HTTP Sessions tracking using in-memory session replication

          I am using wls5.1. I have a situation where the "Java Heap size
          used" exceeds 100%, then it drops down to maybe 20% (after garbage
          collection). What happens to the active Http sessions when using
          in-memory session replication with persistance set to false.
          Are the sessions written to disk? If so, where?
          Are the sessions, before the garbage collection, still active?

Also looks like either your session invalidation time is too high or your
          heap size is too small. Make sure you tune either of this paramater
          properly.
          Viresh Garg
          Principal Developer Relations Engineer
          BEA Systems
          Prasad Peddada wrote:
          > Replicated session data is never written to disk. I dont' understand
          > your second questions. If the session is active and if GC kicks in it
          > won't delete the object since the servlet engine has reference to that
          > session object.
          >
          > Make sure you never swap and your application has enough heap.
          >
          > -- Prasad
          >
          > chris mckinnon wrote:
          >
          > > I am using wls5.1. I have a situation where the "Java Heap size
          > > used" exceeds 100%, then it drops down to maybe 20% (after garbage
          > > collection). What happens to the active Http sessions when using
          > > in-memory session replication with persistance set to false.
          > > Are the sessions written to disk? If so, where?
          > > Are the sessions, before the garbage collection, still active?
          >
          > --
          > Cheers
          >
          > - Prasad

Similar Messages

Session Tracking using Struts

Hello,
I have a small struts application that i am trying to add user validation to. This is what I have done, but I'm not sure if this is the best way. Bear in mind that I am new to struts and JSP.
In may action class I verify the user and create a cookie
public ActionForward execute(
        ActionMapping mapping,
        ActionForm form,
        HttpServletRequest request,
        HttpServletResponse response)
        LoginForm loginForm = (LoginForm) form;
        if(loginForm.getUserName().equals("Sean"))
            Cookie userCookie = new Cookie("userName", loginForm.getUserName());
            response.addCookie(userCookie);
            return mapping.findForward("success");
        else
            return mapping.findForward("failure");
    }Then in all other pages, with the exception of my login page I add
     <logic:notPresent cookie="userName">
          <jsp:forward page="/form/login.jsp"/>
     </logic:notPresent>Thanks.

Hello,
I have a small struts application that i am trying to add user validation to. This is what I have done, but I'm not sure if this is the best way. Bear in mind that I am new to struts and JSP.
In may action class I verify the user and create a cookie
public ActionForward execute(
        ActionMapping mapping,
        ActionForm form,
        HttpServletRequest request,
        HttpServletResponse response)
        LoginForm loginForm = (LoginForm) form;
        if(loginForm.getUserName().equals("Sean"))
            Cookie userCookie = new Cookie("userName", loginForm.getUserName());
            response.addCookie(userCookie);
            return mapping.findForward("success");
        else
            return mapping.findForward("failure");
    }Then in all other pages, with the exception of my login page I add
     <logic:notPresent cookie="userName">
          <jsp:forward page="/form/login.jsp"/>
     </logic:notPresent>Thanks.

Session tracking and Internet Explorer

Hi,
I am currently maintaining a servlet application, on apache/jserv.
This application implements a session tracking using a shared static hashtable of session data, associated with session id's.
This application may open more than one client browser windows.
With netscape, no problem.
With Internet Explorer, since the version 6, when the client close at least one window, the session is closed.
Thus, the application reject any new request from this client, sent by still open windows (session cannot be retrieved in the hashtable).
Has somebody heard about this problem ?
Thanks for any answer.

Thanks.
In fact, I believe that IE keeps the same session for
child windows.
The problem is: when you click on a link which open a
new window, the new open window share the session with
its parent window.
When the new window is closed, the session is also
closed.
It appears that this mechanism only exists since the
version 6 of IE.No. Earlier IE version handle session cookies the same way.

Always use URL Rewriting for session tracking?

