Number of scokets for httpsession replication in multi-tiered architecture

          In the weblogic's document on clustering the
          section "Configuration Notes for Multi-tier Architecture" states
          In particular, during peak socket usage, each WebLogic Server in
          the cluster that hosts servlets and JSPs may potentially use a
          maximum of:
          Two sockets for replicating HTTP session states between primary
          and secondary servers.
          I don't understand why 2 sockets are required. Won't just one do the job. What's
          the 2nd one for?
          pradeep
          

          hi,
          The doc mentions toally 3 sockets. I have pasted the doc
          content FYI
          "In particular, during peak socket usage, each WebLogic Server in the cluster
          that hosts servlets and JSPs may potentially use a maximum of:
          Two sockets for replicating HTTP session states between primary and secondary
          servers, plus
          One socket for each WebLogic Server in the EJB cluster, for accessing remote objects
          Toally 3 sockets. 2 for httpsession replication and 1 for accessing remote objects.
          Why 2 sockets are required for replicating HTTP session states between primary
          and secondary servers
          pradeep
          Kumar Allamraju <[email protected]> wrote:
          ><!doctype html public "-//w3c//dtd html 4.0 transitional//en">
          ><html>
          >the docs also says
          ><p>One socket for replicating HTTP session states between primary and
          >secondary
          >servers, plus
          ><p>One socket for each WebLogic Server in the EJB cluster, for accessing
          >remote objects
          ><p>pradeep bhat wrote:
          ><blockquote TYPE=CITE>In the weblogic's document on clustering the
          ><br>section "Configuration Notes for Multi-tier Architecture" states
          ><p>In particular, during peak socket usage, each WebLogic Server in
          ><br>the cluster that hosts servlets and JSPs may potentially use a
          ><br>maximum of:
          ><br>Two sockets for replicating HTTP session states between primary
          ><br>and secondary servers.
          ><p>I don't understand why 2 sockets are required. Won't just one do the
          >job. What's
          ><br>the 2nd one for?
          ><p>pradeep</blockquote>
          ></html>
          >
          

Similar Messages

  • Strategy for ensuring replication is complete before initating query from source system

    Hi,
    I am using HANA as a side-car scenario with reports running in SAP ECC being accelerated by querying replicated tables in SAP HANA instead. This works well, however I don't have a good mechanism to validate before running a report whether the underlying data has already been replicated to HANA or is still queued up.
    Often users would want to run the reports soon after large data changes have been made in the source tables. It is unknown, based on the overall workload, how long it might take for the most recently written records to get replicated to the corresponding HANA table.
    What is a good practice approach to handle this. I have seen some separate threads on doing record counts between HANA and ECC. I think that is not a good idea at all. Firstly, for large tables, the time overhead of doing the record count is very large. In the time it takes for me to query the record count in ECC, the HANA report could have run 10 times over. But more importantly, for very large source tables, I may have opted to only replicate more recent data, leaving old historical data from ECC un-replicated to HANA.
    I know this is not a new problem. SAP must have already addressed it a number of ways for their own delivered application accelerators. The COPA accelerator for instance must be doing something along these lines. Possibly querying most recent records in ECC and comparing them to the most recent records in HANA for the same tables might be a way to go.
    Does anyone else have insights into how to best approach this? Does SLT expose a mechanism to check whether replication is completed for any given table?
    thanks,
    Nitin Goel

    Hi Nitin,
    SLT can never confirm whether replication is completed for any given table..Replication is a continous process,if records are there in the logging table of ECC then that will be replicated via SLT.
    So the best way to check Replication completion is :
    1. Go to SE16(ECC)--give the Logging table for which you want to check the replication.Copy the Logginng table from LTRC.If the number of entries is 0,then nothing is there in the logging table.
    You can verify the replication in LTRC(SLT)--Expert functions to check replication is working fine or not.
    2. Then go toi SE16(ECC)--Check the number of entries of the original table
    3. Check in HANA--The number of entries should be same as the source table.
    It takes a minute to verfify each table replication,not more than that.
    Regards,
    Joydeep.

