If you enjoyed this post, please considerleaving a comment or subscribing to the RSS feed to have future articles delivered to your feed reader.
Recent Posts
- WebLogic12c with t3s (SSL) secure protocol and the JMX client
- Implementing SOA using Opensource Apache Camel JMS Component on WebLogic 12c
- Why and How to use Apache CXF Service in WebLogic 12c ?
- How to use WebLogic 12c provided Maven Synchronization Plug-In ?
- Apache Camel based EJB Client accessing the EJBs deployed on WebLogic 12c
August 3rd, 2010 on 10:42 am
Hi jay,
Can we use MSI mode with cluster members?
i think so we dont use, the server instance in the cluster always need a updated config file.
then how we can restart failed managed server automatically in the cluster
thanks
venu
August 3rd, 2010 on 1:19 pm
Hi Venu,
In MSI mode the NodeManager is responsible for getting the correct “config.xml” file from the AdminServer. If the AdminServer is not running then the NodeManager will start the ManagedServers using “msi-config.xml” file. “msi-config.xml” is a configuration file which has the Last running configuration file of the Domain.
NodeManager has a feature called as “CrashRecoveryEnabled” Please refer to for more details: (http://download.oracle.com/docs/cd/E13222_01/wls/docs92/server_start/nodemgr.html#wp1082914) To ensure that Node Manager properly restarts servers after a system crash, you must perform the following:
* Ensure that CrashRecoveryEnabled is set to true.
The CrashRecoveryEnabled configuration property allows Node Manager to restart servers after a system crash. The property is not enabled by default.
* You should start the Administration Server via Node Manager.
* All managed servers should be started via the Administration Server. You can accomplish this via WLST or the Administration Console.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
August 3rd, 2010 on 2:07 pm
hi jay
thnx for reply
actually my doubt is:
ms_1 is in the cluster C1 with MSI mode enabled,
if Admin server failed, will the nodemanager restart tht MS_1 with msi-config.xml? or not
thnx in advnce
venu
August 4th, 2010 on 2:07 pm
Hi..
I am starting the Managed server from admin console using node manager and it generates “.out” and “.log” files in Server Log directory .
We can limit the size of “.log” file from admin console and it rotates after that specified size.
Can we do similar configuration for “.out” file ? What i have observed that this file rotate only on Server start . Thus what is happening once “.out” file reach 2Gb , the WL is not writing any logs to it and moreover opening the 2GB in memory is also useless.
Sumit
August 4th, 2010 on 3:08 pm
Hi Sumit,
We dont have much control on STDOUT from WLS side because it is basically controlled by JVM (Not by WebLogic)…But u can do one thing …if you think Large STDOUT are causing some issue….Better redirect the STDOUT to ServerLogs…using the following JAVA_OPTION -Dweblogic.log.RedirectStdoutToServerLogEnabled=true
Because we already have good control over the Server Log Rotation… Even we can force the Server Log to rotate any time we want like: http://middlewaremagic.com/weblogic/2010/04/13/force-serverlog-rotation/
Please read the comments also mentioned in the above link to get more informations of Rotating STDOUT…issues.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
August 4th, 2010 on 7:18 pm
Jay ,
1) I enabling this option from Admin console.
2) Logs are coming in both “.out” and “.log” file
will it possible for u to come online?
Sumit
August 4th, 2010 on 8:41 pm
Hi Sumit,
I m online right now.
.
.
Thanks
Jay SenSharma
August 5th, 2010 on 9:55 am
Jay,
I did one more thing. I keep the severity level to level of warning for std out and thereby logs ( notice and info) are not going to std out and they are redirecting to log file.
I can either turn off the logging to std out and redirect to log file
What is your Opinion on this? Is it a good way ?
Sumit
August 18th, 2010 on 11:28 am
4-node Cluster – not load-balance properly 🙁
Big IP; Solaris 10; iplanet 6.1sp9; Weblogic 10.3
I have 2 boxes (box A and box B)
both boxes have one iplanet webserver each which are my proxy plugin to a Weblogic cluster with 4 nodes; these 2 iplanets are being load-balanced by BigIP (works ok)
box A has 2 WL nodes (node1, node2)
box B has 2 WL nodes (node3,node4)
all 4 nodes is in 1 WL cluster
admin server is in box A (irrelevant i guess)
Weblogic default round-robin algorithim cluster loadbalancing
Problem is, in the entire duration of the JVM life, only 2 nodes will be hit; could be node3 or node1; or node1 or node2; but never all 4 of them; the volume of different users is a lot; this is production live; I saw from proxy logs that it’s only picking one preferred server in iplanet1 and the same thing with iplanet2 so all in all, there are only 2 nodes being hit; ANY INPUT? Thanks!
September 1st, 2010 on 5:22 am
Hi Jay,
Pls help me out the Node manager is starting …but when i sarts the Admin server the Node manger is forcely getting shutdown….
This is the Error it is throwing
java.net.BindException: Address already in use
at java.net.PlainSocketImpl.socketBind(Native Method)
at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:359)
at java.net.ServerSocket.bind(ServerSocket.java:319)
at javax.net.ssl.impl.SSLServerSocketImpl.bind(Unknown Source)
at java.net.ServerSocket.(ServerSocket.java:185)
at java.net.ServerSocket.(ServerSocket.java:141)
at javax.net.ssl.SSLServerSocket.(SSLServerSocket.java:84)
at javax.net.ssl.impl.SSLServerSocketImpl.(Unknown Source)
at javax.net.ssl.impl.SSLServerSocketFactoryImpl.createServerSocket(Unknown Source)
at weblogic.nodemanager.server.SSLListener.init(SSLListener.java:72)
at weblogic.nodemanager.server.NMServer.start(NMServer.java:200)
at weblogic.nodemanager.server.NMServer.main(NMServer.java:339)
at weblogic.NodeManager.main(NodeManager.java:31)
September 1st, 2010 on 11:54 am
Hi Sandeep,
Looks like Port is already in use. Please kill the Port and then restart your processes. Still if it doesnt work then It means there must be something wrong with the configuration.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
September 4th, 2010 on 11:09 pm
Hi Jay sir,
So can i change to other port….
Thanks & Rgds
Sandeep
September 4th, 2010 on 11:32 pm
Yes U can try.
September 6th, 2010 on 1:59 am
Hi sir,
i have tried after changing the port…. bt no use now it is showing javaIO Exception……..
September 6th, 2010 on 5:45 am
Hi Jay sir….
I have tried to change the port and it is not coming into rechable in logs it is showing like this :
Node manager v10.3
Configuration settings:
NodeManagerHome=/export/home/oracle/shri/wlserver_10.3/common/nodemanager
ListenAddress=
ListenPort=5560
ListenBacklog=50
SecureListener=false
AuthenticationEnabled=true
NativeVersionEnabled=true
CrashRecoveryEnabled=false
JavaHome=/export/home/oracle/shri/jdk160_05/jre
StartScriptEnabled=false
StopScriptEnabled=false
StartScriptName=startWebLogic.sh
StopScriptName=
LogFile=/export/home/oracle/shri/wlserver_10.3/common/nodemanager/nodemanager.log
LogLevel=INFO
LogLimit=0
LogCount=1
LogAppend=true
LogToStderr=true
LogFormatter=weblogic.nodemanager.server.LogFormatter
DomainsFile=/export/home/oracle/shri/wlserver_10.3/common/nodemanager/nodemanager.domains
DomainsFileEnabled=true
StateCheckInterval=500
Domain name mappings:
wl_server -> /export/home/oracle/shri/wlserver_10.3/samples/domains/wl_server
domain_1 -> /export/home/oracle/shri/user_projects/domains/domain_1
medrec -> /export/home/oracle/shri/wlserver_10.3/samples/domains/medrec
I have a doubt that i made all the servers down including Admin.
I m trying to start the Node mmanager and then i m trying to start the Admin….
I have done like this in the other setup their it works bt here it is not wrking pls help me out sir ……
Thanks and Rgds
Sandeep
September 6th, 2010 on 11:45 am
Hi Sandeep,
Is your NodeManager coming up properly ?
If Yes, then Have u tried to enroll the NodeManager with the Domain. Enrolling the NodeManager to the domain is the Most important step. Please try to follow the Step7) and Step8) mentioned in the below link: http://middlewaremagic.com/weblogic/2010/04/28/weblogic-clustering-in-remote-boxes/
It would be better if you can paste the Exception Trace which u are getting in the logs…IOException or anything else. While Posting the comments Please replace all > characters with > and Similarly replace all the < characters with < Otherwise your Errors may not be visible in the comments.
.
.
October 8th, 2010 on 4:01 pm
Hello Jay,
Yesterday in production environment i have comes across an strange issue. We are seeing that managed sever is running and it is starting one more instance of itself.
The second process is transient. It comes and goes and thus causing memory exhaustion.
Here is the output of ps -aef | grep java
noaccess 691 1 0 Aug 21 ? 121:12 /usr/java/bin/java -server -Xmx128m -XX:+UseParallelGC -XX:ParallelGCThreads=4
anp 12192 26161 0 Oct 02 ? 7790:16 /opt/fsm/bea/jdk150_12/bin/java -Dweblogic.Name=ANP-lenxksfsmanp01m-8011 -Dbea.
anp 4155 1 0 Sep 05 ? 461:53 /opt/fsm/bea/jdk150_12/jre/bin/java -XX:MaxPermSize=512m -Xmx512m -Djava.awt.he
anp 5341 19046 0 16:25:56 ? 0:01 /opt/fsm/bea/jdk150_12/bin/java -Dweblogic.Name=ANP-lenxksfsmanp01m-8001 -Dbea.
anp 19046 26161 0 06:30:25 ? 1075:00 /opt/fsm/bea/jdk150_12/bin/java -Dweblogic.Name=ANP-lenxksfsmanp01m-8001 -Dbea.
anp 26161 26146 0 Oct 01 ? 37:27 /opt/fsm/bea/jdk150_12/bin/java -client -Xms32m -Xmx200m -XX:MaxPermSize=128m –
PID 19046 was the managed server anp01 ms1
PID 5341 which has a PPID of 19046 appears to be the managed server trying restart itself.
Let me know if u have come across this isssue?
Thanks
Sumit
October 9th, 2010 on 7:40 pm
Jay,
The problem here is our “.out” file has reached to its max limit and thus nothing goes there. There is nothing in server log file. So we dont have logs . The node manager is not playing any role here that is confirmed from logs of node manager.
I just don’t know how to reproduce this.
It looks some people face this issue . Here is the ref
http://forums.oracle.com/forums/thread.jspa?threadID=1065702&tstart=0
Any way we are planning to raise SR on oracle for this and update here on this.
Thanks
Sumit
October 15th, 2010 on 9:33 pm
i have created a horizontal cluster in two physcal boxes.
When i am starting one of the managed sever in cluster on box1 , its giving the below error:
BEA-000170 <Server Server1 did not receive the multicast packets that were sent by itself
####Oct 12, 2010 2:17:52 AM Critical Health ventus.ibm.com Server1 ACTIVE ExecuteThread: '1' for queue: 'weblogic.kernel.Default (self-tuning)' WLS Kernel 1192771072384
BEA-310006 Critical Subsystem Cluster has failed. Setting server state to FAILED.
Reason: Unable to receive self generated multicast messages
Please help me how to solve the above problem
Thanks,
Karthy
November 13th, 2010 on 4:31 am
Hello Admin,
My env is weblogic 9.2.2.0, dev env, and in windows 2k3
I am getting an error while starting one of the managed servers from the same machine as the ADMIN SERVER
Here is the log file that shows the error
#### <>
#### <>
#### <>
#### <>
#### <>
#### <>
Can you please help. It is urgent
Thanks
Rakesh
November 13th, 2010 on 10:32 am
Hi Rakesh,
As the Error / Exception which u pasted are not clear/visible …we are not able to understand the actual issue. Please read the Point4). from the following link: http://middlewaremagic.com/weblogic/?page_id=146.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
November 13th, 2010 on 4:37 am
i dont know why the log file isnt showing because of the brackets
i will paste the error files here
BEA-000114 Failed to find cluster license: Error: Can not find component Cluster for WebLogic Platform 9.2 in the license file.