All you JSP guru:
I am working on a JSP project that requires session tracking. I have successfully implements session tracking with both cookies or URL rewriting. I know that with the HttpSession object, it will always try to use cookie first, if that's disabled, then it'll automatically switch to URL rewriting. However, is there a way to force the HttpSession object to ALWAYS use URL rewriting instead of cookies? I have searched for an answer for a long time and haven't been able to found a solution. Is it possible at all? Thank you very much.

i was going to say that WebSphere always uses URL rewriting if you enable it at all, but someone beat me to it (indirectly) :-)
however, that seemed to me to be a violation of the spec, which seemed to imply the behaviour you're describing (only use URL rewriting if cookies are not supported on the current client)
here's a response someone else made on a websphere newsgroup to a statement in that regard:
I believe you are technically correct. However from my
experience, I think the spec if flawed in this area since
there is no reliable way of determining whether the
client browser supports cookies. The authority on
cookies (www.cookiecentral.com) says:
"To properly detect if a cookie is being accepted via
the server, the cookie needs to be set on one HTTP
request and read back in another. This cannot be
accomplished within 1 request."
This is asking too much of a servlet engine
implementation. Even if it did submit a request for this
purpose, the user could refuse the cookie. So
then technically the browser supports cookies, but the
servlet engine infers it doesn't. So if the servlet engine
infers the browser does not support cookies and so
encodes the URL, it is again out of spec because the
browser really does support cookies. By doing it
however encoding is configured makes things simpler,
robust, consistent and avoids the flaw.
My opinion.so, mostly i'm just rambling, but if you're using websphere, you should get the behaviour your boss wants. if you're using something else, i suppose there's a chance it'll "violate" the spec in this same, potentially helpful way.
btw, i remember somebody else complaining that URL rewriting is less secure than cookies, but i kinda think they're about equal. it seems like either could be intercepted by a sniffer and then used to spoof. but i'm no expert in that stuff...

Out of memory session store + in memory replication

I think I know the answer, but I am going to ask anyway.
          Appearantly, if I use file based session persistance, weblogic will manage
          the in-memory session cache and store least recently used sessions to disk
          in order to preserve memory.
          Of course, file based persistance is slower than in-memory replication, and
          it doesn't work so well if we don't have a shared filesystem for our cluster
          (we don't), but the ability to manage memory in this way is very compelling.
          Is it possible to have a hybrid, where session state is replicated between
          nodes using in-memory-replication techniques, but each node only keeps a
          limited number of sessions in memory and swaps the rest out to disk?
          I think the answer is no, but it seems like a good feature.


          "Erik Speckman" <[email protected]> wrote in message
          news:[email protected]..
          > I think I know the answer, but I am going to ask anyway.
          >
          > Appearantly, if I use file based session persistance, weblogic will manage
          > the in-memory session cache and store least recently used sessions to disk
          > in order to preserve memory.
          >
          > Of course, file based persistance is slower than in-memory replication,
          and
          > it doesn't work so well if we don't have a shared filesystem for our
          cluster
          > (we don't), but the ability to manage memory in this way is very
          compelling.
          >
          > Is it possible to have a hybrid, where session state is replicated between
          > nodes using in-memory-replication techniques, but each node only keeps a
          > limited number of sessions in memory and swaps the rest out to disk?
          >
          > I think the answer is no, but it seems like a good feature.
          >
          >
          I don't think you can explicitly tune it, but of course any stuff in memory
          that doesn't get used for a while is going to end up on disk anyway if your
          physical memory is all allocated. Your swap system will probably do as good
          a job of this as being able to specify how many sessions are kept in memory
          would do.
          Loz

Can we use an overloaded constructor of a Java Bean with Session Tracking

Hi Friends,
If any one can solve my query.... It would be helpful.
Query:
I have a Java Bean with an overloaded constructor in it. I want to use the overloaded constructor in my JSP.
1. One way of doing that is to use it directly in the "Scriptlets" (<% %>). But then I am not sure of the way to do session tracking. I think I can use the implicit objects like "session", "request" etc. but not sure of the approach or of the implementation method.
2. Another way is through the directive <jsp: useBean>. But I cannot call an overloaded constructor with <jsp: useBean>. The only alternative way is to use the directive <jsp: useBean> where I have to write getter and setter methods in the Java Bean and use the <jsp: setProperty> and <jsp: getProperty> standard actions. Then with this approach I cannot use the overloaded constructor.
Can any one suggest me the best approach to solve this problem ?
Thanks and Regards,
Gaive.

