Portal Performance

Hello,
we are running on EP6 SP2 Patch 3 and KMC Patch 3 (Windows machines). We are using an Active Directory server to authenticate against it with a custom login module. The Active Directory Groups are mapped to Portal Roles. There is a role concept in place on which content pages the users see. We have lots of custom iViews which display KM documents (HTML).
We are experiencing problems with the performance of the Portal since couple days now. Log in takes now longer than usual. When logged in, navigating from page to page is very, very slow. A KM Navigation iView loads even longer up to couple minutes to show its content. Couple days ago this behaviour did not exist.
When I logged in as a content administrator and browse through the KM Explorer it takes takes forever, for example clicking one KM folder could take up to 5 min to open it.
This situation seems unsual and cannot be explained. Did anyone experienced the same issue and found the problem maker? Does anybody has an suggestion on how to improve the performance in the content management component?
I would very much appreciate a suggestion that will allow us to improve the performance.
Thanks,
Erik

Erik: is the portal server loaded when you experience the slow performance? That would be the first thing I would check. If the answer is yes, try to find what is causing the load (see portal monitor -> thread activity and get a JVM thread dump). If not, I would figure out what causes the delays. If it is a network issue, check you cabling, port settings (half duplex) and network drivers. If it isn't a network setting it might also be a timeout / caching issue that has been triggered by changes not done by you (database, LDAP, etc).

Similar Messages

  • Portal Performance is Very Slow..Pls Help

    Hi to all.
    Our machine is Sun280R - Sun Solaris 8..
    Using Oracle 9iAS 9.0.2.0.1..
    Our portal is 9.0.2.0.1 Release 2.
    All installed in one machine (infs/apps)
    We have difficulties with the Portal performance.
    It takes slow time to login, and view the pages.
    and when we click the Customize button, it will be more slower (50 sec)...
    Can anybody help?
    How to check what is wrong here?
    Pls..this is urgent..
    Thanks.

    you need to type in the name of a valid tablespace in the database where you are trying to install the owa_perf tables.
    Like it says in the setup guide (readme.htm) that comes with the scripts..
    "During the installation process, you will be asked for three tablespace destinations for:
    * the log tables
    * the log indices
    * the log materialized views
    It is recommended, though not mandatory, that you create and specify individual tablespaces for the storage of these elements. You cannot use DEFAULT as a valid tablespace for these responses."
    ...so you need to type in the name for a valid tablespace for each entry, you can type the same name but it asks you for three choices because each one can get large when your logs are large, asking for them separately would allow you to place them on separate disks or tablespaces that you could drop independently of other data.

  • Portal performance monitoring scripts : (Unable to generate reports)  HELP

    Hi,
    Using 10.1.2.0.0
    I followed README.html document to load the logs files to generate reports for Portal Performance.
    First of all while running loadlogs.pl I keep getting the following error. I even tried adding -nodirect but still gets the same error. Don't know why. But it looks like there is some data loaded in OWA_LOGGER table
    C:\ORACLE_PRODUCTS\PORTAL_AS\portal\admin\plsql\perf\loader>perl loadlogs.pl -logical_host localhost -connection owa_perf/owa_perf@orcl -http_logfile C:\ORACLE_PRODUCTS\PORTAL_AS\Apache\Apache\logs\error_log.1130457600 -webcache_logfile C:\ORACLE_PRODUCTS\PORTAL_AS\webcache\logs\access_log -oc4j_logfile C:\ORACLE_PRODUCTS\PORTAL_AS\j2ee\OC4J_Portal\application-deployments\portal\OC4J_Portal_default_island_1\application -nodirect
    25-Oct-05 13:20:17, Copying abc:C:\ORACLE_PRODUCTS\PORTAL_AS\Apache\Apache\logs
    \error_log.1130241600
    25-Oct-05 13:20:17, Loading C:\DOCUME~1\whitesox\LOCALS~1\Temp\abc_error_log.1130
    241600.20051025.132017
    25-Oct-05 13:20:21, Copying abc:C:\ORACLE_PRODUCTS\PORTAL_AS\j2ee\OC4J_Portal\a
    pplication-deployments\portal\OC4J_Portal_default_island_1\application
    25-Oct-05 13:20:21, Loading C:\DOCUME~1\whitesox\LOCALS~1\Temp\abc_application.20
    051025.132021 -nodirect
    SQL*Loader-350: Syntax error at line 127.
    Token longer than max allowable length of 258 chars
             end",
            ^
    25-Oct-05 13:20:22, Copying abc:C:\ORACLE_PRODUCTS\PORTAL_AS\webcache\logs\acce
    ss_log
    25-Oct-05 13:20:31, Loading C:\DOCUME~1\whitesox\LOCALS~1\Temp\abc_access_log.200
    51025.132022Then I ran reports.sql but I don't see any reports being generated, but running this script did populate some other tables. I tried running some other scripts also but somehow I don't see any reports being generated as opposed to what is said in the README.HTML document i.e. "A sample web page (reports.html) is included which provides links to the generated reports.". How really I get to see the reports, where are the reports generated, is it something else that I am missing. No matter what script I run I don't see any report being generated. The document is not so clear. Can someone please help me out here. Thanks