BEA-000362 Server failed. Reason: Error: Can not find component Cluster for WebLogic Platform 9.2 in the license file.
Can you please help.
If this seems that there is no license file, can you please provide where i can download the license.bea file
Thanks
Rakesh
November 13th, 2010 on 10:41 am
Hi Rakesh,
The above errors in Server Log indicate that eidht er you dont have a valid “license.bea” file or it got corrupted or wrongly edited by someone. Please contact Oracle WebLogic Support Team or your WebLogic Sales Representative person to get the correct license file.
Meaning of above Error Codes:
BEA-000114
Error: Failed to find cluster license: message
Description:Failed to find cluster license: message in the license file.
Cause:You need a valid license file to run clusters.
Action:Contact your sales representative for a valid license.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
November 22nd, 2010 on 10:50 am
Hey Jay,
facing one issue.
My webserver stop connecting for certain time, when I checked the Application server(WL 10.3) logs, I see several entries of
” <TCP/IP socket failure occurred while fetching statedump over HTTP".
I see this is related to addressing. Now, we do have a different cluster in the same domain with the same multicast address and diff port. Do you think that might be the cause? Lately I am facing many issues with multicast addressing, is there anything particular problem/advantage using Unicast? I have a cluster with two managed servers, on diff boxes, I think Unicast might be the better option. Thoughts?
Best Regards,
Roy.
PS Relieved on finding the site again!
November 22nd, 2010 on 10:52 am
left the part : <TCP/IP socket failure occurred while fetching statedump over HTTP"
November 22nd, 2010 on 11:14 am
Hi Roy,
Yes, you are right ..One of the reason of “TCP/IP socket failure occurred while fetching statedump over HTTP” exception is Addressing. It is a very Strict rule that Whenever you want to configure More than one Cluster on the same Network then Every Cluster must have a unique Multicast address. If the Multicast address is same for two or more clusters in that case the there are following Major Disadvantages:
1). There is a very high probability of multicast storm in the network..which usually leads to Bad N/W performance.
2). For every clusters Cluster Master the headache will increase. Every Cluster Master has to validate the incoming multicast message is coming from the correct Member of the Cluster or not. The Validation causes additional performance degrade.
3). You will start seeing Many Side effects of using same Multicast Address….like “TCP/IP socket failure occurred while fetching statedump over HTT” …etc.
Unicast Messaging creates very less burden on the network compared to the Multicast. In Case of Multicast Communication Evey cluster Member sends a small Multicast message to all the other members of the Cluster. But in case of Unicast the complete cluster is divided into multiple groups and Nodes only sends the Unicast messages to their Group Leader only. Group leader is chosen by automatically by the WLS APi…Usually the Older member of the a group becomes the Group leader. If the Gropu Leader goes down/crashes automatically the next older member Server of that group becomes the Gropu Leader.
Unicast Clustering promise little better performance compared to Multicast. Thats why from WLS10 onwards WebLogic Clusters “Messaging Type” is Unicast. Earlier versions of WebLogic used to suppoort only the Multicast.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
November 22nd, 2010 on 5:54 pm
Hai Jay,
WebLogic Server provides two versions of Node Manager.
1. Java based Node Manager
2. Script based Node Manager
The Java based Node Manager used in windows system.
The script base Node Manager is used in UNIX or LINUX system.
The script based Node Manager are using the SSH client.
The main advantages in the script based Node Manager is “Automatic Server Migration” is available in the Script based Node Managed not available in the Java based Node manager.
How to configure the Script based Node manager in LINUX system?
Regards,
S.Vinoth babu
December 6th, 2010 on 5:11 pm
Hai Jay,
I have configured the Script Based Node manager in Linux Environment.
Adminserver , MS1 –> unix1 machine( 192.168.9.11 )
MS2 –> unix2 machine (192.168.9.22 )
The script based node manager only have the “Automatic Server Migration”. But the managed server take the defaultsheap size are
MS1 -> Min Heapsize = 256mb –> in unix1 machine
Max Heapsize = 512mb
MS2 -> Min Heapsize = 256mb –> in unix2 machine
Max Heapsize = 512mb
If i use the java based node manager we have to increase the heap size in the weblogic console page itself
Domain –> Environment -> Servers -> Configuration ->General -> Server Start -> Arguments
Arguments: jvmArgs=”-Xms1024m -Xmx1024m”
But in the Script based nodemanager this page not referred by the nodemanged,then how to increase the heap size for every managed server in the Domain.
Regards,
S.Vinoth Babu.
December 6th, 2010 on 5:57 pm
Hi Vinobabu,
You can use the following option:
please try to add the following 2 entry in the “nodemanager.properties” file and in startWebLogic.cmd please add the JAVA_OPTION=-Xms1024m -Xmx1024m -XX:MaxPermSize=256m
StartScriptName=startWebLogic.sh
StartScriptEnabled=true
Also Please refer to James Bayer’s Blog…http://blogs.oracle.com/jamesbayer/2010/01/weblogic_nodemanager_quick_sta.html….This is one of the best Blog on NodeManagers…
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 6th, 2010 on 6:20 pm
Hai Jai,
The nodemanager.properties file is not available in the script based nodemanager. But in the Java Based nodemanager the nodemanager.properties is available in the nodemanager home path.
The script based nodemanger the only file is available is nodemanager.domains.
[root@localhost nodemanager]# pwd
/prosun/Middleware_1033/user_projects/domains/clustdomain/servers/MS1/data/nodemanager
[root@localhost nodemanager]# ll
total 24
-rw-r–r– 1 root root 142 Dec 6 16:41 boot.properties
-rw-r–r– 1 root root 5 Dec 6 16:41 MS1.lck
-rw-r–r– 1 root root 5 Dec 6 16:42 MS1.pid
-rw-r–r– 1 root root 12 Dec 6 16:42 MS1.state
-rw-r–r– 1 root root 25 Dec 6 16:42 MS1.url
-rw-r–r– 1 root root 353 Dec 6 16:41 startup.properties
i have one idea but if you change the setDomainEnv.sh shell script have the heap size 256mb & 512mb. If i changed into 1024mb means then all the managed server heap size max & mix = 1024mb. But my requirement is this first managed server have 1024mb and the second managed server have 2048mb.
Regards,
S.vinoth babu.
December 6th, 2010 on 6:26 pm
Hai Jai,
I have already post my comments in James Bayer’s Blog but He said i am not configure the script based node manager. I only configure the java based nodemanager.
FYI.
http://blogs.oracle.com/jamesbayer/2010/01/weblogic_nodemanager_quick_sta.html#comment-943355
Regards,
S.Vinoth babu
December 6th, 2010 on 6:46 pm
Hi Vinoth,
When we want to different configurations for ManagedServers then in that case we must have different versions of scripts like: “setDomainEnvMS1.sh” and “setDomainEnvMS2.sh”. Or “startManagedWebLogicOne.sh” and “startManagedWebLogicTwo.sh”
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 7th, 2010 on 12:09 pm
Hi Jay,
I have configured the heap size for every managed server as you use mentioned idea during the scirpt based nodemanager. What are the changes inside the bin directory
MS1 –> heapsize –> 1024
MS2 –> heapsize –> 2048
I checked the heap size in the weblogic console page
Domain –> Environment –> Servers –> MS1 –> Configuration –> Monitoring –> Performance
Domain –> Environment –> Servers –> MS2 –> Configuration –> Monitoring –> Performance
The MS1 max & min heap size is 1024.
The MS2 max & min heap size is 2048.
[root@sis-localhost bin]# pwd
/prosun/Middleware_1033/user_projects/domains/clustdomain/bin
[root@sis-localhost bin]# ls -lrt
total 132
drwxr-x— 2 root root 4096 Dec 3 14:43 service_migration
drwxr-x— 2 root root 4096 Dec 3 14:43 server_migration
drwxr-x— 2 root root 4096 Dec 3 14:43 nodemanager
-rwxr-x— 1 root root 2096 Dec 3 14:43 stopWebLogic.sh
-rwxr-x— 1 root root 2435 Dec 3 14:43 stopManagedWebLogic.sh
-rwxr-xr-x 1 root root 5656 Dec 3 14:43 startWebLogic.sh
-rwxr-xr-x 1 root root 3207 Dec 3 14:43 startManagedWebLogic.sh
-rwxr-x— 1 root root 15961 Dec 3 14:43 setDomainEnv.sh
-rwxr-xr-x 1 root root 5656 Dec 6 12:03 startWebLogic_backup.sh
-rwxr-xr-x 1 root root 3207 Dec 7 10:47 startManagedWebLogic_backup.sh
-rwxr-x— 1 root root 15961 Dec 7 10:47 setDomainEnv_backup.sh
-rwxr-x— 1 root root 15961 Dec 7 10:55 setDomainEnv_MS1.sh
-rwxr-xr-x 1 root root 3215 Dec 7 11:01 startManagedWebLogic_MS1.sh
-rwxr-xr-x 1 root root 3215 Dec 7 11:02 startManagedWebLogic_MS2.sh
-rwxr-xr-x 1 root root 5660 Dec 7 11:04 startWebLogic_MS1.sh
-rwxr-xr-x 1 root root 5660 Dec 7 11:05 startWebLogic_MS2.sh
-rwxr-x— 1 root root 15961 Dec 7 11:07 setDomainEnv_MS2.sh
Point 1: startWebLogic_MS1.sh
# Call setDomainEnv here.
DOMAIN_HOME=”/prosun/Middleware_1033/user_projects/domains/clustdomain”