My first reaction is that you can refactor your overloaded constructor into an init(arguments...) method. Instead of overloaded constructor, you can call that init method. This is the ideal solution if possible.
As to the two choices you listed:
1. This is OK, I believe. You can use scriplet to define the bean and put it into session scope of the pageContext. I am not sure exactly what you meant by session tracking; whatever you meant, it should be doable using HttpSessionAttributeListener and/or HttpSessionBindingListener.
2. Agreed. There is no way that <jsp:useBean> can call a constructor that has non-empty arguments.
Please tell me how it works for you.

How to use session tracking

i am making shopping mall project .
ist page conatins list of product avaiale
2nd page contains list of manufacturuer avaible
problem:-
i want to display on 3rd page the product seleted by user in 1st page
i used session tracking concept.but problem is the value is coming null in third page
please tell me how to solve my problem

If it is like a shopping cart, I suggest you to look for a good shopping cart examples available plenty online.
But if its just about keeping session variables and using them the following works.
Test with a simple example. Have three jsp files like a.jsp, b.jsp and c.jsp.
put the following in a.jsp
<% session.setAttribute("Mobile","Nokia");%>
<%=session.getAttribute("Mobile")%>
<a href="b.jsp">Go to B.jsp</a>Print the value of session variable - <%=session.getAttribute("Mobile")%> in b.jsp and c.jsp
And in b.jsp have a link to c.jsp and so on. Once you set a session variable, it lives as long as your session doesnt expire.
Try it. and also look for more session tracking examples online.
Message was edited by:
passion_for_java

How to find the memory used by specific session in oracle

Hi,
Would require your help to find memory used by specific session/connection in oracle using any query.
**Regards**
**Vijay Salian**

Hi Girish,
which column we have to look at the at in the below ouptut to find the actual memory used by oracle session in our case as mentioned dedicated in ourcase????
SID Oracle User O/S User Session Program Machine PGA Memory PGA Memory Max UGA Memory UGA Memory MAX
3 oracle oracle@alex (LGWR) alex 5,526,516 5,526,516 77,956 77,956
9 oracle oracle@alex (ARC0) alex 4,500,080 4,500,080 77,956 77,956
10 oracle oracle@alex (ARC1) alex 4,500,080 4,500,080 77,956 77,956
11 SYSTEM oracle sqlplus@alex (TNS alex 3,403,988 4,780,244 77,956 1,400,476
2 oracle oracle@alex (DBW0) alex 2,488,624 2,488,624 77,956 77,956
4 oracle oracle@alex (CKPT) alex 1,359,484 1,359,484 77,956 77,956
17 SCOTT oracle sqlplus@alex (TNS alex 577,512 708,584 339,812 339,812
5 oracle oracle@alex (SMON) alex 499,704 499,704 77,956 77,956
8 oracle oracle@alex (QMN0) alex 475,596 541,132 274,348 274,348
19 SCOTT oracle sqlplus@alex (TNS alex 454,964 454,964 143,420 143,420
6 oracle oracle@alex (RECO) alex 237,024 237,024 77,956 77,956
1 oracle oracle@alex (PMON) alex 228,512 228,512 77,956 77,956
7 oracle oracle@alex (CJQ0) alex 228,512 228,512 77,956 77,956
13 rows selected.
Regards
Vijay

How to pass data between two internal sessions using ABAP memory?

Hi,
How to pass data between two internal sessions using ABAP memory?
It would be fine if you could explain with an example.
And also let me clear about the data passing between two main sessions and two external sessions with specific examples.
Thanks.