  • Why we need crm middleware for material replication?

    Hi Experts,
    Why we need CRM middleware for material replication from ECC to SRM when we can do the vendor replication from ECC to SRM with ABAP report? IS there any technical limitation?
    Thanks and regards,
    Ranjan

    Hello,
    Vendors and locations, for example, you can replicate from ECC to SRM using ABAP reports.
    However, if you want to replicate the following information, you should use the middleware:
    - DNL-CUST-BASIS3 - Basis CRM online
    - DNL-CUST-PROD0 - Material number conversions
    - DNL-CUST-PROD1 - Product : categories
    - DNL_CUST_SRVMAS - Customizing: Service Master
    - MATERIAL - R/3 Material Master
    - SERVICE_MASTER - R/3 Service Master
    Check note 872533. It provides more information.
    Regards,
    Ricardo

  • Maximum number of nodes in Multimaster Replication

    Has anyone setup a multimaster replication environment on 8.1.7 with more than 11 nodes?
    What could be the maximum number of nodes for reliable operation?

    Hi,
    With Oracle 8.1 you can have upto 9; but I think you can have under one of the replicated DB another replicated DB. tree like tructure.
    Best,
    EA

  • Is there a need for Partial Replication ?

    Partial replication is the possibility to replicate only a subset of a DIT. It can be Fractional Replication (Subset of attributes per entry), Sparse (Subset of entries) or Subtree (Only a portion of the tree).
    If you have needs for partial replication, it is time to discuss with us about your requirements, your replication scenarios...
    Ludovic Poitou
    Directory Server Replication Architect.

    There are a number of situations where partial replication would be beneficial to most organizations.
    Fractional:
    Various differing national laws prevent companies from sending personal information (such as home addresses and telephone numbers, and government identification numbers) into or outside of particular countries. If you wish to geographically distribute your directory services for performance reasons, non-fractional replication would make that a manual or scheduled process, likely with an LDIF intermediary of some kind. From our perspective, the need for this type of partial replication is greatest.
    Sparse:
    This could again be useful for companies with geographical data distribution concerns, or with specialized/purposed directory servers. How would this be accomplished - using LDAP search filters, or some other method?
    Subtree:
    We would be particularly interested in this type of replication for security reasons. For example, we want to have sensitive authentication/authorization information used for private administrative applications or high-security applications (such as a PKI) stored only on secured servers over which we have physical control. This would also allow us to deploy single-purpose servers (containing only a subset of our entries) where creating an additional database is not a viable option.
    As our environments grow, in number of servers and in global reach, the ability to perform replication in "untraditional" (<i>i.e.</i>, non-database) ways becomes a significant product differentiator.

  • Wanting to deactivate PS Elements on Mac and transfer program to a friend. Is it OK to download a trial to my friend's Mac and use my serial number.  Also is the serial number the same for Mac and Win versions?  Many thanks.

    Wanting to deactivate PS Elements on Mac and transfer program to a friend. Is it OK to download a trial to my friend's Mac and use my serial number.  Also is the serial number the same for Mac and Win versions?  Many thanks.

    Yes it should work transferring PSE10 from Mac to Mac. There is a download link below for PSE10, which your friend could use. There is a formal licence transfer process but it’s not necessary for your friend to register the product (activation is sufficient) although he couldn’t benefit from future upgrade pricing without registration. Serial numbers typically can only be registered one time.
    Boxed versions of PSE are normally multi-platform with disks for both Windows and Mac; so the serial number is the same. Downloads have tended to be single platform only although I believe the current version PSE12 will work on both Windows and Mac with the same serial number – boxed or download.
    http://helpx.adobe.com/photoshop-elements/kb/photoshop-elements-10-11-downloads.html

  • OGG Sybase for Sybase Replication

    Hi!
    I need to know cases of OGG with Sybase for Sybase replication, if possible with details of topology used, transactional model and data volume.
    Can anyone share this information?
    Best & Regards,
    Adelmo

