WL 7.0 sp4 session replication problem

          These are the configuration:
          * BEA WebLogic Server 7.0 SP4
          * Domain : mydomain
          * Machine:
          machine1 (Windows 2003)
          machine2 (Windows 2000)
          * Admin Server :
          myserver (on "machine1")
          * Managed Server :
          server1 (on "machine1") , replication group: groupA , prefered secondary replication
          group: groupB
          server2 (on "machine2") , replication group: groupB , prefered secondary replication
          group: groupA
          proxy (on "machine2")
          * Cluster :
          cluster-1 (contains "server1" and "server2")
          I've deployed an application (one EJB and some JSP, just for testing purpose)
          to "cluster-1".
          And I also deployed a "weblogic.servlet.proxy.HttpClusterServlet" application
          to "proxy",
          which is configured to connect "server1" and "server2".
          Now I open a browser and browse to "proxy" and I can see th result from "server1"
          or "server2".
          But when I shutdown the server which my current session is on, I was redirected
          to another server
          but the session is not replicated.

Did you set the replicated option for your webapp. Did you change the
          cookiename by anychance?
          sree
          "patrick" <[email protected]> wrote in message news:40e53734$1@mktnews1...
          >
          > These are the configuration:
          >
          > * BEA WebLogic Server 7.0 SP4
          > * Domain : mydomain
          > * Machine:
          > machine1 (Windows 2003)
          > machine2 (Windows 2000)
          > * Admin Server :
          > myserver (on "machine1")
          > * Managed Server :
          > server1 (on "machine1") , replication group: groupA , prefered
          secondary replication
          > group: groupB
          > server2 (on "machine2") , replication group: groupB , prefered
          secondary replication
          > group: groupA
          > proxy (on "machine2")
          > * Cluster :
          > cluster-1 (contains "server1" and "server2")
          >
          > I've deployed an application (one EJB and some JSP, just for testing
          purpose)
          > to "cluster-1".
          > And I also deployed a "weblogic.servlet.proxy.HttpClusterServlet"
          application
          > to "proxy",
          > which is configured to connect "server1" and "server2".
          >
          > Now I open a browser and browse to "proxy" and I can see th result from
          "server1"
          > or "server2".
          > But when I shutdown the server which my current session is on, I was
          redirected
          > to another server
          > but the session is not replicated.

Similar Messages

Apache + 2 Tomcats session replication problem.

Greetings everyone.
Before stating the problem, let me explain how my environment is set.
I have two machines. One (PC1) running Apache (HTTP server 2.0.58)
and one instance of Tomcat (5.0.28) and another machine (PC2) with
another instance of Tomcat(5.0.28).
The Apache server
It is configured to handle static content, to redirect dynamic content to a
Tomcat instance through AJP 1.3 connector.
This process is done through the mod_jk and the workers.properties
The workers.properties file is configured to have sticky_session = True
so it assigns a SESSION_ID to the same Tomcat it was first assigned.
The workers.properties file is configured to have
sticky_session_force = True so if the Tomcat the SESSION_ID was
assigned is not available, the server answers with a 500 error.
The Tomcat servers
Both have only the AJP 1.3 connector enabled
Both have the Cluster tag from the server.xml file uncommented
and the useDirtyFlag flag set to false, for not to allow SESSION
replication between Tomcats.
The workers.properties file
workers.apache_log=C:/Apache2/logs
workers.tomcat_home=C:/Tomcat5
workers.java_home=C:/j2sdk1.4.2_13
ps=/
#Defining workers -----------------------------
worker.list=balancer,jkstatus
#Defining balancer ---------------------------
worker.balancer.type=lb
worker.balancer.balance_workers=tel1, tel2
worker.balancer.sticky_session=True
worker.balancer.sticky_session_force=True
worker.balancer.method=B
worker.balancer.lock=O
#Defining status -----------------------------
worker.jkstatus.type=status
worker.jkstatus.css=/jk_status/StatusCSS.css
#Workers properties ---------------------------
worker.tel1.type=ajp13
worker.tel1.port=8009
worker.tel1.host=127.0.0.1
worker.tel1.lbfactor=1
worker.tel1.socket_keepalive=False
worker.tel1.socket_timeout=30
worker.tel1.retries=20
worker.tel1.connection_pool_timeout = 20
#worker.tel1.redirect=tel2
worker.tel1.disabled=False
worker.tel2.type=ajp13
worker.tel2.port=8009
worker.tel2.host=199.147.52.181
worker.tel2.lbfactor=1
worker.tel2.socket_keepalive=False
worker.tel2.socket_timeout=30
worker.tel2.retries=20
worker.tel2.connection_pool_timeout = 20
#worker.tel2.redirect=tel1
worker.tel2.disabled=False
THE PROBLEM
I open a browser in the jk-status page to see how the Tomcat instances are
working, and both are working fine: Stat -> OK, now as the
loadbalancing factor is 1 on both Tomcats, an even alternating session
distribution is set.
While this browser is open to keep an eye on the status, I open a new
browser (B1)to connect to my Web Application, Apache answers
correctly and gives me a SESSION_ID for Tomcat instance 1 [both
instances are OK], if I make a simple refresh, my SESSION_ID is still the
same so I'm assigned to Tomcat instance 1 but this time I get an
ERROR 503 - Service unavailable but looking at the status of the
Tomcat instances both instances are still OK, no-one is down. And it
stays throwing this error for as many refreshes i do.
Now, I open a new browser (B2)and do the same process as before,
as expected, Apache now gives me a SESSION_ID for Tomcat instance 2,
repeating the same refreshing process, the error is thrown again, but still at
the jk-status page, both instances are fine.
Without closing these windows, I make a new refresh try on B1 and
even though the jk-status says both Tomcat instances are OK, the error
is still thrown. I open a third one (B3), and Apache again, correctly
gives me a new SESSION_ID for Tomcat instance 1 and answers
correctly on the first call. But once again if i repeat the refreshing process, the
error is thrown again.
Note: Using a different resolution to always keep and eye on the
instances status and using a refresh rate of 1 second for status, both
servers always were OK.
So the main problem is that somehow when the session is replicated
to the same tomcat, Apache confuses and thinks it is not available, when
asking it through the jk-status it tells it is OK
I've been trying different configurations with both Apache and Tomcat,
but there must be something missing since I don't get it to work correctly
Thanks in advance for all your helping comments.
- @alphazygma

Whew... that was quite an answer... definitely is going to help him a lot. Yeah any n00b by now should know how to use google, but that's not the point in this forums, here we are to help each other. and wether you like it or not many of us deploy applications to tomcat and stumble on this. So dont try to be cool posting this kind of answers like google this or google that if you dont have an answer please dont comment you will appear to be more noobish than you aparently are.
Well enough talking.
I found the following useful: (it comes in the server.xml of the tomcat configuration)

Enabling that entry on both machines should be enough.
Aparently the problem is not with apache. is with tomcat since it can't retain the session apache gives.
more information in the Tomcat help at:
http://tomcat.apache.org/tomcat-5.0-doc/balancer-howto.html#Using%20Apache%202%20with%20mod_proxy%20and%20mod_rewrite