Hi ,
check the example.
Reading Data Objects from Memory
To read data objects from ABAP memory into an ABAP program, use the following statement:
Syntax
IMPORT <f1> [TO <g 1>] <f 2> [TO <g 2>] ... FROM MEMORY ID <key>.
This statement reads the data objects specified in the list from a cluster in memory. If you do not use the TO <g i > option, the data object <f i > in memory is assigned to the data object in the program with the same name. If you do use the option, the data object <f i > is read from memory into the field <g i >. The name <key> identifies the cluster in memory. It may be up to 32 characters long.
You do not have to read all of the objects stored under a particular name <key>. You can restrict the number of objects by specifying their names. If the memory does not contain any objects under the name <key>, SY-SUBRC is set to 4. If, on the other hand, there is a data cluster in memory with the name <key>, SY-SUBRC is always 0, regardless of whether it contained the data object <f i >. If the cluster does not contain the data object <f i >, the target field remains unchanged.
In this statement, the system does not check whether the structure of the object in memory is compatible with the structure into which you are reading it. The data is transported bit by bit. If the structures are incompatible, the data in the target field may be incorrect.
PROGRAM SAPMZTS1.
DATA TEXT1(10) VALUE 'Exporting'.
DATA ITAB LIKE SBOOK OCCURS 10 WITH HEADER LINE.
DO 5 TIMES.
ITAB-BOOKID = 100 + SY-INDEX.
APPEND ITAB.
ENDDO.
EXPORT TEXT1
TEXT2 FROM 'Literal'
TO MEMORY ID 'text'.
EXPORT ITAB
TO MEMORY ID 'table'.
SUBMIT SAPMZTS2 AND RETURN.
SUBMIT SAPMZTS3.
The first part of this program is the same as the example in the section Saving Data Objects in Memory. In the example, the programs SAPMZTS1 and SAPMZTS2 are called using SUBMIT. You can create and maintain the programs called using the SUBMIT statement by double-clicking their names in the statement. For further information about the SUBMIT statement, refer to Calling Executable Programs (Reports)
Example for SAPMZTS2:
PROGRAM SAPMZTS2.
DATA: TEXT1(10),
TEXT3 LIKE TEXT1 VALUE 'Initial'.
IMPORT TEXT3 FROM MEMORY ID 'text'.
WRITE: / SY-SUBRC, TEXT3.
IMPORT TEXT2 TO TEXT1 FROM MEMORY ID 'text'.
WRITE: / SY-SUBRC, TEXT1.
Example for SAPMZTS3:
PROGRAM SAPMZTS3.
DATA JTAB LIKE SBOOK OCCURS 10 WITH HEADER LINE.
IMPORT ITAB TO JTAB FROM MEMORY ID 'table'.
LOOP AT JTAB.
WRITE / JTAB-BOOKID.
ENDLOOP.
The output is displayed on two successive screens. It looks like this:
and
The program SAPMZTS2 attempts to read a data object TEXT3 from the data cluster "text", which does not exist. TEXT3 therefore remains unchanged. The existing data object TEXT2 is placed in TEXT1. In both cases, SY-SUBRC is 0, since the cluster "text" contains data.
The program SAPMZTS3 reads the internal table ITAB from the cluster "table" into the internal table JTAB. Both tables have the same structure, namely that of the ABAP Dictionary table SBOOK.
Pls. reward if useful.....

How can I use active user session that's in an application scope Hashtable?

First of all, is it possible to use the same session after a user exits and then returns?
Second, if it is possible, then please tell me what I need to do. So far, this is what I have been doing...
1.) The user submits login credentials
2.) The user is authenticated.
3.) The user does not have an existing session, so a new one is created.
4.) The user closes the browser.
5.) The user returns to login page and submits login credentials.
6.) The user is authenticated.
7.) The user has an existing session, so it should be used.
This is where I'm having trouble. All active sessions for my application are stored in a Hashtable. I have been using <%@ page session="false" %> so that a new session is not automatically created. Each time before I create a new session, I check my Hashtable to see if that user has an existing session. If a session exists for that user, then I would like to continue using that session if possible. I have used the methods request.getSession() and request.getSession(false), but a new session is always returned.
I could create a new session, copy the attributes from the old session(which is stored in my Hashtable) to the new session, and then invalidate the old session, but I'd rather not do that.
Is there a way that I can use existing sessions that I have stored in my Hashtable?