    It's not going to be much different than any other RDBMS.
    http://download.oracle.com/docs/cd/E18101_01/doc.1111/e17806.pdf
    Plus, like what is stated several times over in the Admin guide:
    Data volume
    The standard configuration is sufficient if:
    ● The transaction load is consistent and of moderate volume that is spread out more or
    less evenly among all of the objects to be replicated.
    and...
    ● There are none of the following: tables that are subject to long-running transactions,
    tables that have a very large number of columns that change, or tables that contain
    columns for which Oracle GoldenGate must fetch from the database (generally
    columns with LOBs, columns that are affected by SQL procedures executed by Oracle
    GoldenGate, and columns that are not logged to the transaction log).
    If your environment does not satisfy those conditions, consider adding one or more sets of
    parallel processes. For more information, see the Oracle GoldenGate Windows and UNIX
    Troubleshooting and Tuning Guide.

  • SQL help: return number of records for each day of last month.

    Hi: I have records in the database with a field in the table which contains the Unix epoch time for each record. Letz say the Table name is ED and the field utime contains the Unix epoch time.
    Is there a way to get a count of number of records for each day of the last one month? Essentially I want a query which returns a list of count (number of records for each day) with the utime field containing the Unix epoch time. If a particular day does not have any records I want the query to return 0 for that day. I have no clue where to start. Would I need another table which has the list of days?
    Thanks
    Ray

    Peter: thanks. That helps but not completely.
    When I run the query to include only records for July using a statement such as following
    ============
    SELECT /*+ FIRST_ROWS */ COUNT(ED.UTIMESTAMP), TO_CHAR((TO_DATE('01/01/1970','MM/DD/YYYY') + (ED.UTIMESTAMP/86400)), 'MM/DD') AS DATA
    FROM EVENT_DATA ED
    WHERE AGENT_ID = 160
    AND (TO_CHAR((TO_DATE('01/01/1970','MM/DD/YYYY')+(ED.UTIMESTAMP/86400)), 'MM/YYYY') = TO_CHAR(SYSDATE-15, 'MM/YYYY'))
    GROUP BY TO_CHAR((TO_DATE('01/01/1970','MM/DD/YYYY') + (ED.UTIMESTAMP/86400)), 'MM/DD')
    ORDER BY TO_CHAR((TO_DATE('01/01/1970','MM/DD/YYYY') + (ED.UTIMESTAMP/86400)), 'MM/DD');
    =============
    I get the following
    COUNT(ED.UTIMESTAMP) DATA
    1 07/20
    1 07/21
    1 07/24
    2 07/25
    2 07/27
    2 07/28
    2 07/29
    1 07/30
    2 07/31
    Some dates donot have any records and so no output. Is there a way to show the missing dates with a COUNT value = 0?
    Thanks
    Ray

  • How to file auto number in save for web ?

    Hi All,
    how to file auto number in save for web ?
    I have make an action script in Photoshop cs5 that
    change the image resolution and SaveForWeb...
    (but at this point user need to supply the file name)
    so, How to write the code/action such that when the
    action button is click the image will save into the file
    file2eMail_001,file2eMail_002,file2eMail_003,file2eMail_004
    or to any unused number ?
    thank for any help!

    You could also ask over in the Photoshop Scripting Forum, I’m pretty certain stuff like progressively numbered copies has been addressed there previously.

  • ALV report shows fixed number of columns for a user ?

    Hello,
    I have an issue wherein I have created an ABAP ALV grid display in a WD ABAP application and hosted it onto portal. Now, all the users can see the correct number of columns for the report which are determined dynamically. But just for one user, irrespective of which report she launches it shows her only 4 columns.
    Can anybody suggest why this kind of behavior.
    Appreciate your help.
    Regards,
    Samta.

