 50% of this time has gone, the client will attempt to renew it from the same server by issuing another Request and awaiting another Acknowledgement. Should there be no Acknowledgement, the client will continue to issue Requests up until the expiry of the original lease time at which point it will drop all the info which it received from the server and then issue another Offer broadcast, looking for a new server (which may or may not be available).
So, first thing, lets see what lease time you are getting from Road Runner and see if this corresponds with the 30 minutes you seem to get from a session...
Ensure you have a cable connection to your router, restart computer. In Applications-> Utilities, start up Terminal. Enter the following line, ending with normal 'return' key (new line)...
ipconfig getpacket en0
That's a zero at the end.
Look for the line similar to this, "lease_time (uint32): 0x1bd8", and post back the string of chars you get at the end of your line (corresponding to the "0x1bd8" part). Quit the Terminal utility.
Can you also confirm what computer model you use, and the system version?
-david
Server 10.4.8

LCM issues with Workflow, Task Lists, etc.

I have found several issues with migrating certain artifacts using LCM v11.1.1.3. Note that I have admin rights in all systems, and that there are no special characters involved. So...
1) Workflow: when migrating Planning Units all of them migrated, including their statuses. But not all PU comments (<annotation>) and logs (<log>) nodes migrated. It happened to almost all users - some who had 1-2 comments in the source system haven't got any in the target one, some got migrated only few (e.g. out of 5 in source, they got 3 in target). Since it happened to all users, it doesn't seem to be security. I have also checked (for those who partially failed) some other nodes, sequence, action, process state - they all seem to be OK. One curiosity I've spotted in few cases, is that in source LCM file (.xml) "priorOwner" node has the name of the user I am investigating, while in the target file (taken after the migration), for that same PU, that node is empty, actually says "null".
Any ideas why this might be happening?
2) Task Lists: when exporting TLs (in LCM) on occassion <dataForm> node in XML file would be missing for the task that actually has it attached. Again, no special characters are used, and I am admin. Then I tried to use TaskListDefUtil, and even if that list has required node, when I try to import, the task is still not attached to the form. I had to delete all TLs, then use Util files to reimport.
Anyone with similar experiences?
smilo

I had issues with tasks lists using LCM in instances where parent task name was same as child task or a task has same name planning web forms. You might also find issues with exporting webforms with business rules attached to forms where form is on plan type A and business rule is on plan type B. In this case, after LCM import, plan B business rule will lose location reference to correct plan type and get associated with web form plan type. In this case you will need to attach business rules again.

I have a 2012 Macbook Pro and the GPU is causing issues with Yosemite. How can I resolve this issue?

I have a 2012 Macbook Pro and the GPU is causing issues with Yosemite. There are a number of issues I have run across:
If I have four tabs open on Safari then I start to get a flickering on the screen from the GPU refreshing
When I leave the computer alone for more than 20 mins the screen saver should start -- what happens is that the screen goes to black. When I move the cursor I do not get the login screen. I have to reboot
FireFox Web browser will not run
I get issues with the computer running very slowly
I have already turned off the transparency option in settings. Any ideas??
Cheers
Matt
MacBook Pro, OS X Yosemite (10.10), GPU / Video Card with Yosemite

Do you have more than one OS on the Mac?
You may want to look at something like http://bjango.com/mac/istatmenus/. This has a Free Trial window, but it is a commercial application. If there are any heat/sensor issues, this may help.
A SMC Reset or PRAM/NVRAM reset may also help.
Intel-based Macs: Resetting the System Management Controller (SMC) - Apple Support
Startup key combinations for Intel-based Macs - Apple Support

I'm using Firefox 3.6.23 on my windows 2000 and lately i've been having issues with web pages loading and need help figuring out why.

Some of the issues : '''1'''. Page loads blank showing web address in tab. Try reloading usually with no results.(problem have the most!) '''2'''. Message appears "problem loading page, server not found" (diff times & diff websites). '''3.''' Pg with msg "oops link broke, DNS error, server not found" then lists suggested links to sites for fix though all links connect to site IGEARED.COM. (new issue) I don't know what to do please help me identify and fix my issue(s) in english, kind of a newbie : ). thx

Try this.
Type in the address bar about:config.
Accept the warning.
In the page that appears, in the Filter box, type network.http.max-connections. Change the value to 32 (which is probably set to 256 in your case).
Close that page. Restart the browser.

Having issues with loading pages with large web articles..

I'm having issues with pages loading all the way when I read articles off of wikipedia. Mainly from pages that have tons and tons of text, but it doesn't make sense. The browser indicates that the page is loaded all the way, but when I make it half way down the page, everything becomes completely white. Like LITERALLY everything becomes white. This prevents me from reading any further down the page, even though I would force the page to scroll down all the way, that doesn't help one bit.
Does anyone have a solution or a fix to this? I would like to continue reading my web articles, without being interrupted with these white pages/bugs.
Post relates to: HP TouchPad (WiFi)

Try clearing the cache for the web browser. While the browser is open, go up to Web/Preferences and choose "Clear cache".
Also try rebooting the TouchPad via Device Info/Reset Options/Restart.
WyreNut
I am a Volunteer here, not employed by HP.
You too can become an HP Expert! Details HERE!
If my post has helped you, click the Kudos Thumbs up!
If it solved your issue, Click the "Accept as Solution" button so others can benefit from the question you asked!