HTTP Session replication problem

Hi all,
After 3 days trying to make the http session replication work in my cluster is time to ask some help.
I´m trying to make a simple application (only a servlet) simulate a http session replication.
I´ve investigated all the possible problems that i encountered in google :
Here is my multicast tests :
SERVER 1
D:\>java -cp d:\bea\weblogic92\server\lib\weblogic.jar utils.MulticastTest -N SE
RVER1 -A 237.0.0.1 -p 12809
***** WARNING ***** WARNING ***** WARNING *****
Do NOT use the same multicast address as a running WLS cluster.
Starting test. Hit any key to abort
Using multicast address 237.0.0.1:12809
Will send messages under the name SERVER1 every 2 seconds
Will print warning every 600 seconds if no messages are received
I (SERVER1) sent message num 1
I (SERVER1) sent message num 2
Received message 2 from SERVER1
I (SERVER1) sent message num 3
Received message 3 from SERVER1
I (SERVER1) sent message num 4
Received message 4 from SERVER1
I (SERVER1) sent message num 5
Received message 5 from SERVER1
I (SERVER1) sent message num 6
Received message 6 from SERVER1
I (SERVER1) sent message num 7
Received message 7 from SERVER1
I (SERVER1) sent message num 8
Received message 8 from SERVER1
I (SERVER1) sent message num 9
Received message 9 from SERVER1
I (SERVER1) sent message num 10
Received message 10 from SERVER1
Received message 11 from SERVER1
I (SERVER1) sent message num 11
I (SERVER1) sent message num 12
Received message 12 from SERVER1
I (SERVER1) sent message num 13
Received message 13 from SERVER1
Received message 14 from SERVER1
I (SERVER1) sent message num 14
I (SERVER1) sent message num 15
Received message 15 from SERVER1
I (SERVER1) sent message num 16
Received message 16 from SERVER1
SERVER 2
D:\>java -cp d:\bea\weblogic92\server\lib\weblogic.jar utils.MulticastTest -N SE
RVER2 -A 237.0.0.1 -p 12809
***** WARNING ***** WARNING ***** WARNING *****
Do NOT use the same multicast address as a running WLS cluster.
Starting test. Hit any key to abort
Using multicast address 237.0.0.1:12809
Will send messages under the name SERVER2 every 2 seconds
Will print warning every 600 seconds if no messages are received
I (SERVER2) sent message num 1
I (SERVER2) sent message num 2
Received message 2 from SERVER2
I (SERVER2) sent message num 3
Received message 3 from SERVER2
I (SERVER2) sent message num 4
Received message 4 from SERVER2
I (SERVER2) sent message num 5
Received message 5 from SERVER2
I (SERVER2) sent message num 6
Received message 6 from SERVER2
I (SERVER2) sent message num 7
Received message 7 from SERVER2
I (SERVER2) sent message num 8
Received message 8 from SERVER2
I (SERVER2) sent message num 9
Received message 9 from SERVER2
I (SERVER2) sent message num 10
Received message 10 from SERVER2
I (SERVER2) sent message num 11
Received message 11 from SERVER2
I (SERVER2) sent message num 12
Received message 12 from SERVER2
Weblogic.XML descriptor
<session-descriptor>
          <session-param>
               <param-name>PersistentStoreType</param-name>
               <param-value>replicated</param-value>
          </session-param>
</session-descriptor>
CODE of the Servlet
HttpSession session = request.getSession();
          String sessionIdentif = (String)session.getAttribute("SESSIONIDENTIF");
          boolean boolNewSession = true;
          if(sessionIdentif == null) {
               session.setAttribute("SESSIONIDENTIF", "1234567890");
          } else
               boolNewSession = false;
As you can see the object that i put in session is a String, so Serializable....
All the requirements are validated (i think...) but it still not working ... i get the SESSION ID : WvHQMKzBh9T35BPhGJxyhQtGks5Q3qJmxnJwGn8mn7V24FM92cnh!949557342! NONE !1284141857480
And the NONE said that the session isn´t replicated.
Any idea what´s wrong with my try of getting http session replication working??
cheers,

Hi,
Thanks for your help.
I´ve tried your application with your proxy application but still no work...
I´m getting :
####<13/Set/2010 15H59m BST> <Debug> <Replication> <DST01VS51> <VIACTT2> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<anonymous>> <> <> <1284389948055> <000000> <Creating primary 1678977133045201960>
####<13/Set/2010 15H59m BST> <Debug> <ReplicationDetails> <DST01VS51> <VIACTT2> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<anonymous>> <> <> <1284389948055> <000000> <Unable to create secondary for 1678977133045201960 on null>
####<13/Set/2010 15H59m BST> <Debug> <ReplicationDetails> <DST01VS51> <VIACTT2> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<anonymous>> <> <> <1284389948181> <000000> <Unable to create secondary for 1678977133045201960 on null>
In my early post i forgot to mention that i´m using an Apache 2.2 in the front-end with Weblogic 9.2 (cluster).
What is happening in this weblogic configuration so that this session replication don´t work.
cheers,

In-memory http session replication problem

          Hi All ,
          I'm testing my cluster configuration with bea's example for session(http) replication,however
          when my cluster runs in one machine everything works fine ,if I run managed server
          from other host session replication does not occure.
          (I hit the firs ManagedServer with my browser,create session and than hit directly
          the second managed server)
          I've made a test which show that multicast I've setup among hosts is available,so
          this is not the case.
          (bind multicast sockets from two hosts and than send some data to it to see if
          both get the packag)
          Info logs from managed server's console shows that everythings work fine when
          startng the cluster - detects Peer Lost,and Peer Join.
          My cluster addres was setup as ip1:port1,ip2:port2.
          I will apricate any opinion, because we are stuck.
          Vladislav


          Hi All,
          10x Sunna , I figured the problem which actually was a metter of blindness :-)
          was that I did't make the request throught load-balancer and the coockies were
          from different hosts,in view of browser they differs.
          Sunna <[email protected]> wrote:
          >Session replication uses IP sockets not multicast. Is there a firewall
          >or something that could be blocking that? See:
          >
          >http://e-docs.bea.com/wls/docs81/cluster/features.html#1031598

Weblogic 7.0 sp1 cluster - session replication problem

Hi,
          I have installed Weblogic 7.0 sp2 on Win NT. To test clustering
          feature, I have installed one admin server and added two managed
          servers. All are running on same box. I could deploy web application
          to the cluster. Connection pools and every other resource is working
          well with the cluster. However I couldn't get session replication to
          work. I have modified web app descriptor, and set 'persistent store
          type' to "replicated".
          I accessed application from one managed server, in the middle of
          session I modified the port number in the URL to point to other
          managed server. It looks like second managed server has no idea of
          that session, my app fails because of this.
          Could you please help me out in this, Do I need to do any thing in
          addition to the above. I couldn't find much in the BEA manual..
          Thanks
          Rao