    Hi,
    Right, click the button "manage layout" or "save layout" and remove any default user settings...
    added: didn't read carefully sorry ;)... better check user settings in SU01 if no ALV parameter is used.
    Kr,
    m.
    Edited by: Manu D'Haeyer on Oct 21, 2011 8:53 PM

  • FMS to Find total number of Quantity for all Sales Orders

    Hi Experts,
    I am trying to get the total number of quantities for the current document. Kindly help me out how to find it out for the current document.
    Currently I am giving the document no. to the query by modifying the query every time before executing it.  but how can i do this without giving the document numbers and executing the query directly.
    Kindly help ...
    Thanks and regards,
    Amit.

    Hi,
    Please follow below steps to get total quantity.
    1. Run below query first in query generator
    SELECT sum(T1.[Quantity]) FROM ORDR T0  INNER JOIN RDR1 T1 ON T0.DocEntry = T1.DocEntry WHERE t0.docnum = $[ORDR.docnum]
    2. After execute, save it under query manager with required name
    3. Open sales order and run saved query
    4. Total quantity will be displayed in query result.
    refer below screen shot:
    Thanks & Regards,
    Nagarajan

  • Different number of rows for different columns in JTable

    hi
    I need to create a JTable with different number of rows for different columns...
    Also the rowheight should be different in each column...
    say there is a JTable with 2 columns... Col1 having 5 rows and column 2 having 2 rows...
    The rowHeight in Col2 should be an integer multiple of Rowheight in Col1
    how do I do this ??
    can anybody send me some sample code ?????
    thanx in advance

    How about nesting JTables with 1 row and many columns in a JTable with 1 column and many rows.
    Or you could leave the extra columns null/blank.
    You could use a GridBagLayout and put a panel in each group of cells and not use JTable at all.
    It would help if you were more specific about how you wanted it to appear and behave.

  • How to restrict number of characters for an input field

    Hi All,
    I have an input field.
    The max number of characters for this input field is 10.
    when a user enter more then 10 characters. it should prompt for an error or the input field should not allow to accpet the 11 character.
    how we do this in VC.
    need your helpful answers
    Rgds
    Srinivas

    Hi Srinu
    You could achieve this by configuring error messages under the formula:-
    Select the control properties and in the Input field at the Display tab write the formula
    "IF((@<LEN(text)>10),'appropriate message','Records available')"
    Note :- there is a LEN(text) under text functions in formula tab.
    Regards
    Navneet
    Message was edited by:
            Navneet Giria

  • How to calculate number of rows for perticular characterstic in SAP BI Bex

    Hi experts,
    Please let me know how to calculate  ' number of rows  ' for perticular characterstic in Bex query. 
    Thanks & Regards,
    Babu..

    Hello,
    You can try this
    Create a CKF and assign the vale 1 to it. Open the query and select Character where you want to display ' number of rows ', go to properties windows, select 'display', in the results row drop down box, select  'always display'.
    Thanks.
    With regards,
    Anand Kumar

  • In what case, we need to use PI for vendor replication in MM-SUS?

    Dear Experts,
    In what scenarios, we need to use or configure the PI settings for vendor replication between MM and SUS? The SAP configuration guide for SRM 7.01 mentions about the PI settings for vendor replication from ECC to SUS.
    However, we are able to replicate the vendor from ECC to SUS using the program BBPGETVD in SUS client?
    In what circumstances, we need to configure the PI settings for vendor replication when we are able to do so in a much simpler way?
    Thanks in advance,
    Ranjan
    Ranjan Sutradhar

    Hi Ranjan,
    In all circumstances you need to replicate vendors from MM to SUS via PI. BBPGETVD is not the right way to replicate vendors to SUS because the information like vendor class, assignment of this vendor class to an object and the vendor classification cannot be transferred via BBPGETVD.
    Hence you will have to use PI. You can try sending a PO from MM to SUS and the PO will fail in SUS with a message like 'Vendor not present in SUS' if you use BBPGETVD.
    Regards,
    Nikhil

Maybe you are looking for