First of all, is it possible to use the same session after a user exits and then returns?No, I don't think so. Let me explain why. When the server creates a session object for each client, it needs to know which client is making the request. Remember that HTTP is a stateless protocol. Every time a client makes a request, it sends some sort of session id to the server to let the server know who is trying to make the request. The server will then check to see if a session object exists for that particular client and if so, it will make sure that the max inactive interval (maximum time alloted for that client's session) has not been exceeded. If every thing is okay, then the client can access the session object to get values that were previously placed there. There are many ways that servers try to keep track of clients. One way is to have the clients write the session ID using cookies. But, many people like disallow cookies. So some servers do what is known as URL rewriting. That is, they will write the session ID on the end of the query string. This can also be accomplished programmatically, but it can be taxing to do. Anways, the point is that the client and the server have to have some sort of link between each other and that link is the session ID. So, if the browser is closed, the session ID is lost. That particular client will be forced to get a new session ID the next time the following code is executed:
//create a session object and set its values
HttpSession session = request.getSession(true);>
Second, if it is possible, then please tell me what I
need to do. So far, this is what I have been doing...
1.) The user submits login credentials
2.) The user is authenticated.
3.) The user does not have an existing session, so a
new one is created.
4.) The user closes the browser.
5.) The user returns to login page and submits login
credentials.
6.) The user is authenticated.
7.) The user has an existing session, so it should
be used.If you really want to do something like this, you could make up your own ID and store it as a cookie on the client. I've never tried anything like this before so you would have to do your own research. Are you sure you want to do something like this. There is a reason why it works the way it does. There is also a reason why you want to keep the session timeout value some what small. Let me give you an example of some craziness with sessions. A client we once had wanted to keep their sessions open for 4 hours because they said there clients simply did not like to log in all the time. I nearly gasped when I was told we needed to do this. When you make the session time out large (i.e. the maxInactiveInterval( )), then session objects stick around longer in the server. Let's say a client logs into the server and receives a session object. Then, the client makes a few requests. The server knows to keep the session alive as long as the time between requests has not exceeded 4 hours. Then the client closes the browser. How is the server suppose to know that the browser was closed. Well, it doesn't. It just knows to check times between requests. So, that particular session object won't be garbage collected until the session times out. What if a whole bunch of clients did this. Yucko. The server would have a whole bunch of session objects living in memory. What a waste. This is all above and beyond the typical security problems that can arise from having a session open for so long. To make a long story short, you really shouldn't do what you are trying to do unless it is the nature of the app.
>
This is where I'm having trouble. All active sessions
for my application are stored in a Hashtable. I have
been using <%@ page session="false" %> so that a new
session is not automatically created. Each time
before I create a new session, I check my Hashtable
to see if that user has an existing session. If a
session exists for that user, then I would like to
continue using that session if possible. I have used
the methods request.getSession() and
request.getSession(false), but a new session is
always returned.
I could create a new session, copy the attributes from
the old session(which is stored in my Hashtable) to
the new session, and then invalidate the old session,
but I'd rather not do that.
Is there a way that I can use existing sessions that I
have stored in my Hashtable?

Is a clustering licence required for using In-Memory replication?

          Hi,
          Could you please tell me if it is possible to use In-Memory replication
          without a clustering licence? (With WebLogic 6 on Linux with eval.
          licence.)
          I ask, because the documentation states that it can be used with
          HW load-balancing devices. If that is the case, how do you configure
          replication without configuring a WebLogic cluster (since you already
          have a HW LB cluster!).
          Regards,
          Frank Olsen