          For Web application like servlets/JSP, it is better to put one web server as proxy
          plugin before your two managed servers and access your application through web
          proxy. (You need set session as in-memory replicated either in weblogic.xml or
          by console editor). Otherwise, you need record the session cookie from the first
          serevr and send the cookie to the second server (not sure if it works). To access
          EJB/JMS, use cluster URL like t3://server1:port1,server2:port2.
          [email protected] (Rao) wrote:
          >Hi,
          >
          >I have installed Weblogic 7.0 sp2 on Win NT. To test clustering
          >feature, I have installed one admin server and added two managed
          >servers. All are running on same box. I could deploy web application
          >to the cluster. Connection pools and every other resource is working
          >well with the cluster. However I couldn't get session replication to
          >work. I have modified web app descriptor, and set 'persistent store
          >type' to "replicated".
          >
          >I accessed application from one managed server, in the middle of
          >session I modified the port number in the URL to point to other
          >managed server. It looks like second managed server has no idea of
          >that session, my app fails because of this.
          >
          >
          >Could you please help me out in this, Do I need to do any thing in
          >addition to the above. I couldn't find much in the BEA manual..
          >
          >
          >Thanks
          >Rao

Session in-memory replication problem

Hi,
          I am running into some cluster HttpSession replication problems. Here is
          the scenario where replication fails (all servers mentioned here are a
          part of a cluster).
          1a - 2 Weblogic servers (A&B) are running - no users logged in,
          2a - user logs in and a new session in server A is created.
          3a - after several interactions, server A is killed.
          4a - after user makes susequent request, Weblogic correctly fails over
          to server B
          Problem: Not entire session data is replicated. The authentication info
          seems to
          be replicated correctly but there are some collections in the session of
          server A
          that did not make it to the session in server B.
          The interesting part is this: If there is only one server A running to
          begin with and a user
          interacts with it for a while and only then server B is started, when
          after server B starts up
          server A dies - the entire session (which is exactly the same as in the
          failing scenario) is
          corretly replicated in B, including collections that were missing in the
          failing scenario.
          How can this be possible ????
          Thanks for any info on this one - it really puzzles me.
          Andrew


Yes, you are on the right track. Everytime you modify the object you should call
          putValue. We will make it more clear in the docs.
          - Prasad
          Andrzej Porebski wrote:
          > Everything is Serilizable. I get no exceptions. I did however read some old
          > posts regarding
          > session replication and I hope I found an answer. It basically seems to boil
          > down to what
          > triggers session sync-up between servers. In my case , I store an object into
          > session and
          > later on manipulate that object directly wihotu session involvment and the
          > results of those manipulations
          > are not replicated - no wonder if HttpSession's putValue method is the only
          > trigger.
          > Am i on the right track here?
          >
          > -Andrew
          >
          > Prasad Peddada wrote:
          >
          > > Do you have non serializable data by any chance?
          > >
          > > - Prasad
          > >
          > > Andrzej Porebski wrote:
          > >
          > > > Hi,
          > > > I am running into some cluster HttpSession replication problems. Here is
          > > > the scenario where replication fails (all servers mentioned here are a
          > > > part of a cluster).
          > > > 1a - 2 Weblogic servers (A&B) are running - no users logged in,
          > > > 2a - user logs in and a new session in server A is created.
          > > > 3a - after several interactions, server A is killed.
          > > > 4a - after user makes susequent request, Weblogic correctly fails over
          > > > to server B
          > > >
          > > > Problem: Not entire session data is replicated. The authentication info
          > > > seems to
          > > > be replicated correctly but there are some collections in the session of
          > > > server A
          > > > that did not make it to the session in server B.
          > > >
          > > > The interesting part is this: If there is only one server A running to
          > > > begin with and a user
          > > > interacts with it for a while and only then server B is started, when
          > > > after server B starts up
          > > > server A dies - the entire session (which is exactly the same as in the
          > > > failing scenario) is
          > > > corretly replicated in B, including collections that were missing in the
          > > > failing scenario.
          > > >
          > > > How can this be possible ????
          > > >
          > > > Thanks for any info on this one - it really puzzles me.
          > > >
          > > > Andrew
          > >
          > > --
          > > Cheers
          > >
          > > - Prasad
          >
          > --
          > -------------------------------------------------------------
          > Andrzej Porebski
          > Sailfish Systems, Ltd. Phone 1 + (212) 607-3061
          > 44 Wall Street, 17th floor Fax: 1 + (212) 607-3075
          > New York, NY 10005
          > -------------------------------------------------------------

Session Replication doesn't work when using a custom Unicast Channel

Hello!
After configure a WLS Cluster for an WebApp with session replication support enabled I faced some issues with cluster configuration.
My LAB env used for this configurations is:
One Solaris 10 SPARC box.
-- One WLS 11g (10.3.6) domain with:
---- 4 Managed servers:
---- Admin server
---- server-1
---- server-2
---- Proxy Server (HttpClusterServlet)
--- 1 Cluster composed by:
---- server-1
---- server-2In that setup I noticed if I define a custom network channel for servers ( server>protocols>channels ) in the cluster and set Cluster Messaging Mode as Unicast* in the Cluster config ( Cluster>Configuration>Messaging>Messaging Mode ), so the session state replication does not work.
When I enable the cluster replication debug for managed servers the following messages appears:
<> <> <1358966729933> <BEA-000000> <[roid:-1772481434088297851] Creating primary for application key /webapp>
####<Jan 23, 2013 4:45:29 PM BRST> <Debug> <ReplicationDetails> <de25503> <server-1> <[ACTIVE] ExecuteThread: '5' for queue: > 'weblogic.kernel.Default (self-tuning)'> <<ano
nymous>> <> <> <1358966729958> <BEA-000000> *<Has secondary servers? false>*
####<Jan 23, 2013 4:45:29 PM BRST> <Debug> <ReplicationDetails> <de25503> <server-1> <[ACTIVE] ExecuteThread: '5' for queue: >'weblogic.kernel.Default (self-tuning)'> <<ano
nymous>> <> <> <1358966729959> <BEA-000000> *<Current secondary server? null>*
####<Jan 23, 2013 4:45:29 PM BRST> <Debug> <Replication> <de25503> <server-1> <[ACTIVE] ExecuteThread: '5' for queue: >'weblogic.kernel.Default (self-tuning)'> <<anonymous>
<> <> <1358966729959> <BEA-000000> <[roid:-1772481434088297851] Unable to create secondary on null>
####<Jan 23, 2013 4:45:31 PM BRST> <Debug> <ReplicationDetails> <de25503> <server-1> <[ACTIVE] ExecuteThread: '5' for queue: >'weblogic.kernel.Default (self-tuning)'> After eliminate all possible issues with my webapp (serialization, weblogic descriptor configuration, etc) and try many cluster network configurations I noticed that this problem only occurs when I use Unicast for Cluster's Messaging.
At the end of the day I really would like to understand why the session replication only works for Cluster's Messaging using Multicast mode. I read a lot the WLS docs (specifically the cluster/network topics) [1][2], but I can't find an official explanation about this.
Can someone help me understand this behavior?
Many thanks.
[1] http://docs.oracle.com/cd/E15523_01/web.1111/e13701/network.htm
[2] http://docs.oracle.com/cd/E15523_01/web.1111/e13709/setup.htm

I have Fluxbox started with Slim and .xinitrc. Dbus works only with:
exec ck-launch-session startfluxbox
you need run Openbox with ck-launch-session:
exec ck-launch-session openbox-session
Bye!!