    Hi!
    You have to change to directory
    ORACLE_HOME$/portal/admin/plsql/perf/scripts
    (you can find reports.sql in it) before you run reports.sql script!
    It will produce several .txt files.
    After running the script just open reports.html, that will point the generated files.
    A better place to ask questions like this:
    Portal Performance and Scalability
    http://forums.oracle.com/forums/forum.jspa?forumID=15

  • Portal performance report

    Hi Portal GURUs,
    We have configured Portal Activity Reports for monitoring hits but we required to monitor Portal Performance.
    Please suggest me if there is any in built features to get these reports?
    Regards
    Kiran

    Hi,
    you can monitor the portal performance by using  CCMS (and GRMG).
    Below link will also help you.
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/48fead90-0201-0010-6e83-b43f5dd4d338https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/3ba6a290-0201-0010-d684-c94b1c765ae9
    Raghu

  • Portal Performance Trend Report

    Hi Portal Gurus,
    I have one request.
    For ABAP Based system we can get historical performance report.
    But for Portal , how we can get daily / weekly or monthly based report for performance.
    We have Mercury Tool to get report for few 'Pages' based on users & response time.
    I have checked response summary from monitoring. But we have multiple dialog instance & node.
    Collecting data from there is not choice.
    I'm looking for better way to way to get portal performance.
    Regards
    Sumanta Chatterjee

    Hi
    There are some portal activity report available in portal.chk user administration->activity report.Also you can customize this report according to your requirement.Basically a user activity report.
    Also in netweaver administration (http://<portal url>:<port>/nwa)you can access all kinds of usage data.
    Shankar

  • Portal performance KPIs

    I need to define an SLA for Portal performance.
    What would be the top KPIs or metrics for this?
    What are the tools/reports to measure it?

    Hi,
    Usually customer expect in SLA to be determined :
    a)  average response time (or processing time) KPI
    This KPI should be carefully defined for well-described process dialog steps (for UI based scenarios), as well as for well-defined amount of processed data.
    For background jobs processing time can be defined, with direct relation to the volume of data to be processed.
    The network environment (LAN/WAN) has significant impact on response time. The system hardware (CPU, Memory, Disk I/O) too. This means that you have to describe the conditions (fast CPU, enough memory, good network, etc) with as concrete as possible parameters, directly in the SLA together with the KPI.
    b) processing throughput or capacity , e.g. number of concurrent users, parallel tasks, orders per hour, and so on
    The KPI of how many users I can run on the system sometimes is a KPI that customers require. This, similar to the response time KPI, should be documented together with the hardware requirements and the volume of data requirements.
    Regards,
    Markus

  • Improve Portal performance with BI-Java by adding additional server node.

    Hi SAP Expert
    I wonder anyone might have attempted this before,
    In an environment with BI-Java and Portal Java stack installed on the same server, is it possible to improve the system performance by adding additional portal server node? or is there any improvement seen by adding the extra server node?
    I am after suggestion and experience on how the portal performance can be improve, with consideration of BI-Java sharing Portal resources.
    Any comment will be most appreciated.