. ${DOMAIN_HOME}/bin/setDomainEnv_MS1.sh $*
Point 2: startWebLogic_MS2.sh
# Call setDomainEnv here.
DOMAIN_HOME=”/prosun/Middleware_1033/user_projects/domains/clustdomain”
. ${DOMAIN_HOME}/bin/setDomainEnv_MS2.sh $*
Point 3: setDomainEnv_MS1.sh –> 1024
. ${WL_HOME}/common/bin/commEnv.sh
WLS_HOME=”${WL_HOME}/server”
export WLS_HOME
XMS_SUN_64BIT=”1024″
export XMS_SUN_64BIT
XMS_SUN_32BIT=”1024″
export XMS_SUN_32BIT
XMX_SUN_64BIT=”1024″
export XMX_SUN_64BIT
XMX_SUN_32BIT=”1024″
export XMX_SUN_32BIT
XMS_JROCKIT_64BIT=”1024″
export XMS_JROCKIT_64BIT
XMS_JROCKIT_32BIT=”1024″
export XMS_JROCKIT_32BIT
XMX_JROCKIT_64BIT=”1024″
export XMX_JROCKIT_64BIT
XMX_JROCKIT_32BIT=”1024″
export XMX_JROCKIT_32BIT
if [ “${JAVA_VENDOR}” = “Sun” ] ; then
WLS_MEM_ARGS_64BIT=”-Xms1024m -Xmx1024m”
export WLS_MEM_ARGS_64BIT
WLS_MEM_ARGS_32BIT=”-Xms1024m -Xmx1024m”
export WLS_MEM_ARGS_32BIT
else
WLS_MEM_ARGS_64BIT=”-Xms1024m -Xmx1024m”
export WLS_MEM_ARGS_64BIT
WLS_MEM_ARGS_32BIT=”-Xms1024m -Xmx1024m”
export WLS_MEM_ARGS_32BIT
fi
Point 4: setDomainEnv_MS2.sh –> 2048
. ${WL_HOME}/common/bin/commEnv.sh
WLS_HOME=”${WL_HOME}/server”
export WLS_HOME
XMS_SUN_64BIT=”2048″
export XMS_SUN_64BIT
XMS_SUN_32BIT=”2048″
export XMS_SUN_32BIT
XMX_SUN_64BIT=”2048″
export XMX_SUN_64BIT
XMX_SUN_32BIT=”2048″
export XMX_SUN_32BIT
XMS_JROCKIT_64BIT=”2048″
export XMS_JROCKIT_64BIT
XMS_JROCKIT_32BIT=”2048″
export XMS_JROCKIT_32BIT
XMX_JROCKIT_64BIT=”2048″
export XMX_JROCKIT_64BIT
XMX_JROCKIT_32BIT=”2048″
export XMX_JROCKIT_32BIT
if [ “${JAVA_VENDOR}” = “Sun” ] ; then
WLS_MEM_ARGS_64BIT=”-Xms2048m -Xmx2048m”
export WLS_MEM_ARGS_64BIT
WLS_MEM_ARGS_32BIT=”-Xms2048m -Xmx2048m”
export WLS_MEM_ARGS_32BIT
else
WLS_MEM_ARGS_64BIT=”-Xms2048m -Xmx2048m”
export WLS_MEM_ARGS_64BIT
WLS_MEM_ARGS_32BIT=”-Xms2048m -Xmx2048m”
export WLS_MEM_ARGS_32BIT
fi
Point 5: startManagedWebLogic_MS1.sh
# Export the admin_url whether the user specified it OR it was sent on the command-line
ADMIN_URL=”${ADMIN_URL}”
export ADMIN_URL
SERVER_NAME=”${SERVER_NAME}”
export SERVER_NAME
DOMAIN_HOME=”/prosun/Middleware_1033/user_projects/domains/clustdomain”
if [ “$1” = “” ] ; then
# Call Weblogic Server with our default params since the user did not specify any other ones
${DOMAIN_HOME}/bin/startWebLogic_MS1.sh nodebug noderby noiterativedev notestconsole noLogErrorsToConsole
else
# Call Weblogic Server with the params the user sent in INSTEAD of the defaults
${DOMAIN_HOME}/bin/startWebLogic_MS1.sh $1 $2 $3 $4 $5 $6 $7 $8 $9
fi
Point 6: startManagedWebLogic_MS2.sh
# Export the admin_url whether the user specified it OR it was sent on the command-line
ADMIN_URL=”${ADMIN_URL}”
export ADMIN_URL
SERVER_NAME=”${SERVER_NAME}”
export SERVER_NAME
DOMAIN_HOME=”/prosun/Middleware_1033/user_projects/domains/clustdomain”
if [ “$1” = “” ] ; then
# Call Weblogic Server with our default params since the user did not specify any other ones
${DOMAIN_HOME}/bin/startWebLogic_MS2.sh nodebug noderby noiterativedev notestconsole noLogErrorsToConsole
else
# Call Weblogic Server with the params the user sent in INSTEAD of the defaults
${DOMAIN_HOME}/bin/startWebLogic_MS2.sh $1 $2 $3 $4 $5 $6 $7 $8 $9
fi
Regards,
S.Vinoth Babu
December 8th, 2010 on 8:31 pm
Hai Jay,
I have configured the Script Based node manager through admin console i start the remote managed servers MS1 , MS2.
192.168.9.11 –> Adminserver , MS1
192.168.9.22 –> MS2
We create Two unix machine Unix1 , Unix1.
Unix1 –> assign servers Adminserver , MS1.
Unix2 –> assing server MS2.
The Two manager servers (MS1,MS2) is configured with Cluster (CLUST).
If i configure the Whole server migration option in the weblogic admin console. I am unable to start Two managed servers. I mentioned the errors are below.
MS1:- ( 192.168.9.11 )
MS2:- ( 192.168.9.22 )
How to resolve this issue during the Whole server migration by using the Script Based Node Manager in the cluster environment.
Regards,
S.Vinoth Babu.
December 9th, 2010 on 10:28 am
Hai Jay,
I have post my error logs my i checked double time all the data is available in the ” Leave Comment ” box but if i post the comment last 34 lines missed in the each managed server error logs.
The managed server missed lines for psMS1:- 192.168.9.11
The managed server missed lines for psMS1:- 192.168.9.22
Regards,
S.Vinoth Babu
December 9th, 2010 on 10:35 am
Hi Vinoth,
We are not able to see the Errors “It is not visible” so you should paste the errors as described in Point4). Diagram in the following Post: http://middlewaremagic.com/weblogic/?page_id=146
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 9th, 2010 on 12:19 pm
Hai Jay,
Kindly remove my last post.
psMS1 Error log :- ( 192.168.9.11 ) –> Unix1 machine
How to resolve the problem during the Script based node manager during the Whole Server Migration in Clustered Environment(CLUST).
In my knowledge the main reason is ” Stopping “async” replication service”
Regards,
S.Vinoth Babu
December 10th, 2010 on 4:52 pm
Hi Jay,
I resolved that issue.In my configuration nodemanager.properties file is not available because i am using the script base node manager. The nodemanager.properties is available in the Jave based node manager. The error main reason is i am not mentioned the ip address for cluster. The log error message shows the cluster ip address is Null. Then i configured the ip address to cluster then the problem is resolved.
Regards,
S.Vinoth Babu
December 10th, 2010 on 5:23 pm
Wow!!!
Vinoth it was a great finding …Now i can see in your Logs that ” Starting “async” replication service with remote cluster address “null”
Null Cluster Address.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 10th, 2010 on 6:30 pm
Hi Vinoth,
Thank you for sharing your Finding and the Resolution of the issue with us….
Magic Team Shares 15-Bonus Magic Points for you.
http://middlewaremagic.com/weblogic/?page_id=2588#comment-2192
And
http://forums.oracle.com/forums/thread.jspa?threadID=2142264&tstart=0
.
.
Keep Posting & Sharing 🙂
Thanks
Jay SenSharma
December 15th, 2010 on 11:00 pm
Hello Jay,
I have a big question.
Is there an effective way to achieve load balancing in weblogic 9.2.2.0 server or weblogic 10.3.0 server without any physical switches or network load balancers.
Can we use cluster technology for this ? if so, can you give me some idea or some process that explains this ???
Thanks so much jay
Rakesh
December 15th, 2010 on 11:13 pm
Hi Rakesh,
You can surely use weblogic cluster to achieve load balancing. To get more information you can have a look at the below link
WebLogic Clustering in Remote Boxes
Hope that would answer your question 🙂
Regards,
Ravish Mody
December 15th, 2010 on 11:24 pm
Hi Rakesh,
Load Balancing Without having a Cluster can be achieved with a Proxy Server like Apache Proxy.
If you have a WebLogic Cluster and if you dont want to use any Apache or additional Server then you can use WebLogic’s Inbuilt proxy Server… as described in the following Link: http://middlewaremagic.com/weblogic/?p=2649
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 15th, 2010 on 11:21 pm
will creating a cluster allow load balancing ??? isn’t there anything else that we need to do ???
I am sorry but can you please explain a little as how the post does load balancing ???
Thanks and Sorry
December 15th, 2010 on 11:33 pm
Hi Rakesh,
Yes, there are few cluster algorithms which can be used depending on the environment requirement, just have a look at the below comment which might help you understand the concept.
http://middlewaremagic.com/weblogic/?page_id=2612#comment-2058
Regards,
Ravish Mody
December 15th, 2010 on 11:35 pm
Hi Rakesh,
Load balancing and Failover are tow basic features of WebLogic Clustering. But without using the Clustering feature as well you can have load Balancing using WebServer like :
If your ABC1 and ABC2 Servers are NOT part of any Cluster …though if you want to Loadbalance the request between them…then in this case as well you need to use the WebLogicCluster parameter in the proxy configuration…like:
Proxy Plugin doesn’t have any additional parameter to specify Load Balancing (Round Robin) among Non-Clustered Managed Servers.
For more information on configuring Apache please refer to : http://middlewaremagic.com/weblogic/?p=1311
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 16th, 2010 on 2:16 am
hello Jay,
can we use this
http://middlewaremagic.com/weblogic/?p=2649
for production environment ???
Thanks
Rakesh
December 16th, 2010 on 3:50 pm
Hi Rakesh,
You can use the WebLogic HttpClusterServlet ProxyServer, however you have to make sure that the managed server which is hosting the ProxyWebApp application should not go down or else your your Proxy Server would go down.
Regards,
Ravish Mody
December 18th, 2010 on 12:34 am
I have configured a singleton service using the steps outlined on this website. The singleton service migrates successfully if the managed server is brought down via Node Manager.
Whenever the Node Manager is brought down abruptly, however, it cannot recover. If I restart it, the server’s state it is managing is UNKNOWN. If the Singleton tries to migrate in this state, the servers sometimes are marked as FAILED.
This is particularly disruptive during weekend server reboots. Here is the scenario:
Server X — Reboots on Saturday
–Hosts MachineA (MS1)
Server Y — Reboots on Sunday
–Hosts MachineB (MS2)
When X is rebooted, the singleton migrates to MS2. When Y is rebooted, MS1 enters the Failed state.
December 18th, 2010 on 9:58 am
Hi Waterboy,
For Singleton Service migration it is must that you use NodeManager and this migration happens by the NodeManager. So Killing the NodeManager is not recommended. Also Whenever you want to start your Managed Server which is part of Cluster and Holding the Singleton Service….Must be started using NodeManager Only.
If you dont start a Managed Server with the Help of Nodemanager then Nodemanager will always get the State of that Server as “UNKNOWN”…Nodemanager can get the correct health State of a ManagedServer only when the Managed Server is started with the help opf NodeManager.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 18th, 2010 on 2:01 am
Hello Ravish, thanks for the input. Also one last question ???
Is the HttpClusterServet method for load balancing more effective or using an IIS plugin for load balancing more effective for production environment ???
Thanks
Rakesh
December 18th, 2010 on 5:30 am
IIS because we are using weblogic in windows 2k3 environment
December 18th, 2010 on 9:54 am
Hi Rakesh,
It totally depends on your business requirement and Business budget to choose what kind of Web proxy should be used. As you mentioned that you are using Windows2K3 which provides IIS by default so it will be a best option to go with IIS.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 21st, 2010 on 8:50 pm
Hi Waterboy,
Please refer to the following Image…http://middlewaremagic.com/weblogic/wp-content/uploads/2009/08/How_to_Post_Comments.jpg
There is no space between “[” and “sourcecode”
Due to the space there we are not able to see any of your Exceptions.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 21st, 2010 on 9:02 pm
December 22nd, 2010 on 12:03 am
Hi Waterboy,
If before the Error code “BEA-310006” somewhere above in the server log if you will see then definitely you will find sone Multicast related Messages or Some Network related Error Messages / Warnings.
Usually above kind of issues happens due to n/w issues. As soon as you see such issue better try applying the multicast test or Check with the N/W Team…95% cases you will see it’s due to Multicase message Broadcasting failure or N/W Issue.
And regarding ” java.lang.UnsatisfiedLinkError: no stackdump in java.library.path”…you need to check whether the “-Djava.library.path” System variable is set correctly or not….For different operating system has different Environment variables like SHLIB_PATH, LD_LIBRARY_PATH etc…so you need to check whether those are pointing to the correct Native Library Modules or not?
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 22nd, 2010 on 12:26 am
The only error above 310006 is this. Note that cfwpoint51 is the OTHER server whose nodemanager was just killed.
####<Dec 21, 2010 3:54:04 AM EST> <Debug> <NodeManager> <cffialxprd492.lehman.com> <logserver_nonprod_1> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <> <1292921644781> <BEA-300047> <getStates weblogic.nodemanager.NMConnectException: Connection refused. Couldn't connect to NodeManager. Check that it is running at cfwpoint51:5556 at java.net.PlainSocketImpl.socketConnect(Native Method) at java.net.PlainSocketImpl.doConnect(PlainSocketImpl.java:333) at java.net.PlainSocketImpl.connectToAddress(PlainSocketImpl.java:195) at java.net.PlainSocketImpl.connect(PlainSocketImpl.java:182) at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:366) at java.net.Socket.connect(Socket.java:519) at weblogic.nodemanager.client.PlainClient.createSocket(PlainClient.java:32) at weblogic.nodemanager.client.NMServerClient.connect(NMServerClient.java:223) at weblogic.nodemanager.client.NMServerClient.checkConnected(NMServerClient.java:195) at weblogic.nodemanager.client.NMServerClient.getStates(NMServerClient.java:40) at weblogic.nodemanager.mbean.NodeManagerRuntime.getStates(NodeManagerRuntime.java:248) at weblogic.cluster.leasing.databaseless.MachineState.run(MachineState.java:111) at weblogic.work.ServerWorkManagerImpl$WorkAdapterImpl.run(ServerWorkManagerImpl.java:518) at weblogic.work.ExecuteThread.execute(ExecuteThread.java:209) at weblogic.work.ExecuteThread.run(ExecuteThread.java:181) >December 22nd, 2010 on 4:31 am
Also, why would it say DatabaseLessLeasing has failed – I am using using consensus leasing, not database leasing.