Session Replication and Session Timeout -- Urgent

WLS -4.5.1 SP 11
          OS - both NT and Sol
          HI All
          The session does not seems to timeout when running the Weblogic servers in
          clustered mode with in-memory session replication on.
          This problem seems to be there in version5.1 too , as per another post , it
          seems to be solved in WLS5.1 sp 8
          Could anyone pls confirm in which patch in WLS 4.5.1 is the problem
          addressed . I'm using patch 11 and am getting the problem.
          Thanks in advance..
          -Surya


Try SP13 and if the problem persists, contact [email protected]
          -K
          Surya B Maithani wrote:
          > WLS -4.5.1 SP 11
          > OS - both NT and Sol
          >
          > HI All
          >
          > The session does not seems to timeout when running the Weblogic servers in
          > clustered mode with in-memory session replication on.
          > This problem seems to be there in version5.1 too , as per another post , it
          > seems to be solved in WLS5.1 sp 8
          >
          > Could anyone pls confirm in which patch in WLS 4.5.1 is the problem
          > addressed . I'm using patch 11 and am getting the problem.
          >
          > Thanks in advance..
          >
          > -Surya

Failover is work, but Session replication doesn't work!

Dear all,
          My Cluster can failover, but can't session replicate....
          I start up the 2 WLS 6.0 and the Web Server, then I visit
          http://web_server_ip/myWebApp/SessionServlet serveral times, then the
          counter variable in the session raise up, then I down the "target" WLS, then
          I try http://web_server_ip/myWebApp/SessionServlet now, I can call but the
          counter restart.... so the session seems not yet replicated. what I can
          do...?!
          My Setup: I setup a Netscape Web Server 3.6.3 as the proxy using libproxy.so
          with the cluster setting, and use 2 WLS 6.0 (both are running in 7001,7002)
          as the Application Server Cluster (WLCluster). one WLS is on Unix Machine
          and the other in on a NT. I set the 2 WLSs: the Primary Group is
          "WL_rep_group1", and set the secondary is "WL_rep_group2".
          Deployment: I deploy the application "myWebApp" (.war) on the servers unix
          and nt, adn also deploy to the cluster WLCluster.
          I deploy the examplesWebApp to my created domain "demo", and modify the
          weblogic.xml:
          <weblogic-web-app>
          <session-descriptor>
          <session-param>
          <param-name>PersistentStoreType</param-name>
          <param-value>replicated</param-value>
          </session-param>
          </session-descriptor>
          <jsp-descriptor>
          <jsp-param>
          <param-name>
          pageCheckSeconds
          </param-name>
          <param-value>
          1
          </param-value>
          </jsp-param>
          <jsp-param>
          <param-name>
          verbose
          </param-name>
          <param-value>
          true
          </param-value>
          </jsp-param>
          </jsp-descriptor>
          </weblogic-web-app>
          so ... anything I did wrong, so the WLS can't do Session replication??!!
          Please Help or Comment!
          Thanks in advance!
          with thanks,
          Alex Wong