Okay, so then the advice to contact your BEA account rep for a clustering eval
          license is correct.
          Frank Olsen wrote:
          > Hi and thanks for your help,
          >
          > My license file does not have support for clustering/replication.
          > There's "only": WebLogic, SSL/Export, 2PC, EJB 1.1, EJB 2.0, JMS,
          > jDriver/MSSQLServer4, jDriver/Informix4, jDriver/Oracle.
          >
          > BTW, sorry for misreading the first response -- maybe I can put
          > it on the account on the end of the day with a flu ;-)
          >
          > Robert Patrick <[email protected]> wrote:
          > >My wife has the now expired eval license that came packaged
          > >with WLS 6.0 GA and
          > >it includes licenses for clustering and in-memory replication
          > >for both servlets
          > >and EJBs.
          > >
          > >Hope this helps,
          > >Robert
          > >
          > >Prasad Peddada wrote:
          > >
          > >> I don't think replication license is part of it.
          > >>
          > >> -- Prasad
          > >>
          > >> Robert Patrick wrote:
          > >>
          > >> > I was under the impression that our eval licenses
          > >now include clustering.
          > >> > Please open up your license.bea file to see if this
          > >is the case...
          > >> >
          > >> > Prasad Peddada wrote:
          > >> >
          > >> > ? Frank,
          > >> > ?
          > >> > ? I meant you need clustering license. Contact
          > >your sales rep to get an
          > >> > ? eval license.
          > >> > ?
          > >> > ? You can ignore groups for now.
          > >> > ?
          > >> > ? -- Prasad
          > >> > ?
          > >> > ? Frank Olsen wrote:
          > >> > ?
          > >> > ? ? Prasad Peddada [email protected]? wrote:
          > >> > ? ? ?No, you can't use In Memory replication without
          > >clustering
          > >> > ? ? ?license.
          > >> > ? ? ?
          > >> > ? ?
          > >> > ? ? Great!
          > >> > ? ?
          > >> > ? ? Could you please tell me how to configure this?
          > >> > ? ?
          > >> > ? ? The documentation took me from configuring replication
          > >of HTTP
          > >> > ? ? sessions in:
          > >> > ? ? http://e-docs.bea.com/wls/docs60/cluster/servlet.html
          > >> > ? ? to configuring Replication Groups in:
          > >> > ? ? http://e-docs.bea.com/wls/docs60/cluster/setup.html#683053
          > >> > ? ? which tells you to:
          > >> > ? ? 1. Open the Administration Console.
          > >> > ? ? 2. Select the Servers node.
          > >> > ? ? 3. Select the server to configure.
          > >> > ? ? 4. Select the Cluster tab.
          > >> > ? ? 5. Type in values for the following attribute
          > >fields:
          > >> > ? ? a) Replication Group: Enter the replication
          > >group name to
          > >> > ? ? which this server belongs.
          > >> > ? ? b) Preferred Secondary Group: Enter the name
          > >of the replication
          > >> > ? ? group you would like to use to host replicated
          > >HTTP session states
          > >> > ? ? for this server.
          > >> > ? ? 6. Apply the changes.
          > >> > ? ?
          > >> > ? ? It's all well to give names of these groups, but
          > >where do you specify
          > >> > ? ? the servers in the group?
          > >> > ? ?
          > >> > ? ? There's a link to:
          > >> > ? ? http://e-docs.bea.com/wls/docs60/adminguide/config.html#config_005
          > >> > ? ? which gives a list of configuration tasks (Configuring
          > >an individual
          > >> > ? ? server, Cloning an indv. server, ...).
          > >> > ? ?
          > >> > ? ? Where to I go from here?
          > >> > ? ?
          > >> > ? ? Regards,
          > >> > ? ? Frank Olsen
          > >