    Hi Jim,
    We've this configuration at our site, Portal and BI running together (not federated).  Recommendations would be to:
    1. Set a sensibly large max heap size on each server node of Portal (at least 2GB if not larger) and implement several nodes at least.  For example we have 4 x physical nodes, each running 3 server nodes of 2GB max heap size apiece so a total of 12 nodes.
    2. Implement the BI Safety Belt for large results sets (1127156)
    3. Implement the latest JVM / JDK you can on your environment.  We found much improved performance after implementing JDK SR10 with the J9/2.3 options enabled e.g.,  -Xjvm:j9vm23
    -Xsoftrefthreshold0
    4. Patch your BI components on the Portal (BIBASES, BIWEBAPP etc) to the latest available patch level for your Portal SP level
    5. Make sure all your RFC connections are load balanced to the backend and tuned for the kind of load you expect on your reports, and the BI is sized appropriately in terms of app servers and dialog work processes, RFC-enabled login groups in SMLG etc.
    6. Consult notes 1048691 (especially useful, opions &PROFILING=X&TRACE=X added to reports for example for performance tracing), 937697, 1021921, 948158 for information about problem analysis for this scenario
    7. Implement SAP Web Dispatcher for Portal to reduce load on static mime files
    8. Tune the Portal application.  There's plenty information on SDN related to Portal performance tuning.
    I hope this helps!
    Cheers,
    Marc

  • SSL impact on Portal performance.

    I was wondering if anyone knows of any benchmarks of Portal performance with SSL enabled versus a non SSL enabled Portal.
    I realize that SSL does create some additional overhead on the server during the encyrpt/decrypt process. I'm trying to determine if a dedicated reverse proxy may be more beneficial to support approximately 1000 concurrent connections.
    Thanks in advance, and as always points awarded.

    I haven't seen any official figures, but you also have to consider the applications running in the portal. For example, if you plan to encrypt SAPGUI for HTML, that will have an imapact etc.
    Reallistically you need to do your own tests, as relying on someone else's figures based on their specific portal applications may lead you to problems later on.

  • Supply recommendations for Improvement of portal performance and create POC

    HI All
    i am not aware of Supply recommendations for Improvement of portal performance and create POC*s, any body let me know wht POC means.
    thanks
    Udhaya.

    POC can stand for a lot of different things: http://www.answers.com/topic/poc
    In a supply management context, it could be Purchase Order Change or Purchase Order Confirmation.
    Regards,
    Sean

  • How to restrict logging on portal (Performance issue)

    Hi guys,
    does anyone know how to restrict logging of the portal? Even though I made a change on Visual Administrator -> Services -> Log Configurator setting ALL categories to severity = 'Error', I keep getting 'Info' severity traces logged in defaultTrace.trc (viewed by Logg Viewer). Which in practise means tons of logged trace information! I think this might be one of the reasons we have been experiencing performance issues lately.
    Does anyone know how to restrict this logging?
    Any help greatly appreciated!
    Cheers,
    Frank

    Hi,
    I would not only set the severity of categories to 'Error' but also locations to 'Error'.
    A log controller - can be a "category" or a "location".     Categories are used for "logging" and represent problem areas such as network, database, security, user interface, etc. Identified by a slash-separated ('/') path of names, prefixed with an additional slash.     Locations are used for "tracing" and correspond to places in the development components. Identified by a dot-separated ('.') path of names, which is often the name of the respective java package.
    Greetings,
    Praveen Gudapati

  • Calculating roundtime for iviews - Portal performance monitoring

    Dear Gurus,
    I have a bunch of iviews that need to be monitored. I am running these BI integrated iviews and want to calculate the rountime for them. currently the way I am doing it, is I enter the variable values for each of the iviews and run them, I switch on a stop clock and then wait till the results come up in the portal. The average run time right now is around 8 mins. This is not because of the portal but because of the time lapse in the back end as well.
    Are there any performance parameters I can activate, or a log I can look into to see the rountime for the iviews without having to manually track the time.
    Any help is appreciated, and generous points for useful answers.