Dear Vinod and Kevin,
          Some more information that I think it may help.
          First, I want to correct 1 point in my previous message, the Replication
          Group is "WLRepGroup1" and the Preferred Secondary Group is "WLRepGroup2"
          for the 2 machine WLSs.
          Second, I found that the Multicast Communication is not functioning becoz
          the switch of my office block the 224.0.0.1's communication, so now I added
          a hub for this 2 machine, as a result the multicast is wokring... at least
          pass the java utils.MulticastTest....
          Thrid, after the mutlicast work, I try the SessionServlet, it is working at
          beginning... but after some more clicks, one server will cause the following
          error (this time crash on x.y.z.b--NT machine)....
          -------- (a long list, the IPs are replaced: x.y.z.a UNIX machine, x.y.z.b
          NT machine, x.y.z.c My PC)
          Screen caption on demo>Servers>wlserver2 (monitoring> cluster)
          Unexpected Error
          Connected to x.y.z.a:7001 Active Domain: demo Jun 11, 2001 5:06:23 PM
          GMT+08:00
          An unexpected error was encountered in processing your request.
          Exception
          java.lang.NullPointerException
          <>
          Current Date
          Mon Jun 11 17:06:23 GMT+08:00 2001
          Console Release Build
          6.0 Service Pack 2
          Console Build
          6.0 Service Pack 2 05/24/2001 11:55:28 #117037
          Server Release Build
          6.0 Service Pack 2
          Server Build
          6.0 Service Pack 2 05/24/2001 11:55:28 #117037
          All Server Product Versions
          WebLogic Server Build: 6.0 Service Pack 2 05/24/2001 11:55:28 #117037
          WebLogic XML Module: 6.0 Service Pack 2 05/24/2001 12:34:27 #117037
          Request Info
          Protocol: HTTP/1.1
          ServerName: x.y.z.a
          ServerPort: 7001
          Secure: false
          ContextPath: /console
          ServletPath: /panels/mbean/Server.jsp
          QueryString:
          mbean=demo%3AName%3Dwlserver2%2CType%3DServer
          PathInfo: null
          PathTranslated: null
          RequestURI: /console/panels/mbean/Server.jsp
          AuthType: null
          ContentType: null
          CharacterEncoding: null
          Locale: zh_TW
          Method: GET
          Session:
          weblogic.servlet.internal.session.MemorySessionData@661950
          RequestedSessionId:
          OyQxOz4LG0WnLjufYcZDkIGGXIVAhHNASdBLUs2seZnvoQ4RXekX/console
          RequestedSessionIdFromCookie: true
          RequestedSessionIdFromURL: false
          UserPrincipal: system
          RemoteUser: system
          RemoteAddr: x.y.z.c
          RemoteHost: x.y.z.c
          Parameters
          mbean = demo:Name=wlserver2,Type=Server
          Attributes
          console.original./console/panels/mbean/Server.jsp.ContextPath = /console
          console.original./console/panels/mbean/Server.jsp.Method = GET
          console.original./console/panels/mbean/Server.jsp.QueryString =
          mbean=demo%3AName%3Dwlserver2%2CType%3DServer
          console.original./console/panels/mbean/Server.jsp.RemoteUser = system
          console.original./console/panels/mbean/Server.jsp.RequestURI =
          /console/panels/mbean/Server.jsp
          console.original./console/panels/mbean/Server.jsp.ServletPath =
          /panels/mbean/Server.jsp
          console.preferences.ContextKey =
          /panels/mbean/Server.jsp
          javax.servlet.include.context_path = /console
          javax.servlet.include.request_uri =
          /console/common/requestinfo.jsp
          javax.servlet.include.servlet_path =
          /common/requestinfo.jsp
          javax.servlet.jsp.jspException =
          java.lang.NullPointerException
          weblogic.httpd.user = system
          weblogic.management.console.tags.ContentTag =
          java.lang.Object@545f89
          weblogic.management.console.tags.HeaderTag =
          java.lang.Object@4c5c6c
          weblogic.management.console.tags.TabbedDialogTag =
          java.lang.Object@7eafc
          Headers
          Accept = image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, */*
          Accept-Encoding = gzip, deflate
          Accept-Language = zh-tw
          Connection = Keep-Alive
          Cookie =
          JSESSIONID=OyQxOz4LG0WnLjufYcZDkIGGXIVAhHNASdBLUs2seZnvoQ4RXekX!143710744761
          7668081!-896735123!7001!7002!NONE
          Host = x.y.z.a:7001
          User-Agent = Mozilla/4.0 (compatible; MSIE 5.01; Windows 98)
          BrowserInfo
          User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows 98)
          IE: true
          Netscape: false
          Supported: true
          JavscriptHrefs: false
          TableCellClick: true
          DocumentReloadedOnResize: false
          DropdownStretchable: true
          CellSpacingBlank: false
          EmptyCellBlank: false
          ImgOnclickSupported: true
          TableBorderFancy: true
          PartialToWideTables: false
          Server System Properties
          bea.home = /u0/weblogic
          file.encoding = ISO8859-1
          file.encoding.pkg = sun.io
          file.separator = /
          java.awt.fonts =
          java.awt.graphicsenv = sun.awt.X11GraphicsEnvironment
          java.awt.printerjob = sun.awt.motif.PSPrinterJob
          java.class.path =
          /u0/weblogic/wlserver6.0:/u0/weblogic/wlserver6.0/lib/weblogic_sp.jar:/u0/we
          blogic/wlserver6.0/lib/weblogic.jar:/opt/oracle/product/8.1.6/jdbc/lib/class
          es12.zip
          java.class.version = 47.0
          java.ext.dirs = /u0/weblogic/jdk130/jre/lib/ext
          java.home = /u0/weblogic/jdk130/jre
          java.io.tmpdir = /var/tmp/
          java.library.path =
          /u0/weblogic/jdk130/jre/bin/../lib/sparc/hotspot:/u0/weblogic/jdk130/jre/bin
          /../lib/sparc::/lib:/usr/local/easysoft/oob/client:/usr/local/easysoft/lib:/
          u0/weblogic/wlserver6.0/lib/solaris:/u0/weblogic/wlserver6.0/lib/solaris/oci
          816_8:/usr/lib
          java.naming.factory.initial =
          weblogic.jndi.WLInitialContextFactory
          java.naming.factory.url.pkgs = weblogic.jndi.factories
          java.protocol.handler.pkgs =
          weblogic.utils|weblogic.utils|weblogic.net|weblogic.management|weblogic.net|
          weblogic.net|weblogic.utils
          java.runtime.name = Java(TM) 2 Runtime Environment,
          Standard Edition
          java.runtime.version = 1.3.0
          java.security.policy =
          =/u0/weblogic/wlserver6.0/lib/weblogic.policy
          java.specification.name = Java Platform API Specification
          java.specification.vendor = Sun Microsystems Inc.
          java.specification.version = 1.3
          java.vendor = Sun Microsystems Inc.
          java.vendor.url = http://java.sun.com/
          java.vendor.url.bug =
          http://java.sun.com/cgi-bin/bugreport.cgi
          java.version = 1.3.0
          java.vm.info = mixed mode
          java.vm.name = Java HotSpot(TM) Client VM
          java.vm.specification.name = Java Virtual Machine
          Specification
          java.vm.specification.vendor = Sun Microsystems Inc.
          java.vm.specification.version = 1.0
          java.vm.vendor = Sun Microsystems Inc.
          java.vm.version = 1.3.0
          javax.rmi.CORBA.PortableRemoteObjectClass =
          weblogic.iiop.PortableRemoteObjectDelegateImpl
          javax.rmi.CORBA.UtilClass = weblogic.iiop.UtilDelegateImpl
          javax.xml.parsers.DocumentBuilderFactory =
          weblogic.xml.jaxp.RegistryDocumentBuilderFactory
          javax.xml.parsers.SAXParserFactory =
          weblogic.xml.jaxp.RegistrySAXParserFactory
          jmx.implementation.name = JMX RI
          jmx.implementation.vendor = Sun Microsystems
          jmx.implementation.version = 1.0
          jmx.specification.name = Java Management Extensions
          jmx.specification.vendor = Sun Microsystems
          jmx.specification.version = 1.0 Final Release
          line.separator =
          os.arch = sparc
          os.name = SunOS
          os.version = 5.6
          path.separator = :
          sun.boot.class.path =
          /u0/weblogic/jdk130/jre/lib/rt.jar:/u0/weblogic/jdk130/jre/lib/i18n.jar:/u0/
          weblogic/jdk130/jre/lib/sunrsasign.jar:/u0/weblogic/jdk130/jre/classes
          sun.boot.library.path =
          /u0/weblogic/jdk130/jre/lib/sparc
          sun.cpu.endian = big
          sun.cpu.isalist = sparcv8plus+vis sparcv8plus
          sparcv8 sparcv8-fsmuld sparcv7 sparc
          sun.io.unicode.encoding = UnicodeBig
          user.dir = /u0/weblogic/wlserver6.0
          user.home = /
          user.language = en
          user.name = root
          user.region = US
          user.timezone = Asia/Hong_Kong
          weblogic.Domain = demo
          weblogic.Name = wlserver1
          weblogic.security.jaas.Configuration =
          weblogic.security.internal.ServerConfig
          weblogic.security.jaas.Policy =
          /u0/weblogic/wlserver6.0/lib/Server.policy
          ---------- End of the message -----------
          any comment?? is that I deploy on wlserver1, wlserver2, WLCluster cause