December 22nd, 2010 on 10:35 pm
HI Waterboy,
I suspect that your “config.xml” and “xxx_DataSource_jdbc.xml” files might be configured to use Database Leasing. Or can you please post the complete Stack Trace where it says DataBaseLeasing has Failed….
May be few lines above to that error we can find something related to it?
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 23rd, 2010 on 1:27 pm
Hello Jay/Ravish
I enabled the the Administration port and by default it runs on 9002.
However we have already a different weblogic admin application running on 9002 port . So i was redirected to this application admin console.
Now i don’t have any way to login to my first application. So i thought of kill the admin server and restart . But start of the admin server fails as 9002 port in already occupied
Thus to come out of this i had to stop the weblogic running on 9002.
Do we have any way to come out of this scenario? This is valid mistake for default configured ports/values in the scenario where we have 2 weblogic installed on the same box
Thanks
Sumit
December 23rd, 2010 on 7:57 pm
Hi Sumit,
Just open the “config.xml” file of the Server and then search for Tag <administration-port-enabled>false</administration-port-enabled> then add the following line of Tag just below to it:
<administration-port>9999</administration-port>
Put your desired Port number there and then start your AdminServer.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 28th, 2010 on 6:02 pm
Hi Jay,
Is there any limitation using clustered architecture for weblogic. In our architecture, we have clustered environment and weblogic 10.3.0 is our app server. When we create concurrent instances (say 500-600), instance creation stops after 200. Is this because of the cluster configuration??? Is there anyway to find the cause why the instance creation fails? anything to check from weblogic side?
December 28th, 2010 on 11:39 pm
Hi Charan,
The important thing here is “How actually you are trying to add 500-600 concurrent instances ? How exactly are you trying?”
From WebLogic side there is no restriction on the Number of Members for a Cluster…So as far as your Hardware/Network supports You can add as many Members to the Cluster.
But please remember that as soon as you increase the Number of nodes in a cluster …after a certain number (based on your N/W and Hardware combinations) the performance of overall Cluster starts decreasing…Because Every Cluster Member has to send some Heart Beat messages to other running members of the Cluster, The Cluster Master has to maintain the ActiveServerList (Dynamic Server List), Maintenance of Session Stickyness ….etc are various factors which creates an over burden on the Network because usually in Clusters the cluster members communicates with each other using some Multicast messages which creates a Multicast storm on the Network as soon as we keep on adding many cluster Members to the Cluster….
Even Chosing Unicast Clustering Communication Mode will not help much in case of a very large cluster. Unicast clustering uses TCP/IP sockets to pass cluster messages between members. To avoid requiring each cluster member
to have connectivity to every other cluster member, WebLogic Server uses a group leader strategy whereby the oldest member of the group (in other words, the server that was started first) is designated the group leader. All members of the cluster connect to the group leader so that the group leader acts as the relay point for cluster messages between members. If the group leader goes down, the next oldest member becomes the new group leader.
As you can imagine, the group leader strategy works well for small groups but becomes less efficient as the number of members of the group grows large. As such, WebLogic Server uses a multiple group leader strategy where it limits the number of members in a group to 10. If the cluster is larger than 10 members, WebLogic Server splits into two or more groups, each with their own group leader. The group leaders themselves are all interconnected to minimize the number
of hops that a cluster message must traverse to reach all cluster members.
The downside to unicast-based clustering is that as the size of the cluster gets very large, the number of sockets grows large and the burden on the group leaders becomes much higher. If the group leaders become overloaded, this can lead to cluster members dropping out and rejoining the cluster frequently. As such, it is recommend that you consider using multicast-based clustering for very large clusters. You should test whatever clustering mechanism you choose in your
environment with your applications under realistic loads to ensure that you system behaves as expected. This is especially important for large, high volume systems.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
December 29th, 2010 on 11:27 am
Jay,
Is group leader is different than cluster master?
It is good to know that there is a concept of group leader for 10 members. Can you please let me know some document where i can read more on these concept?
Thanks
Sumit
December 30th, 2010 on 3:10 pm
Hi Jay,
Now i have configure the cross domain user configuration in the oracle weblogic 10.3.3 server. But i am not able to configure.
I have mentioned the below oracle document to configure the cross domain configuration.
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/secmanage/domain.html#domain_interop
http://download.oracle.com/docs/cd/E14571_01/web.1111/e13752/toc.htm#INTRO120
http://download.oracle.com/docs/cd/E14571_01/apirefs.1111/e13952/taskhelp/security/EnableTrustBetweenDomains.html
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/ConsoleHelp/taskhelp/security/ConfigureConnectionFiltering.html
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/security/con_filtr.html#wp1030656
Regards,
S.Vinoth Babu
January 3rd, 2011 on 3:59 pm
Hi Jay,
Can you explain me the use of Data folder under AdminServer/managed server?
I tried starting admin server in production mode and it was giving me an exception that in production mode, password encrypted field cannot have clear text value (java.lang.illegalArgumentException).
When i saw the config file, for Admin Server, I was not able to find the tag. Can you let me know if this is fine? I tried adding this tag manually to the config file with the password encrypted as node manager password. Still i was getting the same error.
Thanks,
Charan
January 3rd, 2011 on 5:42 pm
Hi Charan,
Please refer toi the following Thread where i replied the same query… http://forums.oracle.com/forums/thread.jspa?threadID=2152413&tstart=0
If you are getting following kind of error:
<BEA-000362> <Server failed. Reason: [Management:141266]Parsing Failure in config.xml: java.lang.IllegalArgumentException: In production mode, it’s not allowed to set a clear text value to the property: PasswordEncrypted of ServerStartMBean
As the error suggests that the “config.xml” might have some Clear Text Passwords which is not allowed when you run it in production mode , Or it is also possible that due to some Wrong editing this files got wrong Encrypted Password…
So please try the below…
Step1). in your “setDomainEnv.sh” script set the “PRODUCTION_MODE=false” or use the following option in your server StartScript: ” -Dweblogic.ProductionModeEnabled=false”
Step2). Now enter the clear Text Passwords in the “config.xml”
Step3). Now Start your Server. To see everything is OK or not.
Step4). Now before changing your Server to PRODUCTION MODE back….
Open a command prompt and then Move inside bin directory
Now run the “setDomainEnv.sh” to set the DOMAIN Environment variables
please use the following command to Encrypt the ClearText Passwords ..
java weblogic.security.Encrypt
Enter Password: something
{3DES}7sdsdHJs23ddsJTw=
Step5). The above encrypted values you need to put in the “config.xml” file
Step6). Now Enable the PRODUCTION_MODE=true and then start your Server.
NOTE: Before making any changes to “config.xml” please take a Backup of it.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
January 4th, 2011 on 3:41 pm
Hi Jay,
Thanks for your prompt reply. I have a boot.properties file in the data-> node manager folder.
This is what I tried before your post.
I changed in config.xml to false. I restarted the server and everything was working fine.
I found in the config.xml that the Admin Server didn’t have tag and there are no tag(which is inside tag in other managed servers)
Am not sure from where it is complaining of clear text passwords. Can you let me know from where it would be looking??
After starting server in dev mode, I encrypted the clear text password in test server and added the tags and tag in admin server and restarted adminserver.Still it was throwing the same error.
Can you please help me out??
Thanks,
Charan
January 5th, 2011 on 8:41 pm
Hi Jay,
one more point. Whatever is the password i give in the password-encrypted tag it complains clear text value error. Is it referring to values from some other file rather than config.xml? am confused. Now my production mode is in dev mode. please help
January 5th, 2011 on 9:39 pm
Hi Charan,
Can u please paste the error stack Trace that will give us much idea what is actually happening at your end.
Also please check the following file which contains the encrypted UserName and Password …. “<DOMAIN_HOME>servers<YOUR_SERVER_NAME>securityboot.properties”
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
January 6th, 2011 on 1:49 pm
Hi Jay,
please find the below error Stack trace:
Starting WLS with line:
G:OracleMIDDLE~1JROCKI~1binjava -jrockit -Xms512m -Xmx1024m -Dweblogic.system.BootIdentityFile=G:OracleMiddlewaredomainsBpmPRODDomainserversAdminServerdatanodemanagerboot.properties -Dweblogic.nodemanager.ServiceEnabled=true -Dfuego.fdi.hybrid.ldap.debug=false -Dfuego.fdi.polling.debug=false -Dfuego.directory.hybrid.ldap.person.idAttribute=true -Dfuego.directory.polling.disable=true -Xnohup -Djavax.net.ssl.trustStore=G:OracleMiddlewaredomainsBpmPRODDomainserversC4MS1Psecuritycacertscacerts -Djavax.net.ssl.trustPassword=changeit -Djava.protocol.handler.pkgs=com.sun.net.ssl.internal.www.protocol -da -Dplatform.home=G:OracleMIDDLE~1WLSERV~1.3 -Dwls.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.management.discover=true -Dwlw.iterativeDev=false -Dwlw.testConsole=false -Dwlw.logErrorsToConsole=false -Dweblogic.ext.dirs=G:OracleMIDDLE~1patch_wlw1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_wls1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_cie660profilesdefaultsysext_manifest_classpath -Dweblogic.Name=AdminServer -Djava.security.policy=G:OracleMIDDLE~1WLSERV~1.3serverlibweblogic.policy weblogic.Server
I have found one more thing. If i set production mode to true in config.xml then am unable to start Admin server. If I set SET production mode = true in setDomainEnv.cmd, adminServer.out file shows WLS Start Mode = Production, but the config.xml still shows production mode is disabled (I checked in console too, the production mode checkbox was not selected).
Can you help me out??