    The method System.currentTimeMillis() returns the current time in milliseconds of type LONG.
    Define a long attribute in the component controller context. map it to the first and last views in the iview. Then in the doInit of the first view of the iview, set it's value to System.currentTimeMillis(). Then, at the end of the last view, use the formula
    Long l = System.currentTimeMillis() - wdContext.currentContextElement().get<YourLongAttribute>();
    This will give the roundabout time taken by the entire iview to execute.
    Divide it by 1000 and u can get the time in seconds, and then divide it by 60 and it gets converted into minutes. print it using ur message manager.
    Below is the small snippet of code i used.
    // Init of First View in the IView
    public void wdDoInit()
       //@@begin wdDoInit()
       wdContext.currentContextElement().setTime(System.currentTimeMillis());
       //@@end
    // DoExit of last View in the IView
    public void wdDoExit()
      //@@begin wdDoExit()
      long l = System.currentTimeMillis() - wdContext.currentContextElement().getTime();
      float f = (float)l/(float)1000;
      wdComponentAPI.getMessageManager().reportSuccess("Time Taken by the IView "f" Seconds");
      //@@end
    Note: wdDoExit() was My last method to get executed. if u use something else, substitute this there.

  • Docs. for Monitoring Installing Portal Performance steps/scripts etc.......

    Hello can someone please point me to the documentation on how to setup the Portal (10.1.2) for generating and loading performance reports. What files do I need to modify and what scripts to run. I want to generate reports to check the performance of the Portal pages, apps. and overall performance.
    Thanks

    I ran the following command the data got loaded but I am not sure if all the data is in the tables due to the following error. Can someone please tell me what's causing the following error occurs when running loadlogs.pl:
    SQL*Loader-350: Syntax error at line 127.
    Token longer than max allowable length of 258 chars
             end",
            ^Here is the complete output:
    C:\ORACLE_PRODUCTS\PORTAL_AS\portal\admin\plsql\perf\loader>perl loadlogs.pl -lo
    gical_host localhost -connection owa_perf/owa_perf@orcl -http_logfile C:\ORACLE_
    PRODUCTS\PORTAL_AS\Apache\Apache\logs\error_log.1130241600 -webcache_logfile C:\
    ORACLE_PRODUCTS\PORTAL_AS\webcache\logs\access_log -oc4j_logfile C:\ORACLE_PRODU
    CTS\PORTAL_AS\j2ee\OC4J_Portal\application-deployments\portal\OC4J_Portal_defaul
    t_island_1\application
    25-Oct-05 13:20:17, Copying localh:C:\ORACLE_PRODUCTS\PORTAL_AS\Apache\Apache\logs
    \error_log.1130241600
    25-Oct-05 13:20:17, Loading C:\DOCUME~1\admin\LOCALS~1\Temp\abc_error_log.1130
    241600.20051025.132017
    25-Oct-05 13:20:21, Copying localh:C:\ORACLE_PRODUCTS\PORTAL_AS\j2ee\OC4J_Portal\a
    pplication-deployments\portal\OC4J_Portal_default_island_1\application
    25-Oct-05 13:20:21, Loading C:\DOCUME~1\admin\LOCALS~1\Temp\abc_application.20
    051025.132021
    SQL*Loader-350: Syntax error at line 127.
    Token longer than max allowable length of 258 chars
             end",
            ^
    25-Oct-05 13:20:22, Copying localh:C:\ORACLE_PRODUCTS\PORTAL_AS\webcache\logs\acce
    ss_log
    25-Oct-05 13:20:31, Loading C:\DOCUME~1\mine\LOCALS~1\Temp\abc_access_log.200
    51025.132022Thanks!

  • Improving Portal Performance

    Hi,
    We are looking at improving our Portal's performance.  At the moment, it seems like we can achieve this by adding either Java Instance or Server Instance.  I have read that if there are frequent access to the Portal, it would be best that additional server instances/processes can be added.  I have followed the guidelines in help.sap.com to add server instances.  However, I am not able to find any documentation which recommends/advices on how many server instances a server needs in order to achieve maximum performance.  Would it be as simple as 1 server process per processor?  Hence, assuming that I have a 4-CPU server, I should at least have 4 server processes?
    Aside from looking at adding Java instances and Server instances, we are also increasing the heap size of each local disks. 
    Any other pointers to which we can further improve the performance of the server are welcomed.  Please excuse my misuse of terms and poor-understanding of concepts.  Please feel free to correct as needed.  If there is any other posts or boards I can refer to, please feel free to redirect.
    Thank You.