          error??!! anything I did wrong??!
          please help!! or Comment
          with thanks,
          Alex Wong
          "Alex Wong" <[email protected]> ¼¶¼g©ó¶l¥ó
          news:[email protected]...
          > Hi Vinod and Kevin,
          >
          > Thanks Vinod first, and after I try the session.getClass(), I get the
          > following output
          > class weblogic.servlet.internal.session.ReplicatedSessionData
          > I think that's means I am using "replicated" already, rite?!
          >
          > About the License, I cat the license.bea, there is a license to the
          > "In-memory Replication - Servlet" and "In-memory Replication - EJB",
          however
          > it is a eval license but seems not yet expired, so I don't think that is
          the
          > problem, rite?
          >
          > And a quick question, you tell me "your configuration looks ok" means the
          > H/W configuration, or the WLS configuration??!!
          > Maybe I tell you more about my WLS configuration, so that, you can
          > determinate where I go wrong. (thanks)
          >
          >
          > Vinod and Kevin
          > (Kevin, I am Alex Wong from ASL, if you have any comment, you can simply
          > reply my personal email, thx)
          >
          > Steps I did to create the domain, WLS and Cluster and deploy the .war
          > --- on Unix ---
          > I create a demo domain and wlserver1 using port 7001 and 7002 by the
          > installation program (ip x.y.z.a)
          > --- on NT ---
          > I create a demo domain and wlserver2 using port 7001 and 7002 by the
          > installation program (ip x.y.z.b)
          >
          > --- on Unix ---
          > start the wlserver1 with startWebLogic.sh
          >
          > --- My PC ---
          > go http://x.y.z.a:7001/console
          >
          > create one more server named "wlserver2" using port 7001 and 7002
          > create the cluster named "WLCluster" and fill the the cluster ip
          > "x.y.z.a:7001,x.y.z.b:7001"
          > fill the multi-cast ip 224.0.0.1
          > add the "wlserver1" and "wlserver2" to the "WLCluster"
          > stop the unix WLS
          > start the unix WLS with startWebLogic.sh
          >
          > --- on NT ---
          > start the WLS with startManagedWebLogic.sh wlserver2 http://x.y.z.a:7001
          >
          > --- on Unix ---
          > install the NES 3.6.3 and create a WEB server instance and modify the
          > obj.conf
          >
          > DETAIL
          > ----- start of the obj.conf -----
          > Init fn="load-modules" funcs="wl-proxy,wl-init"
          > shlib="/opt/nes/plugins/weblogic/libproxy.so"
          > Init fn="wl-init"
          > .
          > .
          > .
          > <Object name="weblogic" ppath="*">
          > Service fn=wl-proxy \
          > WebLogicCluster="x.y.z.a:7001,x.y.z.b:7001"
          > </Object>
          >
          > <Object name=default>
          > *NameTrans ...
          > *NameTrans ...
          > *NameTrans ...
          > *NameTrans fn=document-root root="/opt/nes/docs"
          > Service method="(GET|HEAD|POST|PUT)" type=text/jsp fn=wl-proxy \
          > WebLogicCluster="x.y.z.a:7001,x.y.z.b:7001"
          > .
          > .
          > .
          > </Object>
          >
          > <Object name=cgi>
          > .
          > .
          > .
          > </Object>
          > ----- End of the obj.conf ----- (I suppose will redirect all HTTP request
          to
          > WLS from NES)
          >
          > --- My PC ---
          > go http://x.y.z.a:7001/console
          > assign the DefaultWebApp_wlserver1 to wlserver1
          > assign the DefaultWebApp_wlserver2 to wlserver2
          > assign the console the wlserver1 only
          > assign the certificate to wlserver1, wlserver2, WLCluster
          > go SERVER-wlserver1 assign "WLRepGroup1" to Replication Group and
          Preferred
          > Secondary Group
          > go SERVER-wlserver2 assign "WLRepGroup1" to Replication Group and
          Preferred
          > Secondary Group
          >
          > --- on Unix ---
          > go to ./config/examples/applications/examplesWebApp/WEB-INF
          > modify the weblogic.xml
          > go to ./config/examples/applications/examplesWebApp
          > jar cvf myWebApp.war *
          > copy the myWebApp.war to ./config/demo/applications/.
          > and ftp myWebApp.war ro (NT)./config/demo/applicaitons/.
          >
          > --- My PC ---
          > And then now, I can visit http://x.y.z.a:7001/console
          > now I can configuration the myWebApp.
          > assign the myWebApp to wlserver1,wlserver2,WLCluster
          >
          > --- on Unix ---
          > stop the WLS and start with startWebLogic.sh
          >
          > --- on NT ---
          > stop the WLS and start with startManagedWebLogic.sh wlserevr2
          > http://x.y.z.a:7001
          >
          > --- My PC ---
          > visit http://x.y.z.a/myWebApp/SessionServlet (this time no port 7001,
          becoz
          > I want to use the NES web server to proxy the HTTP request)
          >
          > Finally... I can failover the SessionServlet, SnoopServlet and all the
          other
          > html, jsp, servlet, but seems can't load-balancing and Session
          Persistence.
          >
          > Is that anything I did wrong, please comment!
          >
          >
          > with thanks,
          >
          > Alex Wong
          >
          >
          >
          > "Vinod Mehra" <[email protected]> ¼¶¼g©ó¶l¥ó
          > news:[email protected]...
          > > Your configuration looks ok but I guess somehow your clustering is not
          > setup
          > > properly. Check if all the servers are listeniong on the same port and
          > part
          > > of
          > > the same cluster. Also check if you have clustering license.
          > >
          > > Make sure your sessions are of type : "replicated". For that print
          > > the value of session.getClass(), the name will give you an idea. Check
          > > your logs if you got any error message.
          > >
          > > --Vinod.
          > >
          > >
          > > "Alex Wong" <[email protected]> wrote in message
          > > news:[email protected]...
          > > > Dear all,
          > > >
          > > > My Cluster can failover, but can't session replicate....
          > > >
          > > > I start up the 2 WLS 6.0 and the Web Server, then I visit
          > > > http://web_server_ip/myWebApp/SessionServlet serveral times, then the
          > > > counter variable in the session raise up, then I down the "target"
          WLS,
          > > then
          > > > I try http://web_server_ip/myWebApp/SessionServlet now, I can call but
          > the
          > > > counter restart.... so the session seems not yet replicated. what I
          can
          > > > do...?!
          > > >
          > > > My Setup: I setup a Netscape Web Server 3.6.3 as the proxy using
          > > libproxy.so
          > > > with the cluster setting, and use 2 WLS 6.0 (both are running in
          > > 7001,7002)
          > > > as the Application Server Cluster (WLCluster). one WLS is on Unix
          > Machine
          > > > and the other in on a NT. I set the 2 WLSs: the Primary Group is
          > > > "WL_rep_group1", and set the secondary is "WL_rep_group2".
          > > >
          > > > Deployment: I deploy the application "myWebApp" (.war) on the servers
          > unix
          > > > and nt, adn also deploy to the cluster WLCluster.
          > > >
          > > > I deploy the examplesWebApp to my created domain "demo", and modify
          the
          > > > weblogic.xml:
          > > > -------------------
          > > > <weblogic-web-app>
          > > >
          > > > <session-descriptor>
          > > > <session-param>
          > > > <param-name>PersistentStoreType</param-name>
          > > > <param-value>replicated</param-value>
          > > > </session-param>
          > > > </session-descriptor>
          > > >
          > > > <jsp-descriptor>
          > > > <jsp-param>
          > > > <param-name>
          > > > pageCheckSeconds
          > > > </param-name>
          > > > <param-value>
          > > > 1
          > > > </param-value>
          > > > </jsp-param>
          > > > <jsp-param>
          > > > <param-name>
          > > > verbose
          > > > </param-name>
          > > > <param-value>
          > > > true
          > > > </param-value>
          > > > </jsp-param>
          > > > </jsp-descriptor>
          > > >
          > > > </weblogic-web-app>
          > > > ---------------------
          > > >
          > > > so ... anything I did wrong, so the WLS can't do Session
          replication??!!
          > > >
          > > >
          > > > Please Help or Comment!
          > > >
          > > > Thanks in advance!
          > > >
          > > > with thanks,
          > > >
          > > > Alex Wong
          > > >
          > > >
          > > >
          > > >
          > > >
          > > >
          > > >
          > > >
          > > >
          > >
          > >
          >
          >