ITS, New user session denied due to memory bottleneck

Dear Friends,
We were facing the following issues
1) when we access Shopping cart functionality (using Internal ITS)
through web and it is given error
500 SAP Internal Server Error (Error: New session was refused due to
memory bottleneck). For every error in web, we could find the entry
in SM21 as mentioned below,
10:39:35 DIA 000 300 PHG02777 W0 D ITS, New user
session denied due to memory bottleneck.
10:39:35 DIA 000 300 PHG02777 W0 3 ITS, User
session could not be created
10:39:35 DIA 000 300 PHG02777 D0 1 Transaction
Canceled ITS_P 027 ( )
2) ST22 Dumps: SYSTEM_NO_SHM_MEMORY
We have followed SAP note #Note 1232973 - SYSTEM_NO_SHM_MEMORY short
dump and increased the parameter abap/shared_objects_size_MB from 320
to 640.After that we have not any SYSTEM_NO_SHM_MEMORY error and we
monitored in SHMM transaction about the free memory and
available. Before this change, we used to clean manually the ICF shared
memory when it got accumulated with high value. This implies that the
problem regarding SYSTEM_NO_SHM_MEMORY is solved.
But still we were facing issues mentioned in point no 1). When we have
more than 100 sessions. This is occurring more from 8:00 CET to 12:00 CET
(peak load)
Later we changed the value of abap/shared_objects_size_MB from 640 to
1664 and also we reduced em/initial_size_MB from 12Gb to 11Gb (since we
are not using full amount of em initial size).
****Important *****
after this change we are facing more often related to point no 1).
Moreover, We could find the following in the trace file of W0 as,
Trace file of W0 is showing,
Tue Apr 14 12:38:42 2009
WARNING => ipl_MemAvail already 80 % memory used (itsp:
highwater = 80 %) [itsplmem.c 534]
***LOG W0D=> NULL& [itspxx.cpp 697]
ERROR => itsp_OpenSession failed rc = 1, send icf error page
[itsplxx.c 878]
***LOG W03=> [itsplxx.c 879]
ERROR => ipl_OpenSession returns 1(ITSPE_FAILURE) [itsplxx.c
881]
ERROR => Raise Last error:[27 from: itspxx.cpp :699]
[itsplxx.c 1214]
ERROR => RaiseError(sapdext) ITS_P:27 [itspxx.cpp 699]
Since the parameter itsp/max_eg_mem_percent was set to 80.
It is not clear that from which memory layout the eror is occuring.
According to note :Note 785365 - Extended Global Memory (EG) increased
from 2 GB to 8 GB
we could see that The size of the EG is obtained from the total of the
parameter values em/global_area_MB and abap/shared_objects_size_MB
We would like you check the system and update us about the memory
layout defined in our system, more specically em/global_area_MB (we are
planning to increase the value according to note SAP Note 742048) from
896 (current) to 1200 (10 % of em/initial_size_MB) and reduce the
abap/shared_objects_size_MB from 1664 to 640 .
Any help will be highly appreciated.
Thanks
Venkat

to 1)
You can monitor the ITS memory usage in transaction SITSPMON, but you are on the right track with increasing em/global_area_MB.
If you see its sessions lasting for a long time, then you might need to adapt the session timeout to save memory (TIMEOUT of your icm server port). Have a look at note [885580 - Integrated ITS: Configuration Parameters|https://service.sap.com/sap/bc/bsp/spn/sapnotes/index2.htm?numm=885580] for details.
Regards, Michael

New session rejected due to memory bottleneck in testing proxy2proxy

Hi all!
i am testing abap proxy2proxy on the client 105 in help varient 5 of simple use cases docu.
even though i got sending successful from SXIPATT1 on the client 105, i got the following error.
my server has RAM 2GB and have four client(100,105,106,107), and only i use for testing, NOT another person, so i do not think memory bottleneck issue.
checked list.
1. SXMB_MONI, i could see inbound message(SENDER) <b>only.</b>
   from 105 to XI, no problem I thought, but very strange there, becasue there is only inbound message(sender).
2. status flag on sxmb_moni
   Message Recorded (Commit follows) (TO BE DELIVERED)
does anyone experience such issue?
do i have to up physical memory?
Error when processing your request
What has happened?
The URL http://sapxi.hh.net:8000/sap/bc/gui/sap/its/webgui/! was not called due to an error.
Note
The following error text was processed in the system XID : <b>New session rejected due to memory bottleneck</b>
The error occurred on the application server SAPXI_XID_00 and in the work process 0 .
The termination type was: ABORT_MESSAGE_STATE
The ABAP call stack was:
Function: ICF_ATTACH_ITS_PLUGIN of program SAPLHTTP_RUNTIME
Form: EXECUTE_REQUEST of program CL_HTTP_SERVER================CP
Function: HTTP_DISPATCH_REQUEST of program SAPLHTTP_RUNTIME
Module: %_HTTP_START of program SAPMHTTP
What can I do?
If the termination type was RABAX_STATE, then you can find more information on the cause of the termination in the system XID in transaction ST22.
If the termination type was ABORT_MESSAGE_STATE, then you can find more information on the cause of the termination on the application server SAPXI_XID_00 in transaction SM21.
If the termination type was ERROR_MESSAGE_STATE, then you can search for more information in the trace file for the work process 0 in transaction ST11 on the application server SAPXI_XID_00 . In some situations, you may also need to analyze the trace files of other work processes.
If you do not yet have a user ID, contact your system administrator.
Error code: ICF-IE-http-c:105-u:XISUPER -l:E-s:XID-i:SAPXI_XID_00 -w:0-d:20060410-t:195211-v: ABORT_MESSAGE_STATE-e:
HTTP 500 - Internal Server Error
Your SAP Internet Communication Framework Team