    Hi,
    Note that a java process cannot as far as I know use more than 1 CPU. Therefore, if you have 4 CPUs and only two java process, you basically have to idle CPUs.
    Memory is also important. Normally, java cannot address more than 2 GB of memory, and you don't really want it to, because the full garbage collection that will eventually occur, will block the entire JVM for many seconds (the more memory, the more time).
    Regards
    Dagfinn

  • SAP NetWeaver 2004 Portal performance

    Hi
        can anybody say how do we improve the performance of SAP NW 04
    with regs
    vij

    Hi Vijay
    Check this links ..to see whether it will help you out.
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/webcontent/uuid/10986f61-0a01-0010-bdad-cdd93247dbc1 [original link is broken]
    https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/30c221aa-b4e5-2910-b899-d268d84c19c5
    https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uui
    Regards
    Nissy
    Pls give points if useful.

  • Portal Performance - Potential Solutions?

    In an Article about Performance testing found on the SDN
    https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/969fded9-0901-0010-77b6-a05d645ca05c
    It talks about the J2EE engine being configured with multiple J2EE server instances :
    "We found that using multiple J2EE server instances was a suitable approach to make the best use of the large number of CPUs. The J2EE engine was configured with 4 server instances. These J2EE server instances were configured as standalone cluster. The J2EE servers process the most of the workload in this scenario. Having many processes handling smaller chunks of the work as opposed to have fewer processes handling larger chunks each was found to be the best approach to achieve scalability."
    I have a out-of-the-box install of EP6.0, and I can open the config tool (usrsap<INSTANCE>JC00j2eeconfigtool) and see that I have just one server instance and one dispatcher.
    I can click "Server" -> Add Server
    and a new Server_ID<> is added (just like magic huh )
    My question: Does this make a "Standalone Cluster" or are there other settings I must attend to in order to get the increased performance I hope to get... or at least have the system share the resource requests between the two server processes.
    Any assistance, comments or documentation you can point me to is appreciated.

    Hi Marty,
    I am not familiar with the term 'Standalone Cluster', but having multiple server nodes is indeed a way to improve performance.
    If you add server processes via Config tool, you should be aware that changes you made to parameters using ConfigTool prior to the addition of the new server processes, are not reflected to the new server processes, ie. you'll have to change these settings by hand.
    For more info, see http://help.sap.com/saphelp_nw04/helpdata/en/2e/611724f410254ca12a3f396ec5ae85/frameset.htm
    Best regards,
    Robin van het Hof

Maybe you are looking for

  • HT2513 Is there any way to sync between laptop, desktop, iPhone and iPad without using the iCloud?

    I want to sync my iCal between all my Apple products but this doesn't seem to be possible? It all used to sync with MobileMe, but I didn't upgrade, so it stopped syncing. If I use a 3-rd party calendar, say Google Calendars, in my iCal would that wor

  • How do I move my iWeb website to a new machine?

    Greets all, I am currently using a buddies MBP while I'm away from home (deployed to Baghdad) lucky Me Anyway, he just upgraded his to Leopard and iLife '08. I have had a little free time lately and decided to play with iWeb...I like it. So, I will m

  • Shortcut keys not working

    On my Mac (OSX 10.9.5), the 'step backwards' shortcut key (cmd-alt-z) isn't working; nothing happens. On my Windows (Windows 7) pc, the 'step forwards' shortcut key (ctrl-shift-z) isn't working. If I go into the shortcut menu and change the shortcut

  • New ipod nano...

    I had an older ipod nano and recieved a new one for Christmas. Can I still use the music library from the old one on the new one?

  • TableView LINEEDIT works differently when viewed through Portal

    I have a simple table with a few columns of data that I display and allow the user to change a row using the selectionMode = "lineedit".  The columns are all controlled using htmlb tableViewColumns. When I test my application from the WebAs the table