Strange errors in session replication

          I get following errors in session replication :
          weblogic.utils.NestedError: Tried to update secondary, but it thought
          it was the primary
          - with nested exception:
          [weblogic.cluster.replication.BadStatusException: updateSecondary
          found
              -148099416498695151 but it is not the primary]
          Using Weblogic451, SP7 on Solaris 2.6.
          Any clues ??
          -Kuntal Shah.


Hello,
          > > Alteon is set to use "permanent hashing" to route the client into
          > > the same server in each request.
          >
          > How do you expect the switch to know about the secondary node
          > if the primary fails? This information is stored in the URL and
          > parsed by a proxy.
          It took a couple of days and fair amount of documentation to understand
          the behaviour of the web cluster...;-)
          We changed the configuration to include two proxy servers between the
          Alteon switch and web cluster, but we ended up with a different problem.
          The setup at the moment is:
          - Sun Solaris 2.6, native threads, JDK 1.1.7B_008
          - two WLS 4.51 SP7 running in cluster using in-memory replication
          - only servlets deployed in the servers
          - Two WLS proxies between the webcluster and alteon switch. Both proxies are
          configured to use both of the web servers. they do not belong to the web
          cluster.
          The proxies are configured in the following way:
          weblogic.httpd.session.enable=true
          weblogic.httpd.session.timeoutSecs=1800
          weblogic.httpd.session.invalidationIntervalSecs=600
          weblogic.httpd.session.URLRewriting.enable=true
          weblogic.httpd.session.cookies.enable=true
          weblogic.httpd.session.cookie.maxAgeSecs=-1
          weblogic.httpd.session.cookie.domain=some.domain.name
          weblogic.httpd.register.cluster=weblogic.servlet.internal.HttpClusterServlet
          weblogic.httpd.initArgs.cluster=defaultServers=x.y.z.a1:9001|x.y.z.a2:9001
          weblogic.httpd.defaultServlet=cluster
          weblogic.allow.execute.weblogic.servlet=everyone
          Probably most of the configuration parameters are not necessary, can
          somebody tell me which are
          and which are not?
          The servers in the web cluster are configured in the following way:
          The following parameters are in the start-up script of the servers:
          -Dweblogic.cluster.enable=true \
          -Dweblogic.system.name=somehost \
          -Dweblogic.cluster.name=someclustercluster \
          -Dweblogic.cluster.multicastAddress=237.0.0.1 \
          -Dweblogic.system.bindAddr=x.y.z.a1 \
          -Dweblogic.system.nativeIO.enable=true \
          -Dweblogic.debug.replication=1 \
          And the following are in the cluster-wide properties file:
          weblogic.httpd.clustering.enable=true
          weblogic.httpd.session.persistence=true
          weblogic.httpd.session.persistentStoreType=replicated
          weblogic.httpd.session.enable=true
          weblogic.httpd.session.timeoutSecs=1800
          weblogic.httpd.session.invalidationIntervalSecs=600
          weblogic.httpd.session.URLRewriting.enable=true
          weblogic.httpd.session.cookies.enable=true
          weblogic.httpd.session.cookie.domain=some.domain.name
          The IP-address of the Alteon box in front of the proxies is bound to
          'some.domain.name'.
          The first problem seems to be very poor performance. The overall throughtput
          is vveeeerrryyy slow, comparing
          to the configuration where there is no proxy sitting in front of the
          cluster. Furthermore, at some stage the proxy
          starts to print out error messages: java.lang.InvalidStateException: broken
          pipe [...] . After those exceptions no
          client get any response anymore.
          Could someone give some hints what are the most essential items in the
          weblogic.properties -file for the
          proxy server? Also, if someone has some experience from setting up a
          configuration similar to ours: one hardware
          load-balancer in front of two proxies, and eventually two web servers
          running in a web cluster.
          Cheers,
          Jarno.

Cluster session replication

Hi,
CFMX 7.01 MULTISERVER:
I am facing a problem, session are not replicating on two CF
instances on same server. Below is Jrun log file error detail
30/05 17:18:10 error Setup of session replication failed.
[1]java.rmi.RemoteException: The web application
'cfusion.ear#cfusion.war' could not be found to accept sessions for
replication.
at
jrun.servlet.session.SessionReplicationService.replicate(SessionReplicationService.java:8 0)
at sun.reflect.GeneratedMethodAccessor51.invoke(Unknown
Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown
Source)
at java.lang.reflect.Method.invoke(Unknown Source)
at sun.rmi.server.UnicastServerRef.dispatch(Unknown Source)
at sun.rmi.transport.Transport$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Native
Method)
at sun.rmi.transport.Transport.serviceCall(Unknown Source)
at sun.rmi.transport.tcp.TCPTransport.handleMessages(Unknown
Source)
at
sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(Unknown
Source)
at java.lang.Thread.run(Unknown Source)
CFMX 7.01 MULTISERVER:
My set-up, installed as multiserver. create new instance from
CF Admin. Cluster them from Jrun Admin. I followed almost all
instructions from live doc, and also help from other blogs.
So far i could not findout why session can not replicate on
same machine.
Is any body have idea about above error.
Thanks for your help in-advance

In my config.xml, I have:
<cluster>
<name>MyCluster</name>
<cluster-address/>
<default-load-algorithm>round-robin</default-load-algorithm>
<cluster-messaging-mode>unicast</cluster-messaging-mode>
<frontend-host>192.168.6.6</frontend-host>
<frontend-http-port>80</frontend-http-port>
<frontend-https-port>443</frontend-https-port>
</cluster>
I already posted my weblogic.xml

Clustering/Session Replication

Hi,
I know that's a fairly basic question, but I just want to make sure I am not missing anything here.
We have an application which runs on a cluster of WebLogics and the user sometimes is requested to sign in again (it seems like it hit another instance and the session is not there).
The admins from Weblogic say there is NO WAY to disable Session Replication when running Weblogic in a cluster, which makes total sense. They say, however, that if the application doesn't serialize its session, then Weblogic can't handle the serialization. In short, they say it's an application issue.
The developers say the application is written properly, and that the problem is in Weblogic's side.
Although I am still a developer (for my own projects), my role here doesn't allow me to look into the code, how can I show the developers that WebLogic is setup correctly and that they need to fix their code?
Regards

Information about session replication can be found here: http://docs.oracle.com/cd/E21764_01/web.1111/e13709/failover.htm#i1024620
it contains a section 'Requirements for HTTP Session State Replication' that might be helpful to you.
Also important to note is, how the load balancer handles session binding and the WebLogic cookie
http://docs.oracle.com/cd/E21764_01/web.1111/e13709/load_balancing.htm#i1026939

Http session replication fails in cluster

Hello everybody.
I have some problems with HTTP session replication in WebLlogic cluster environment. I have a cluster with 2 nodes and application deployed there. Application is configured with:
<session-param>
<param-name>PersistentStoreType</param-name>
<param-value>replicated</param-value>
</session-param>
in weblogic.xml
WebLogic plug-in for Apache webserver is configured properly as described in documentation.
But when I try to make experiment to enter the application, make some activities, look in console where I was redirected by apache proxy and manually shut down the node where request was sent, I loose my HTTP session with all data there (all the beans stored in session are Serializable). Replication doesn't work correctly. May be I've missed something in configuration? How can I configure my application to provide correctness session replication?
Thanks for advice.
Thanks

Thanks for response!
I'm using WebLogic 8.1 SP 4 and configured Apache proxy as described in documentation for load balancing. The only section I have in httpd.conf is next:
<Location /HTTPClnt>
     SetHandler weblogic-handler
</Location>
<IfModule mod_weblogic.c>
     WebLogicCluster serv1:7541,serv2:7541
     MatchExpression *.*
     Debug ON
     WLLogFile /www/tmp/global_proxy.log
     WLTempDir "/www/tmp"
     DebugConfigInfo On
     KeepAliveEnabled ON
     KeepAliveSecs 15
</IfModule>
<Location /myApp>
     SetHandler weblogic-handler
     WebLogicCluster serv1:7541,serv2:7541
</Location>and the string to include weblogic proxy module for apache:
LoadModule weblogic_module     modules/mod_wl_20.soI've configured CookiesEnabled=true in weblogic.xml, but it didn't help. About session specification in httpd.conf - where can I read about that? I've just configured apache according to manual from here:
[url http://e-docs.bea.com/wls/docs92/plugins/apache.html]http://e-docs.bea.com/wls/docs92/plugins/apache.html
Thanks