did you have this entries in SXMB_ADM of XI server
catgory:RUNTIME     parameter:ENGINE_TYPE     value:     HUB
catgory:RUNTIME     parameter:IS_URL     value:http://server:port/sap/xi/engine?type=entry

Disable non-SSL session tracking?

Hi, all,
I wonder if one can disable all session tracking in JSP's whenever SSL is not being used? I would like to turn off all cookie-setting and URL-rewriting and use SSL-session tracking only (if I use session-tracking at all on a given page). I also want to specify this behavior programmatically (inside my JSP's) and not in my server's config files.
I'm basically concerned that if my user leaves one of my HTTPS pages, they will still retain a non-secure cookie with their session information. This seems to be indeed the default behavior: when I run my tests and transition from an HTTPS page to an HTTP one, the browser does store a cookie. I know I can invalidate the session as the next step, but I'd rather have the cookie not being set altogether to begin with. Imagine the situation where the user leaves my HTTPS page for a totally different (HTTP) website: in this setting I won't get a chance to invalidate the session and delete the cookie.
Any ideas, therefore, on how to programmatically disable non-SSL session-tracking?
Thanks,
Dmitri.

I don't think you can do this programatically.
However I also don't think it is a problem.
Cookies are related to zone names aren't they?
http://mysite and https://mysite are two different
zones as far as cookies are concerned. One should
not be able to see the other.
It issues a new cookie for the http site you are just
navigating to. That cookie has nothing to do with
the secure site you just came from, and shouldn't be
able to tell them any info about the secure site.
I think you are worrying about something that isn't
really there.
What is your concern? That they pick up a JSESSIONID
from the cookie and can then pretend to be a
different user?Yes. A cookie is transmitted and stored unencrypted, I imagine (in any case, it should be more easily crackable than SSL). I wish Sun came up with an extension to the Session API where you would be able to explicitly specify which session-tracking protocols you want used and which ones you don't. At the moment their API abstracts and manages too much detail for you.
I mean, if my site is supposed to be secure while I'm using SSL, then you'd expect that no information about those secure sessions should leak outside the SSL protocol, wouldn't you say?

Oracle Forms Session Tracking mechanism

Hi,
In this doc http://www.oracle.com/technology/products/forms/pdf/10g/troubleshooting_fls.pdf we can read the following:
The JsessionID, which uniquely identifies a Forms session. The Forms Listener Servlet uses two session tracking mechanisms:
- Cookies, where the Servlet container sends a cookie to the client.
The client returns the cookie to the server upon each HTTP
request, thereby associating the session with the cookie.
- URL rewriting, where the Servlet container appends a session ID
to the URL path, for example:
http://host[:port]/forms90/l90servlet;jsessionid=a23445bcde89
Does this means that forms uses one of those, or uses both mechanisms simultaneous?
anyone?
Regards
Ricardo
Edited by: user12015527 on Mar 10, 2010 2:39 PM

duplicate post: Oracle forms session crashes.

HTTP Sessions tracking using in-memory session replication

Similar Messages

Maybe you are looking for