Issue with testing web service via SICF

Hi, I am trying to test service via SICF. This service is for a FM based Web Service. When I test the service from SICF it passes the authorization checks but then error page, "Page not Found" is shown with error "HTTP 500 - Internal server error Internet Explorer".
The trace file shows the errors:
Error in module XMLConverterReadTag:3059
can not read XML data stream(2097184)
Does anyone know what is causing this issue. The FM which the web service is based on is a test FM with one line of code and returns the character 'S'.
Any idea or suggestion will be great.

Dear Shay,
Thanks for your prompt response.
You are right. JDeveloper 11g uses JDK 6 style annotations for the clients it creates. But you can change the JRE Version used at compile time by following these steps:
1. In the Applications Navigator, right-click the Project Nanem node and select Project Properties... from the context menu.
2. Select the Compiler node and check the Source Files and Generated Class Files dropdown lists. You may change these versions depending on the version of the JRE you are using with Forms to ensure that the compiled
classes from JDeveloper can be read by the JRE used by Form.
So I selected JDK version 1.4 there.
Sorry that I did not mention that we are using Oracle Forms 10g. That is the reason I selected JDK 1.4
Thank you.
Scott.

HANA XS - How to fix Inactive Testing issue of items not loading (when using the cookie)

Is anyone else having issues with the inactive testing using the cookie (sapXsDevWorkspace)? My team has abandoned using it because it is causing us a lot of trouble. It initially worked well for a few hours, but eventually it stopped loading resources. We got errors loading the SAP js libs or even the newly commited ( but not activated) code changes.
I have seen the feedback on SCN post http://scn.sap.com/thread/3517042 but we cannot seem to fix the issue of the system not loading things. I am on version SPS7 (1.00.70.00.386119) on AWS and I have seen it work for a short period then it seems to get flaky and stop working. I don't understand what is causing the error to occur as I cannot make it do the same thing twice.
Here are some screenshots:
Cookie defined in Chrome
Repository Name with my user:
This is the result:
Sometimes it loads more of the files and fails later:
I verified that I logged in under the same user and the cookie name. The system has developer_mode set to true. Is there something we're missing?
Jim Giffin.

Hi Jim, Hi Matthias
we are facing the exact same problem with our project.
Jim, we had the same phenomenon that it worked for 1-2 reloads and then stopped at all.
After looking into it, I figured out, that the browser was using cached files from all resources except from the "inactive" new files, which we could access with the cookie. This made it work, however, when we refreshed again, the browser eventually removed its cache and tried to load inactive versions from files that didn't had any.
We ended up in empty responses (no content, no headers)
We are really looking forward to find some answers here, as it constantly causes trouble to work with multiple developers on the same activated code base.
Any input or suggestions are appreciated.
Best Regards,
Axel

Web sites will not fully load. MSNBC, Fox and CNN load a partial page or load a text only version. this issue just started about a week ago.

Some news web sites will not fully load. MSNBC will only show the head line and advertisers. CNN and Fox show a text only version with very limited pictures and the same for . Durdge report loads fully as will the Chicago Trib. seems like only certain news sites are subject to this partial display. have not seen this on commercial sites.

Try deleting cookies and cache:
1. Tools| Clear recent history
2. Time range to clear: Everything
3. If it isn't already selected, select '''Cookies''' and '''Cache'''
4. '''Clear now'''
Test for the problem
Also see: [[Websites look wrong]] and [[Troubleshooting Extensions and themes]]

Load Testing Discoverer Plus 11g Web Using HP Load Runner 11

Hi,
I am doing load test for "Oracle Business Intelligence Discoverer Plus 10g" using HP Load Runner 11. Could you tell me which is the best protocol to record script?
I am looking for a protocol which I can parametrize after recording.
These are recommended protocols from Load Runner when I run Protocol Analyzer
AJAX(Click and Support) : Seems to work fine but Can't parametraize Query after recording
Java Over HTTP
Web (HTTP/HTML)
There are other protocols too.
Could you provide any pointers or best practices. Thanks a lot
Edited by: 936062 on May 22, 2012 2:56 PM

Hi,
You can refer this forum
[http://groups.google.com/group/LR-LoadRunner/browse_thread/thread/f1fb1c6430e39f94/565baac51a6d35c0]
Thanks and Regards,
Dhaya.G.

Webi XI 3.1 SP2 LOAD TESTING ISSUE WITH LOAD RUNNER

Similar Messages

Maybe you are looking for