January 6th, 2011 on 1:54 pm
Hi Jay,
full stack trace was not paste above :
Stack Trace :
Starting WLS with line:
G:OracleMIDDLE~1JROCKI~1binjava -jrockit -Xms512m -Xmx1024m -Dweblogic.system.BootIdentityFile=G:OracleMiddlewaredomainsBpmPRODDomainserversAdminServerdatanodemanagerboot.properties -Dweblogic.nodemanager.ServiceEnabled=true -Dfuego.fdi.hybrid.ldap.debug=false -Dfuego.fdi.polling.debug=false -Dfuego.directory.hybrid.ldap.person.idAttribute=true -Dfuego.directory.polling.disable=true -Xnohup -Djavax.net.ssl.trustStore=G:OracleMiddlewaredomainsBpmPRODDomainserversC4MS1Psecuritycacertscacerts -Djavax.net.ssl.trustPassword=changeit -Djava.protocol.handler.pkgs=com.sun.net.ssl.internal.www.protocol -da -Dplatform.home=G:OracleMIDDLE~1WLSERV~1.3 -Dwls.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.management.discover=true -Dwlw.iterativeDev=false -Dwlw.testConsole=false -Dwlw.logErrorsToConsole=false -Dweblogic.ext.dirs=G:OracleMIDDLE~1patch_wlw1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_wls1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_cie660profilesdefaultsysext_manifest_classpath -Dweblogic.Name=AdminServer -Djava.security.policy=G:OracleMIDDLE~1WLSERV~1.3serverlibweblogic.policy weblogic.Server
January 6th, 2011 on 1:55 pm
HI Jay,
Am unable to paste the full stack trace 🙁
January 6th, 2011 on 4:14 pm
Hi Charan,
Please refer to the Post Image …to see how to post the Exceptions…. Point4). http://middlewaremagic.com/weblogic/?page_id=146
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
January 6th, 2011 on 8:04 pm
Hi Jay,
Please find the Stack Trace as in AdminServer.out
Starting WLS with line:
G:OracleMIDDLE~1JROCKI~1binjava -jrockit -Xms512m -Xmx1024m -Dweblogic.system.BootIdentityFile=G:OracleMiddlewaredomainsBpmPRODDomainserversAdminServerdatanodemanagerboot.properties -Dweblogic.nodemanager.ServiceEnabled=true -Dfuego.fdi.hybrid.ldap.debug=false -Dfuego.fdi.polling.debug=false -Dfuego.directory.hybrid.ldap.person.idAttribute=true -Dfuego.directory.polling.disable=true -Xnohup -Djavax.net.ssl.trustStore=G:OracleMiddlewaredomainsBpmPRODDomainserversC4MS1Psecuritycacertscacerts -Djavax.net.ssl.trustPassword=changeit -Djava.protocol.handler.pkgs=com.sun.net.ssl.internal.www.protocol -da -Dplatform.home=G:OracleMIDDLE~1WLSERV~1.3 -Dwls.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.home=G:OracleMIDDLE~1WLSERV~1.3server -Dweblogic.management.discover=true -Dwlw.iterativeDev=false -Dwlw.testConsole=false -Dwlw.logErrorsToConsole=false -Dweblogic.ext.dirs=G:OracleMIDDLE~1patch_wlw1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_wls1030profilesdefaultsysext_manifest_classpath;G:OracleMIDDLE~1patch_cie660profilesdefaultsysext_manifest_classpath -Dweblogic.Name=AdminServer -Djava.security.policy=G:OracleMIDDLE~1WLSERV~1.3serverlibweblogic.policy weblogic.Server
<Jan 6, 2011 3:09:36 AM EST> <Info> <WebLogicServer> <BEA-000377> <Starting WebLogic Server with BEA JRockit(R) Version R27.6.0-50_o-100423-1.6.0_05-20080626-2105-windows-ia32 from BEA Systems, Inc.>
<Jan 6, 2011 3:09:37 AM EST> <Info> <Management> <BEA-141107> <Version: WebLogic Server 10.3 Fri Jul 25 16:30:05 EDT 2008 1137967 >
<Jan 6, 2011 3:09:40 AM EST> <Critical> <WebLogicServer> <BEA-000362> <Server failed. Reason: [Management:141266]Parsing Failure in config.xml: java.lang.IllegalArgumentException: In production mode, it’s not allowed to set a clear text value to the property: PasswordEncrypted of ServerStartMBean>
<Jan 6, 2011 3:09:40 AM EST> <Notice> <WebLogicServer> <BEA-000365> <Server state changed to FAILED>
<Jan 6, 2011 3:09:40 AM EST> <Error> <WebLogicServer> <BEA-000383> <A critical service failed. The server will shut itself down>
<Jan 6, 2011 3:09:40 AM EST> <Notice> <WebLogicServer> <BEA-000365> <Server state changed to FORCE_SHUTTING_DOWN>
<Jan 6, 2011 3:09:40 AM> <Debug> <NodeManager> <Waiting for the process to die: 5736>
<Jan 6, 2011 3:09:40 AM> <Info> <NodeManager> <Server failed during startup so will not be restarted>
<Jan 6, 2011 3:09:40 AM> <Debug> <NodeManager> <runMonitor returned, setting finished=true and notifying waiters>
Thanks,
Charan
January 6th, 2011 on 9:12 pm
HI Charan,
Please try the below steps to debug the issue:
1). Add the JAVA_OPTION “-Dweblogic.ProductionModeEnabled=false” in the Server Start Script Just beside the Heap Settings
2). Now restart your Server
3). If the Server Comes up successfully then do the Following (If not then send the Updated StackTrace)
4). Now open a command prompt to Encrypt the WebLogic password:
Run the “setDomainEnv.sh” in the command Shell.
Run the following command to encrypt the DataBase Password of the DataSource.
java weblogic.security.Encrypt YOURPASSWORD
{3DES}0HYiEySifDiz0IcqDhelGA==
(NOTE: the above command you need to run from inside the directory or u need to use -Dbea.home option to specify the Domain Directory path)
5). Now Place this Encrypted Password in your “config.xml” (Better please take a back up of it)
6). Now retry to start your Server….And Make Sure that the “<DOMAIN_HOME>serversAdminServersecurityboot.properties” file has the correct username and password…
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
January 8th, 2011 on 5:54 am
Hey Jay/Ravish
The environment is weblogic 9.2.2 production environment in windows 2k3.
I was trying to create a cluster with 2 managed servers in a domain.
after creating, when i start the managed server using startmanagedweblogic.cmd managed1 http://… the managed server starts well and it comes into RUNNING mode, after 5 or 10 seconds, there is a thread dump file and then i receive these messages
BEA-000170 Server managed1 did not receive the multicast packets that were sent by itself
BEA-310006 Critical Subsystem Cluster has failed. Setting server state to FAILED.
Reason: Unable to receive self generated multicast messages>
The same thing happens for server 2 too.
when i did the same thing in weblogic 10.3, i got the same error.
But i changed the messaging mode from multicast to unicast in weblogic 10.3 and then the cluster worked fine.
I searched for unicast in 9.2 but seems like unicast in available only from 10.0.
In weblogic 9.2, i tried with different multicast addresses and ports, still the same error, the managed servers works fine for 5 to 10 seconds and then they show FAILED state in the console.
I saw a few dforums in google, and some people say its a network issue or something.
can you please tell me how to solve this issue, its really been bugging me from long time.
Thanks in advance
Rakesh
January 8th, 2011 on 11:05 am
HI Rakesh,
The error codes clearly indicates that you network has some issues in few nodes. Please consult with your N/W team to find out why the Multicast communication is not happening properly between nodes.. From your side you can max perform the MulticastTest to prove that there is a Multicast issue, rest your N/W team need to fix.
Please refer to the following posts to know more about What i Multicast Failure…and It’s Causes:
http://middlewaremagic.com/weblogic/?p=980
And
http://middlewaremagic.com/weblogic/?p=4611
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
February 9th, 2011 on 1:22 pm
HI,
We have clustered environment.
We are connecting Oracle BPM and weblogic 10.3.0.
When we execute a batch file for creating 600 records, the transaction fails.
Please find the below error in weblogic managed server logs :
I have increased my JTA Time out from 1800 – 3600.
Is this correct to increase to such a high numbmer as against the default 30 seconds.
ERROR DETAILS:
<> <[STUCK] ExecuteThread: '13' for queue: 'weblogic.kernel.Default (self-tuning)' has been busy for "645" seconds working on the request "weblogic.work.SelfTuningWorkManagerImpl$WorkAdapterImpl@208acb6", which is more than the configured time (StuckThreadMaxTime) of "600" seconds. Stack trace:
Thread-72 "[STUCK] ExecuteThread: '13' for queue: 'weblogic.kernel.Default (self-tuning)'" {
jrockit.net.SocketNativeIO.readBytesPinned(SocketNativeIO.java:???)
jrockit.net.SocketNativeIO.socketRead(SocketNativeIO.java:31)
java.net.SocketInputStream.socketRead0(SocketInputStream.java:???)
java.net.SocketInputStream.read(SocketInputStream.java:107)
java.io.BufferedInputStream.read1(BufferedInputStream.java:249)
java.io.BufferedInputStream.read(BufferedInputStream.java:308)
^– Holding lock: java.io.BufferedInputStream@d85f5e[thin lock]
java.io.FilterInputStream.read(FilterInputStream.java:116)
java.io.BufferedInputStream.fill(BufferedInputStream.java:189)
java.io.BufferedInputStream.read1(BufferedInputStream.java:249)
java.io.BufferedInputStream.read(BufferedInputStream.java:308)
^– Holding lock: weblogic.xml.sax.OpenBufferedInputStream@d88b1a[thin lock]
com.sun.org.apache.xerces.internal.impl.XMLEntityManager$RewindableInputStream.read(XMLEntityManager.java:2921)
com.sun.org.apache.xerces.internal.impl.io.UTF8Reader.read(UTF8Reader.java:275)
com.sun.org.apache.xerces.internal.impl.XMLEntityScanner.load(XMLEntityScanner.java:1736)
com.sun.org.apache.xerces.internal.impl.XMLEntityScanner.skipSpaces(XMLEntityScanner.java:1490)
com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl$TrailingMiscDriver.next(XMLDocumentScannerImpl.java:1367)
com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl.next(XMLDocumentScannerImpl.java:647)
com.sun.org.apache.xerces.internal.impl.XMLNSDocumentScannerImpl.next(XMLNSDocumentScannerImpl.java:135)
com.sun.org.apache.xerces.internal.impl.XMLDocumentFragmentScannerImpl.scanDocument(XMLDocumentFragmentScannerImpl.java:447)
com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(XML11Configuration.java:765)
com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(XML11Configuration.java:729)
com.sun.org.apache.xerces.internal.parsers.XMLParser.parse(XMLParser.java:106)
com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser.parse(AbstractSAXParser.java:1198)
com.sun.org.apache.xerces.internal.jaxp.SAXParserImpl$JAXPSAXParser.parse(SAXParserImpl.java:516)
weblogic.xml.jaxp.WebLogicXMLReader.parse(WebLogicXMLReader.java:132)
weblogic.xml.jaxp.RegistryXMLReader.parse(RegistryXMLReader.java:144)
javax.xml.parsers.SAXParser.parse(SAXParser.java:384)
org.apache.axis.encoding.DeserializationContext.parse(DeserializationContext.java:223)
org.apache.axis.SOAPPart.getAsSOAPEnvelope(SOAPPart.java:660)
org.apache.axis.Message.getSOAPEnvelope(Message.java:424)
org.apache.axis.handlers.soap.MustUnderstandChecker.invoke(MustUnderstandChecker.java:54)
org.apache.axis.client.AxisClient.invoke(AxisClient.java:76)
org.apache.axis.client.Call.invokeEngine(Call.java:2765)
org.apache.axis.client.Call.invoke(Call.java:2618)
org.apache.axis.client.Call.invoke(Call.java:2385)
org.apache.axis.client.Call.invoke(Call.java:2331)
org.apache.axis.client.Call.invoke(Call.java:1756)
fuego.soaptype.SoapCall.invoke(SoapCall.java:217)
fuego.soaptype.SoapObject.invoke(SoapObject.java:294)
fuego.lang.Invokeable.invokeImpl(Invokeable.java:229)
fuego.lang.Invokeable.invokeDynamic(Invokeable.java:180)
MY.SBDTrackerAssessment.Default_1_0.Instance.CIL_prepareAssessments(Instance.xcdl:12)
MY.MyTest.Default_1_0.Instance.CIL_prepareAssessments(Instance.xcdl:???)
sun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethodAccessorImpl.java:???)
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:27)
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
java.lang.reflect.Method.invoke(Method.java:575)
fuego.server.execution.EngineExecutionContext.invokeMethodAsCil(EngineExecutionContext.java:1083)
fuego.server.execution.EngineExecutionContext.runCil(EngineExecutionContext.java:1295)
fuego.server.execution.microactivity.ComponentExecutionMicroActivity.runCil(ComponentExecutionMicroActivity.java:120)
fuego.server.execution.microactivity.ComponentExecutionMicroActivity.execute(ComponentExecutionMicroActivity.java:72)
fuego.server.execution.microactivity.MicroActivityEngineExecutionHandler.executeActivity(MicroActivityEngineExecutionHandler.java:51)
fuego.server.execution.ImmediateActivity.execute(ImmediateActivity.java:42)
fuego.server.execution.DefaultEngineExecution$AtomicExecutionTA.runTransaction(DefaultEngineExecution.java:287)
fuego.transaction.TransactionAction.startNestedTransaction(TransactionAction.java:518)
fuego.transaction.TransactionAction.startTransaction(TransactionAction.java:547)
fuego.transaction.TransactionAction.start(TransactionAction.java:212)
fuego.server.execution.DefaultEngineExecution.executeImmediate(DefaultEngineExecution.java:92)
fuego.server.execution.DefaultEngineExecution.executeAutomaticWork(DefaultEngineExecution.java:59)
fuego.server.execution.EngineExecution.executeAutomaticWork(EngineExecution.java:42)
fuego.server.execution.ToDoItem.executeAutomaticWork(ToDoItem.java:236)
fuego.ejbengine.ItemExecutionBean$1.execute(ItemExecutionBean.java:212)
fuego.server.execution.DefaultEngineExecution$AtomicExecutionTA.runTransaction(DefaultEngineExecution.java:287)
fuego.transaction.TransactionAction.startBaseTransaction(TransactionAction.java:462)
fuego.transaction.TransactionAction.startTransaction(TransactionAction.java:547)
fuego.transaction.TransactionAction.start(TransactionAction.java:212)
fuego.server.execution.DefaultEngineExecution.executeImmediate(DefaultEngineExecution.java:92)
fuego.server.execution.EngineExecution.executeImmediate(EngineExecution.java:66)
fuego.ejbengine.ItemExecutionBean.processMessage(ItemExecutionBean.java:202)
fuego.ejbengine.ItemExecutionBean.onMessage(ItemExecutionBean.java:78)
weblogic.ejb.container.internal.MDListener.execute(MDListener.java:403)
weblogic.ejb.container.internal.MDListener.transactionalOnMessage(MDListener.java:352)
weblogic.ejb.container.internal.MDListener.onMessage(MDListener.java:264)
weblogic.jms.client.JMSSession.onMessage(JMSSession.java:4526)
weblogic.jms.client.JMSSession.execute(JMSSession.java:4022)
weblogic.jms.client.JMSSession.executeMessage(JMSSession.java:3708)
weblogic.jms.client.JMSSession.access$000(JMSSession.java:114)
weblogic.jms.client.JMSSession$UseForRunnable.run(JMSSession.java:5058)
weblogic.work.SelfTuningWorkManagerImpl$WorkAdapterImpl.run(SelfTuningWorkManagerImpl.java:516)
weblogic.work.ExecuteThread.execute(ExecuteThread.java:198)
weblogic.work.ExecuteThread.run(ExecuteThread.java:165)
}
Please advice.