WS7 cluster session replication

Now that my WS7 cluster is up and running, I'm attempting to deploy a simple JSF webapp that incorporates session replication between the nodes in my cluster.
I've followed the instructions in the WS7 admin guide to enable session replication, and I've added the following to my sun-web.xml deployment descriptor:
<session-config>
    <session-manager persistence-type="replicated"/>
</session-config>The first problem I discovered was that my JSF managed bean was not serializable, so I fixed that. Now, however, when I load the initial page in my app, I am getting an NPE that looks like it is coming from within the WS7 session replication code:
[23/Dec/2009:08:46:51] warning ( 3692): for host 172.16.2.47 trying to GET /sessionTest/index.jsf while trying to GET /sessionTest/, service-j2ee reports: executePhase(RENDER_RESPONSE 6,com.sun.faces.context.FacesContextImpl@2de69e99) threw exception
javax.faces.FacesException
        at com.sun.faces.lifecycle.RenderResponsePhase.execute(RenderResponsePhase.java:135)
        at com.sun.faces.lifecycle.LifecycleImpl.phase(LifecycleImpl.java:251)
        at com.sun.faces.lifecycle.LifecycleImpl.render(LifecycleImpl.java:144)
        at javax.faces.webapp.FacesServlet.service(FacesServlet.java:245)
        at org.apache.catalina.core.ApplicationFilterChain.servletService(ApplicationFilterChain.java:398)
        at org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:277)
        at org.apache.catalina.core.StandardContextValve.invokeInternal(StandardContextValve.java:255)
        at org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:188)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:586)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.web.replication.session.SessionLockingStandardPipeline.invoke(SessionLockingStandardPipeline.java:71)
        at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:187)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:586)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.webserver.connector.nsapi.NSAPIProcessor.service(NSAPIProcessor.java:160)
Caused by: ClientAbortException: java.io.IOException: WEB8001: Write failed
        at org.apache.coyote.tomcat5.OutputBuffer.realWriteBytes(OutputBuffer.java:385)
        at org.apache.tomcat.util.buf.ByteChunk.flushBuffer(ByteChunk.java:404)
        at org.apache.coyote.tomcat5.OutputBuffer.doFlush(OutputBuffer.java:334)
        at org.apache.coyote.tomcat5.OutputBuffer.flush(OutputBuffer.java:313)
        at org.apache.coyote.tomcat5.CoyoteResponse.flushBuffer(CoyoteResponse.java:622)
        at com.sun.webserver.connector.nsapi.NSAPIResponse.flushBuffer(NSAPIResponse.java:127)
        at org.apache.coyote.tomcat5.CoyoteResponseFacade.flushBuffer(CoyoteResponseFacade.java:291)
        at com.sun.faces.application.ViewHandlerImpl.renderView(ViewHandlerImpl.java:203)
        at com.sun.faces.lifecycle.RenderResponsePhase.execute(RenderResponsePhase.java:106)
        ... 14 more
Caused by: java.io.IOException: WEB8001: Write failed
        at com.sun.webserver.connector.nsapi.NSAPIResponseStream.doWrite(NSAPIResponseStream.java:61)
        at org.apache.coyote.Response.doWrite(Response.java:575)
        at org.apache.coyote.tomcat5.OutputBuffer.realWriteBytes(OutputBuffer.java:380)
        ... 22 more
[23/Dec/2009:08:46:51] warning ( 3692): for host 172.16.2.47 trying to GET /sessionTest/index.jsf while trying to GET /sessionTest/, service-j2ee reports: REPL0080: Received exception while reading session from store [{0}]
java.lang.NullPointerException
        at com.sun.web.replication.session.LWSFPersistentManager.doRead(LWSFPersistentManager.java:162)
        at com.sun.web.replication.session.LWSFValve.invoke(LWSFValve.java:31)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:577)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.web.replication.session.SessionLockingStandardPipeline.invoke(SessionLockingStandardPipeline.java:71)
        at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:187)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:586)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.webserver.connector.nsapi.NSAPIProcessor.service(NSAPIProcessor.java:160)I thought that perhaps once the session was created, I would stop seeing this error. It isn't exactly the same, but I am still seeing an NPE even after the session has been created:
[23/Dec/2009:08:47:33] warning ( 3692): for host 172.16.2.47 trying to POST /sessionTest/index.jsf, service-j2ee reports: REPL0080: Received exception while reading session from store [{0}]
java.lang.NullPointerException
        at com.sun.web.replication.session.LWSFPersistentManager.doRead(LWSFPersistentManager.java:162)
        at com.sun.web.replication.session.LWSFValve.invoke(LWSFValve.java:31)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:577)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.web.replication.session.SessionLockingStandardPipeline.invoke(SessionLockingStandardPipeline.java:71)
        at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:187)
        at org.apache.catalina.core.StandardPipeline.doInvoke(StandardPipeline.java:586)
        at org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.java:556)
        at com.sun.webserver.connector.nsapi.NSAPIProcessor.service(NSAPIProcessor.java:160)Any idea what's going wrong? Is it something I've done?
I can give you the code for my webapp if you need it.
Thanks,
Bill

1. First try deploying your application without having this parameter, This is need only when you need session replication at web application level not at web server level..
2. Try with this option as well:
<sun-web-app>
          <session-config>
<session-manager persistence-type="replicated">
</session-manager>
</session-config>
</sun-web-app>

CF7 Clustering - session replication broken

Further to my
earlier
thread on CF7 clustering and the "network error",
I think session replication is broken. True, I can't find any
documentation which explains how it is supposed to behave, but this
is what I'm seeing:
If you have a two node cluster, and person A has a session on
node 1 and person B a session on node 2, session variables for each
user DO NOT appear on both nodes
Sessions on one node will only appear on the other when the
first node fails over
Moreover, they will only appear when the user makes another
request
CFLOGIN, when using loginstorage="session" does not replicate.
This means you have to use sticky sessions, and if there is a node
failure, the user will have to login in again. This can be avoided
by using loginstorage = "cookie" but this means the username and
password are stored unencrypted in the cookie.
Problems I have with this:
It's impossible to have a cluster wide view of sessions. E.g.
you can't count how many sessions you have at any one time. Any
code that makes use of being able to dig in to multiple sessions -
made possible by J2EE sessions - will not work reliably.
CFLOGIN is broken. It would be better to not use it for sites
that have to be secure as keeping the unencrypted username and
password in a cookie is very insecure. The only way to protect the
username and password being stored in the cookie is SSL; and that
means that every single item on the site needs to be protected by
SSL to stop the browser from serving up the cookie over an
unprotected connection.
It's not clear what is responsible for replication. If sessions
are only replicated when an instance goes down, rather than
continously, then I assume that the master instance of CF on each
server is responsible for replication. What happens if the server
crashes?

Information about session replication can be found here: http://docs.oracle.com/cd/E21764_01/web.1111/e13709/failover.htm#i1024620
it contains a section 'Requirements for HTTP Session State Replication' that might be helpful to you.
Also important to note is, how the load balancer handles session binding and the WebLogic cookie
http://docs.oracle.com/cd/E21764_01/web.1111/e13709/load_balancing.htm#i1026939

WL 7.0 sp4 session replication problem

Similar Messages

Maybe you are looking for