Thanks and Regards,
Charan
February 9th, 2011 on 2:29 pm
Hi Charan,
The this is a Server Hang/Slow responsiveness scenario and these scenarios requires at least 5-6 Thread dumps taken in the interval of 9-10 seconds each when you see the Server Hang. Without looking at the thread dumps it is difficult to say anyting on this. Plese followi the instructions mentione din the link to collect Thread dumps and sent it to us on contact@middlewaremagic.com
http://middlewaremagic.com/weblogic/?p=823
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
February 11th, 2011 on 1:46 pm
HI Jay,
I have sent the thread dump yesterday. Please advice.
Thanks and Regards,
Charan
February 11th, 2011 on 3:37 pm
Hi Charan,
Can you let us know where did you sent your thread dump?
February 14th, 2011 on 12:11 pm
Hi Ravish,
I sent the threadDump to contact@middlewaremagic.com on 2/10/2011.Please advice if I should send it again.
There is a thread stuck from the beginning of the task.
Initially my JTA time out was 1800 seconds. I increased JTA timeout to 5400 seconds and now users are able to make 570 transactions.
The thread stuck is still there. I dont know how to find where it is stuck. Also please advise me if in future the users want to make more transations i think I need to increase JTA time out again. I think this is not the best solution. Please advice me on optimum value of JTA and why am able to make 200 transactions only when my JTA is 1800 and why am able to make 570 transactions after i increased my JTA.
February 14th, 2011 on 12:17 pm
Hi Charan,
For some reason we dint get the file, so it would be better if you send it once again.
February 14th, 2011 on 2:51 pm
Hi Ravish,
I have resent the thread dumps.
Please have a look at them and help clarify my doubt.
Thanks and Regards,
Charan
February 14th, 2011 on 10:11 pm
Hi Charan,
Sorry to say but still we dint received any thread dumps. Make sure you are send it to “contact@middlewaremagic.com” and what is the size of the dumps.
February 17th, 2011 on 7:56 pm
Hi Ravish,
The size of the dumps is less than 1MB together.
But still I have split up the files and sent you.
Can you please give me any other alternative Id(gmail or yahoo)??
I sent to contact@middlewaremagic.com only.
I verified 🙁
Thanks and Regards,
Charan
February 17th, 2011 on 9:13 pm
Hi Charan,
Sorry for the late reply was very busy from last few days.
Yes, we had got your thread dumps and have gone through it. However in the thread dump it looks that everything is fine, now that be possible because you might have not taken the thread dump during the issue occurred or have taken the thread dump of a wrong server.
But I was going through your snip-code of the error in your previous comment it seems that the time taken for executing the batch file which is creating 600 records take more then 600 seconds thus the transition is been ended check out the below error message:
Now there are two things which can be done in your case
1) Try to enhance the batch code so that it completes the process within the 600 secs.
2) Or else if the batch cannot be enhance the code and it would take more then 600 secs then you would have to increase the StuckThreadMaxTime value as its an requirement of your application. To increase the “StuckThreadMaxTime” you can go to the following path of console and do it.
Servers –> YOUR_SERVER_NAME –> Configuration [tab] –> Tuning [sub-tab] –> Stuck Thread Max Time:
Note: You have to restart the same server to take the value which has been changed.
February 28th, 2011 on 6:26 pm
hi Middleware magic team,
As im new with weblogic,was trying clustering in weblogic.
I deployed an application on Managed server_1,1.1.1.1:7003 and then deployed the same application on Managed Server_2 1.1.1.1:7004.(Im doing this setup as a Backup ie if the application fails in 1 Managed Server den another managed server should be ready to accept the request)
Now i created a cluster of Managed server 1 and 2.
Kept the Cluster as unicast and Default load algorithm:Random/set the cluster ip address 1.1.1.1:7005
Now wen i try to access the application from cluster ip address 1.1.1.1:7005.failed in accessing.But when i try to access individual managed server deployment.Its workin fine.
Am i missin something?
Regards
Fabian
February 28th, 2011 on 7:17 pm
Hi Fabian,
To test a cluster configuration you do not have to use the IP and Port address of multicast / Unicast address as they are only used for internal communication for the managed servers.
If you want to test your configuration then you have to use a web-server like Apache which kept in the front of the WLS, you can go through the below posts
Topic: WebLogic HttpClusterServlet ProxyServer
http://middlewaremagic.com/weblogic/?p=2649
Topic: Apache Proxy as Static Content Repository
http://middlewaremagic.com/weblogic/?p=1311
Note: HttpClusterServlet is not recommend to be used in production environment
February 28th, 2011 on 8:05 pm
hi Ravish,
Thanks for those useful documents will try HttpClusterServlet ProxyServer.
And just want to ask
when i set SecureListener=false in nodemanager.properties
And Node manager type “plain”
I m getting my node status as inactive and even get this error
javax.net.ssl.SSLHandshakeException
Can you pls help me in this.
Regards
Fabian
March 1st, 2011 on 3:45 pm
Hi Fabian,
Whenever we create a Machine we must provide the IP Address for the NodeManager where we want to run it. The IP Address Must be correct and Valid. Please check that your NodeManager IP Address is correct or not?
.
.
Keep Posting 🙂
Thanks
Jay Sensharma
March 9th, 2011 on 2:20 pm
Hi Ravish,
Sorry for the delay in response.
Thank you for your thread dump analysis.
I tried increasing the StuckThread Timeout but still the execution stopped in the middle. When I increased JTA timeout to 5400 seconds then the execution was completed without any problem.
Is it right to increase JTA timeout like this as this is domain level change?
also the execution of the particular module takes 1 and a half hours for completing. If in future the completion takes more time again should i increase JTA time out??? is this the correct approach??
Please advice.
Thanks and Regards,
Charan
March 9th, 2011 on 3:48 pm
Hi Charan,
The transaction timeout can be set at the application level as well …based on the application type. Like there are different ways of setting timeout for a Transaction initiated by an EJB , WebServices, …etc
So you will need to check with your Developers to add the appropriate timeout for the application components in the application deployment descriptors.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
March 9th, 2011 on 6:53 pm
Hi Jay,
Thank you for your response.
how to set the transaction timeout at application level? Is it possible to set the transaction time out for the particular cluster alone where the application is deployed?
If so how?
Kindly advice.
Thanks and Regards,
Charan
March 9th, 2011 on 7:16 pm
Hi Charan,
Cluster is a “Logical Unit”..It is just a logical grouping of WebLogic Servers so it does not have any timeout parameter/attribute.
As i mentioned earlier that you can set Transaction Timeout for your application beased on the nature of your Application like example:
If you are using EJBs then you can set the timeout at application level like following:
http://download.oracle.com/docs/cd/E13222_01/wls/docs81/ejb/DDreference-ejb-jar.html#1116332
If you are using WebServices then you can set the timeout at application level like following:
transaction-timeout : The child element of the element specifies a timeout value for the current transaction, if the Web Service operation(s) are running as part of a transaction.
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/webserv_ref/dd.html
So please consult with your application developers to know what kind of timeout then need to set because it depends on the nature of the application.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
March 10th, 2011 on 3:10 pm
Hi Jay,
Thank you for your response.
I will check with application team for this. Where can i find this weblogic-ejb-jar.xml and weblogic-webserivces.xml?
Can you share your suggestion on JTA timeout value of 6000? Is it good,bad?? if there are negative impacts what would they be? how to check whether other applications or affected or not?
Is it like even if the transaction has completed within 1800 seconds since we have increased to 6000 the thread will still hold the connection something like that??
What is the best or optimum value for JTA?
Thanks and Regards,
Charan
March 10th, 2011 on 9:05 pm
Hi Charan,
“timeout” value should be set according to the requirement. Because if the timeout value is set to very low then most of the valid transactions also will fail due to transaction timeout.
If the timeout value is set to too high then invalid transactions also will wait until the timeout will not occur which will consume WebLogic Threads to be stuck in processing a Transaction…And these threads will be dedtcately working on those bad transactions until the Transaction will timeout …So like this it will consume server resources for a longer period unnecessarily.
So the exact value (approximate) of transaction timeout need to be set or else it will be harmful.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
March 14th, 2011 on 4:29 pm
Hi Jay,
Thank you very much for your reply.
I have one more doubt.
We have set the JTA time out value to 2000 in environment1 with same configuration but using WLS 9.2 and only one cluster. we have environment 2 using WLS 10.3.0 and 6 clusters..
The appliction(batch file) runs perfectly without any interuption or error in environment 1 where as it fails with XAConnection error in environment 2 (In both the cases the batch file rns for 1 and a half hours)
Can you help me understand this behaviour?
Is there anything I need to look into?
Thanks and Regards,
Charan
March 14th, 2011 on 10:00 pm
Hi Charan,
It looks like this issue need to be investigated in depth, So we would request you to please open a Support Ticket with Oracle. Because if changing is making much difference in Transaction Timeout and XAConnection Factory error then it is bad.
Oracle WebLogic Support person can check and compare the WLS9.2 and WLS10.3.0 codes to find out what is actually casing the issue or may provide you some debug patch to get more informations regarding the issue.
.
.
Keep Posting 🙂
Thanks
Jay SenSharma
April 19th, 2011 on 8:16 am
Hi Jay,
What is the difference between Dynamic and Static Cluster?
Thanks
Sathya
April 19th, 2011 on 10:53 am
Hi Sathya,
From the plugin side when “DynamicServerList” set to OFF, the plug-in ignores the dynamic cluster list used for load balancing requests from the plug-in and only uses the static list specified with the “WebLogicCluster” parameter, this is the called as Static Cluster.
When “DynamicServerList” set to ON (i.e. default) then the plug-in would take the dynamic cluster list for load balancing the requests, this is called as Dynamic Cluster.
To get more information you can refer this link
Topic: General Parameters for Web Server Plug-Ins
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/plugins/plugin_params.html#wp1143055
And if you your question was not from the plugin side, then can you let us know where did you get/read about these terminology, as till now I have never heard these type of cluster.
Regards,
Ravish Mody
April 20th, 2011 on 6:57 am
Hi Ravish,
I was going through documentation on Websphere application server, where the concept of dynamic Cluster and Static cluster is dicussed.. So was looking out if there is something of similar kind in Weblogic..
http://publib.boulder.ibm.com/infocenter/wxdinfo/v6r0/index.jsp?topic=/com.ibm.websphere.xd.doc/info/odoe_task/rodrcompareclusters.html
AND
http://publib.boulder.ibm.com/infocenter/wxdinfo/v6r0/index.jsp?topic=/com.ibm.websphere.xd.doc/info/odoe_task/codrdynamiccluster.html
Thanks
Sathya
April 20th, 2011 on 10:11 am
Hi,
Terminologies and Container Vendor specific Features offered by every application server are little different so of them can be compared and some not.
.
.
Thanks 🙂
Jay SenSharma
April 20th, 2011 on 7:02 am
Hi Jay/Ravish,
Say I have a clustered environment plugged with IPlanet.
Cluster is configured for Round robin algorithm.
When monitored using a HP Probe..
It was observed a within a Single domain the count on each Node was observed to be different continously…
In the way load was not getting equally distributed between nodes/MS within a cluster…
Suggest..
Thanks
Sathya
April 20th, 2011 on 7:18 am
Hi Jay/Ravish..
After my posting.. When through the other posting and saw this above
“chabuebeb ”
August 18th, 2010 on 11:28 am
4-node Cluster – not load-balance properly
Big IP; Solaris 10; iplanet 6.1sp9; Weblogic 10.3
I have 2 boxes (box A and box B)
both boxes have one iplanet webserver each which are my proxy plugin to a Weblogic cluster with 4 nodes; these 2 iplanets are being load-balanced by BigIP (works ok)
box A has 2 WL nodes (node1, node2)
box B has 2 WL nodes (node3,node4)
all 4 nodes is in 1 WL cluster
admin server is in box A (irrelevant i guess)
Weblogic default round-robin algorithim cluster loadbalancing
Problem is, in the entire duration of the JVM life, only 2 nodes will be hit; could be node3 or node1; or node1 or node2; but never all 4 of them; the volume of different users is a lot; this is production live; I saw from proxy logs that it’s only picking one preferred server in iplanet1 and the same thing with iplanet2 so all in all, there are only 2 nodes being hit; ANY INPUT? Thanks!
Scenario is the same….
Thanks
Sathya
May 2nd, 2011 on 2:06 pm
Hi Jay,
Can you please explain how to stop Nodemanager using WLST command. I am using startNodeManager() to start the nodemanager, but no method to stop it. Please help me. Thanks.
May 2nd, 2011 on 4:26 pm
Hi Rajendra,
Yes, you are right there is no method to stop Nodemanager using WLST, hence you cannot do it. You would have to explicitly find the PID of Nodemanager and kill it. Also if you have noticed that there is no shutdown script for Nodemanager given by Oracle Weblogic.
Regards,
Ravish Mody
June 29th, 2011 on 4:08 am
Don’t if this is the right forum but here goes
1 – WLS/SOA Suite 11gR1PS3 environment on OEL R5U4
2 – Geographically separate nodes, (approx 50km)
3 – We want active-active (ie both geographic regions supporting load(50-50) for the SAME application)
Apparent constraints. Both geographic regions sharing the SAME physical RAC instance hosted at one site (ie the RAC nodes DO NOT span geography)
1 – MDS (eg SOA_INFRA) data cannot be shared between clusters
2 – Geographic separation (latency/bandwidth) may “potentially” result in an unstable cluster (eg false warnings on WLS health)
Does anybody out there have a similar use case they can share
eg Workarounds/success/constraints etc..
thanks
July 19th, 2011 on 11:40 pm
Hi Magic Team,
Env: WL 10.3 & QA region
We are seeing huge Cluster logging information in .out files. So we want to disable the Cluster Logging option. We disbled cluster logging option from WL Admin conole –>server–> Debug. But still the cluster logs are floating into the .out file.
This is real urgency for us.Hence Please take a quick look and reply me by the earliest.
Thanks,
Praveen
September 1st, 2011 on 4:15 am
Hi,
I was wondering if you have any posts or threads related to using WL 9.2 in a cluster/node manager using a Hardware Load balancer. It seems our WL is having issues translating the host name. We use a dns with a https address but our WL uses http/non SSL protocals. I do not know where to look because we cannot login to our application and recieve access is denied Javascript errors when we force our way in via URL Login scripts.
Any suggestions?
Thx in Advance
Krelek
September 13th, 2011 on 6:39 pm
Hi Jay
I am using custom identity and custom trust in my application. ofcourse both are loading in node manager and admin server. However I was the below messages/errors in admin and nodemanager log files.
Sep 13, 2011 7:08:25 AM EDT> <Demo trusted CA certificate is being used in production mode:
Sep 13, 2011 7:38:17 AM weblogic.nodemanager.server.Handler run
WARNING: Uncaught exception in server handlerjavax.net.ssl.SSLKeyException: [Security:090482]BAD_CERTIFICATE alert was received from – . Check the peer to determine why it rejected the certificate chain (trusted CA configuration, hostname verification). SSL debug tracing may be required to determine the exact reason the certificate was rejected.
I have also passed ssl arguments for admin and node manager such as hostname verification and enfore. Still I am getting the Bad certificate error in node manager log file. Please help me on this. Thanks.
October 10th, 2011 on 7:58 am
Hi Jay,
We are upgrade the weblogic server 8.1 to 10.3.3 with Horizontal cluster domain.
Machine1 : MS1 –> App1:17001
Machine2 : MS2 –> App1:17001
MS1+MS2= Cluster_APP1 ( Multicast )
We have deploy the existing deployed 8.1 application into 10.3.3 weblogic server domain. then The applications are deployed successfully. But If i am accessing the application to enter the username & password then press enter key the same page will appear.
Then if i shutdown any one managed server in the clustered environment then we have to access the application the application working properly.
Kindly give your advise what is the root cause for those issue
Regards,
S.Vinoth Babu.
October 19th, 2011 on 9:18 am
Team ,
Which type of nodemanager is more feasible ?
Java Based Node manager or Shell Based Node Manager .
and why ?
Just a back ground for the question .
Issue – Managed Servers re-start on their own.
Cause – Failed to renew lease with the Cluster Leader.
Support team suggested to migrate from our existing shell based node manager to java based node manager .
Is this a feasible option ?
Thanks,
Kiran
October 19th, 2011 on 11:51 am
Hi Kiran,
First of all, magic team would like to congratulate you to become Magic Beginners with Magic Point 510 🙂
I believe the below link would help you to choose which based Node manger you should use
Topic: Determining Which Node Manager Version to Use
http://download.oracle.com/docs/cd/E12840_01/wls/docs103/nodemgr/overview.html#wp1074804
However I dint understand your issue, let me know if my understanding is correct regarding your issue.
“You do NOT want your managed servers to re-start on their own. ”
Irrespective of the issue, I believe your support team would be in a better position to explain you why are they suggesting it, hence would suggest you to ask them the reason for there given suggestion
Regards,
Ravish Mody
October 20th, 2011 on 6:57 am
Thanks Ravish for the explanation .
November 3rd, 2011 on 5:03 am
Hi Team ,
Can you let me know how do I migrate from script based node manager to java based node manager ? If you have any document or detailed steps will be really helpful .
thanks,
Kiran
February 9th, 2012 on 8:14 pm
Hi Team,
We have a cluster of 16 managed server’s. Currently for load balancing we have configured weblogic HTTPCluster load balancer, but some Apache and other can also be used.
Could you please suggest which load balancing method is considered best performance wise.
Weblogic version:10.3.5
System: Exadata
OS: OEL5.7
Thanks and Regards!
March 6th, 2012 on 4:53 pm
Hi Ravish,
We are facing ‘Failure of Server Apache Bridge Error’. Following are the details:
System: Windows 2008 64 bit
Forms/reports [64 Bit] Version 11.1.2.0.0
weblogic server :10.3.5
Apache Version:
D:SoftwareOracleMiddlewareOracle_FRHome1ohsbin>httpd -version
Server version: Oracle-HTTP-Server/2.2.15 (Win32)
Server built: Apr 2 2011 17:14:20
Server label: APACHE_11.1.1.5.0_WINDOWS.X64_110325.2000
We have tried to give static IP address in moduleconf folder forms.conf and reports_ohs.conf like this:
SetHandler weblogic-handler
WebLogicHost 10.161.168.119
WebLogicPort 9001
#mod_weblogic related entry
#
SetHandler weblogic-handler
WebLogicHost 10.161.168.119
WebLogicPort 9002
#
And from weblogic console also we have changed the listen address to this static IP and restarted the ohs,form, report server.
But still facing the same issue.
Please suggest.
Thanks in advance!!
March 21st, 2012 on 5:11 pm
Hi,
I have a requirement to delete a managed server, say ‘server1’ from a domain. The managed server is part of a cluster, ‘cluster1’. So before deleting the managed server, I need to unassign the server from the cluster.
For this, I am using the below code:
readDomain()
unassign(“Server”, “server1”, “Cluster”, “cluster1”)
updateDomain()
closeDomain()
While executing the above commands, I do not get any exception. But when I login to the console, I can see that ‘server1’ is still assigned to the cluster ‘cluster1’).
Is there any other step that I may be missing out to complete this process of unassignment?
On the other hand, I tried the unassignment by setting the Cluster of ‘server1’ to ‘standalone’ using the below command:
set(‘Cluster’,’standalone’)
and then saving the changes.
This command is able to successfully unassign server1 from cluster1.
Please suggest which is the correct way for unassignment.
Thanks,
Ramya
March 21st, 2012 on 5:13 pm
Hi,
In my earlier post, pls read
readDomain() as readDomain(‘domainDir’)
Thanks,
Ramya
March 21st, 2012 on 11:55 pm
What you do is OK
wls:/offline> readDomain(‘/home/oracle/weblogic12.1.1/configuration/domains/base_domain’);
wls:/offline/base_domain>unassign(‘Server’,’someserver’,’Cluster’,’loadtest-cluster’)
wls:/offline/base_domain>updateDomain()
wls:/offline/base_domain>closeDomain()
After performing these commands you can check the config.xml file. Here is indeed shows that the server is no longer part of the cluster, which is also reflected in the Admin console, after you restart the admin server.
March 22nd, 2012 on 12:05 am
Hi all,
our domain is on 10.3.2 version we have currently 2 managed servers in a cluster,but only one of these are currently running.My problem is when I try to start a deployment targeting to the cluster using the below command only to target a module in an early deployed(myEnterpriseApp) application, it deploys whole application again:
java weblogic.Deployer -url t3://IP:7001 -username xxxxx -password xxxxx -name myEnterpriseApp -targets /mywebmodule@myCluster -deploy -source .
i point the flags following;
-name [myEnterpriseApp]-deployed enterprise application as exploded archive format
-targets [/mywebmodule]-using context-uri which is tagged in application.xml
-source . -pointing to the current directory because i’ve explodeded the .ear archive and myweb.war is a part of myEnterpriseApp.
weblogic command-line reference guide defines the issue “Targeting Web Application Modules”
But i could not figure at why it tries to deploy whole application?
I’ll be so glad to hear if any one of you had a similar experience while deploying applications using command-line reference.
April 16th, 2012 on 8:54 pm
Hi,
When i start the nodeManager by running
C:OracleMiddlewarewlserver_10.3serverbinstartNodeManager
i am getting below exception:
C:OracleMIDDLE~1WLSERV~1.3commonNODEMA~1>”C:OracleMIDDLE~1JROCKI~1.1-3b
injava.exe” -jrockit -Xms128m -Xmx256m -Dcoherence.home=C:OracleMIDDLE~1COHE
RE~1.6 -Dbea.home=C:OracleMIDDLE~1 -Xverify:none “-Djava.security.policy=C:O
racleMIDDLE~1WLSERV~1.3serverlibweblogic.policy” “-Dweblogic.nodemanager.ja
vaHome=C:OracleMIDDLE~1JROCKI~1.1-3” weblogic.NodeManager -v
[ERROR][model ] Failed to patch initial classes
Could not create the Java virtual machine.
C:OracleMIDDLE~1WLSERV~1.3commonNODEMA~1>goto finish
C:OracleMIDDLE~1WLSERV~1.3commonNODEMA~1>ENDLOCAL
What can be the reason for this?
April 16th, 2012 on 9:52 pm
Thanks , I got the solution .. i changed the java path to
C:Program Files (x86)Javajre6 in commEnv.cmd file
May 16th, 2012 on 3:17 pm
Hi all.
I have two HP server with 32Gb RAM, run on RHEL 5.x. At this moment, I run a cluster with 2 managed servers and face to low HA. The available physical resource is big, so I want to add managed server to cluster. How to determine number of managed server in cluster that can improve HA of system?
Thank in advance.
June 4th, 2012 on 5:14 pm
Hi,
I am having a serious problem with weblogic and unable to reproduce.
It started with our operation team reported that weblogic ejb failover is not working. I tried to investigate with config files for settings but seems it did not help.
now I had setup a simple ejb application and tried to test and found that failover to work takes 5 min on server
here is configuration
Solaris 10 10/09 s10s_u8wos_08a SPARC
WebLogic Server 10.0 MP2
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_33-b09)
I had horizontal clustering
domain1
cluster1
m1 (simpleejb)
m2 (simpleejb)
domain2
cluster1
m3 (Webapp simple servlet calling ejb)
m4
I did not put proxy-plugin inbetween here as i want to create production env.
simultaion, both applicaation talking from diff domain over t3
I tried taking test
in test there are two threads calling ejb with gap of 40 sec.
in each call in servlet will make 3 times call to ejb with gap of 2 sec
in normal situation found that call goes to both machine m1 and m2 ejb in
round robiing fashion.
but when i put down network interface of m1 immideate after call1>>method1 but
request gets hang for about 5 min to take failover.
second thread call goes fine after 40 second to machine2 and all its method invoked on same machine.
I tried with many possiblilities by changing parameters and startup weblogic parameters
but it did not help me.
I tried adding JNDI parameters for
weblogic.rmi.clientTimeout and weblogic.jndi.requestTimeout
>> did not helped
I tried adding managed server startup parameter
-Dweblogic.t3.ConnectTimeout=10 -Dweblogic.client.SocketConnectTimeout=10 -Dweblogic.client.SocketConnectTimeoutInSecs=10
>> did not help
I tried
>> Solaris Thread(IO) timeout to 30 sec
can you please help me to find what is going wrong
August 29th, 2012 on 6:10 pm
Hi everyone,We are facing a a issue in weblogic with Apche CXF,(Have been looking for a solution since long)
We are using weblogic 11g(OCCAS) on linux machine.
Its in cluster ,application gets deployed through the admin server on engines.
We are using APACHE CXF for the REST based request processing in the application.
The issue which we are facing is :
If we stop and then start the deployed application from the console, It shows active on the Page.Now if we hit the REST web service using REST client it gives 500 -internal server error.
But only the POST REST requests are giving this error however the GET requests works fine.But if we wait for some time(ranging from 15 sec to 12 min) the POST request also start responding successfully.
The application is deployed in a cluster ,we have noted this issue on the Envoirnments where the admin-server is on another machine and the engines are on differnet machines.Its works fine on the ENV where the both the admin server and engine are on the same machine.
We have enabled the CXF logging as well but have not been able to notice any exception/error in the logs.
Exception trace is :
2012-08-22 01:54:44,812 WARN [org.apache.cxf.phase.PhaseInterceptorChain] Application has thrown exception, unwinding now
org.apache.cxf.interceptor.Fault: argument type mismatch while invoking public com.nsn.see.application.callerid.Registration com.nsn.see.application.callerid.CallerIdService.registration(com.nsn.see.application.callerid.Register) with params [com.nsn.see.application.callerid.Register@1d03015d].
at org.apache.cxf.service.invoker.AbstractInvoker.createFault(AbstractInvoker.java:159)
at org.apache.cxf.service.invoker.AbstractInvoker.invoke(AbstractInvoker.java:133)
at org.apache.cxf.jaxrs.JAXRSInvoker.invoke(JAXRSInvoker.java:130)
at org.apache.cxf.jaxrs.JAXRSInvoker.invoke(JAXRSInvoker.java:82)
at org.apache.cxf.interceptor.ServiceInvokerInterceptor$1.run(ServiceInvokerInterceptor.java:58)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:441)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
at java.util.concurrent.FutureTask.run(FutureTask.java:138)
at org.apache.cxf.workqueue.SynchronousExecutor.execute(SynchronousExecutor.java:37)
at org.apache.cxf.interceptor.ServiceInvokerInterceptor.handleMessage(ServiceInvokerInterceptor.java:106)
at org.apache.cxf.phase.PhaseInterceptorChain.doIntercept(PhaseInterceptorChain.java:236)
at org.apache.cxf.transport.ChainInitiationObserver.onMessage(ChainInitiationObserver.java:109)
at org.apache.cxf.transport.servlet.ServletDestination.invoke(ServletDestination.java:98)
at org.apache.cxf.transport.servlet.ServletController.invokeDestination(ServletController.java:394)
at org.apache.cxf.transport.servlet.ServletController.invoke(ServletController.java:133)
at org.apache.cxf.transport.servlet.AbstractCXFServlet.invoke(AbstractCXFServlet.java:142)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.handleRequest(AbstractHTTPServlet.java:179)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.doPost(AbstractHTTPServlet.java:103)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:727)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.service(AbstractHTTPServlet.java:159)
at weblogic.servlet.internal.StubSecurityHelper$ServletServiceAction.run(StubSecurityHelper.java:227)
at weblogic.servlet.internal.StubSecurityHelper.invokeServlet(StubSecurityHelper.java:125)
at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:300)
at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:183)
at weblogic.servlet.internal.WebAppServletContext$ServletInvocationAction.doIt(WebAppServletContext.java:3692)
at weblogic.servlet.internal.WebAppServletContext$ServletInvocationAction.run(WebAppServletContext.java:3656)
at weblogic.security.acl.internal.AuthenticatedSubject.doAs(AuthenticatedSubject.java:321)
at weblogic.security.service.SecurityManager.runAs(SecurityManager.java:121)
at weblogic.servlet.internal.WebAppServletContext.securedExecute(WebAppServletContext.java:2268)
at weblogic.servlet.internal.WebAppServletContext.execute(WebAppServletContext.java:2174)
at weblogic.servlet.internal.ServletRequestImpl.run(ServletRequestImpl.java:1451)
at weblogic.work.ExecuteThread.execute(ExecuteThread.java:201)
at weblogic.work.ExecuteThread.run(ExecuteThread.java:173)
Caused by: java.lang.IllegalArgumentException: argument type mismatch
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at org.apache.cxf.service.invoker.AbstractInvoker.performInvocation(AbstractInvoker.java:173)
at org.apache.cxf.service.invoker.AbstractInvoker.invoke(AbstractInvoker.java:89)
at org.apache.cxf.jaxrs.JAXRSInvoker.invoke(JAXRSInvoker.java:130)
at org.apache.cxf.jaxrs.JAXRSInvoker.invoke(JAXRSInvoker.java:82)
at org.apache.cxf.interceptor.ServiceInvokerInterceptor$1.run(ServiceInvokerInterceptor.java:58)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:442)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
at java.util.concurrent.FutureTask.run(FutureTask.java:139)
at org.apache.cxf.workqueue.SynchronousExecutor.execute(SynchronousExecutor.java:38)
at org.apache.cxf.interceptor.ServiceInvokerInterceptor.handleMessage(ServiceInvokerInterceptor.java:106)
at org.apache.cxf.phase.PhaseInterceptorChain.doIntercept(PhaseInterceptorChain.java:236)
at org.apache.cxf.transport.ChainInitiationObserver.onMessage(ChainInitiationObserver.java:109)
at org.apache.cxf.transport.servlet.ServletDestination.invoke(ServletDestination.java:100)
at org.apache.cxf.transport.servlet.ServletController.invokeDestination(ServletController.java:394)
at org.apache.cxf.transport.servlet.ServletController.invoke(ServletController.java:133)
at org.apache.cxf.transport.servlet.AbstractCXFServlet.invoke(AbstractCXFServlet.java:142)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.handleRequest(AbstractHTTPServlet.java:180)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.doPost(AbstractHTTPServlet.java:104)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:727)
at org.apache.cxf.transport.servlet.AbstractHTTPServlet.service(AbstractHTTPServlet.java:159)
at weblogic.servlet.internal.StubSecurityHelper$ServletServiceAction.run(StubSecurityHelper.java:227)
at weblogic.servlet.internal.StubSecurityHelper.invokeServlet(StubSecurityHelper.java:125)
at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:300)
at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:184)
… 9 more
I can share the detailed logs if required as well.
Thanks in advance.
August 29th, 2012 on 6:47 pm
One more query I would like to ask.I have enabled the class loaders logging in Weblogic11g.Bu ti am not sure where i can see the logs for the same.I have seen the server log files as well as the engine log files but couldn’t locate.Can anyone help.
Thanks in advance
August 31st, 2012 on 11:38 am
Hi René van Wijk,
Thanks for the response.We are sending a small XML file with 2 string values through REST.Actually we didn’t suspected the param As:
1> The application starts working after some time.
2> The issue comes up only when the admin server and engines are on different machines.
I tried one thing which worked was I packaged the CXF.jar into the EAR itself and set prefer-web-inf-classes Element.Then the application started behaving perfectly fine.SO by this its seems that the server is not able to initialize the CXF properly.This seems to be an issue with class loading .
Either the CXF jar is tried to beloaded from Admin first and then from local lib.. I tried enabling the class loader logs but couldn’t locate the logs i.e where the logs are getting added.
Will appreciate any help on this.
September 1st, 2012 on 2:53 pm
Usually the (JVM verbose) logging will be added to the normal log files. Depending on how you started WebLogic it will be
added to a .log (normal) or .out (with nodemanager). The log files can be found in the ${DOMAIN_HOME}/servers/${SERVER_NAME}/logs
directory.
Maybe it is better to define a separate log file for the verbose output, when working with JRockit you can use -XverboseLog
(http://docs.oracle.com/cd/E15289_01/doc.40/e15062/optionx.htm#i1029749). In the case of HotSpot you can use
-Xloggc:/directory/verbose.log (http://docs.oracle.com/javase/7/docs/technotes/tools/windows/java.html)
November 1st, 2012 on 12:46 am
Hi Jay/Ravish,
WE have clustering in weblogic with BPM (Oracle BPM). However, we see that because of clusters nodes, the serialization between objects fail and causes classcastException intermittently. Would you have any suggestions on the cause of the issue ? ( we have enabled round robin affinity for clusters as well. Even that doesnt seem to help)
Thanks and Regards,
Amrita
November 1st, 2012 on 1:10 pm
Hi Jay/Ravish,
I am using clusters in my environment which spans across 2 different servers. However, when I deploy my application it throw intermittent classcastException which I feel is because of marshalling and unmarshalling of objects within clusters (Managed server). If I shutdown one managed server, I dont encounter this issue. I have enabled round-robin affinity as well to the load balancer. Can you kindly help?
Thanks,
November 2nd, 2012 on 7:47 pm
Hi Rene,
Thanks for the link. Can you also help me understand how round-robin affinity affects CPU and server resource utilization? Will this cause the requests to be routed to one node only?
Thanks and Regards,
Amrita
November 6th, 2012 on 7:34 pm
Can some one help me over monitoring jvm utilization over 100 weblogic instances over windows.how can we monitor the jvm ?Is there any open source tool available for this?Is there any script available?
Weblogic10.0.3 server is going down frequently and jrocket dumps are not getting generated.Can anyone help over this
November 9th, 2012 on 7:24 pm
I’m setting up WebLogic cluster (horizontal with 2 managed servers spanning across 2 machines).I would want to have admin server on each machine and only one admin server will be active and the other will be in passive mode.Can you please explain how I can set this up with 2 adminsitration servers and how do I make it active-passive mode for admin server and how i can bring it up in case of 1st admin server going down.
Please help as I have to implement this in a production environment in a week.
December 4th, 2012 on 7:13 pm
Hi Jai,
If i configure the Java based node manager. If i need add some application jar files in the weblogic CLASSPATH and also in need alter the JVM heap size and also i need to update the latest Jrockit JVM. If i tried in weblogic server admin console “serverstart” column. But i see the managed server has taken only the default heap size like 512MB and also the application JAR files are not added the weblogic classpath. How to alter the heap size in admin console and how to add the application jar files in the weblogic classpath and how to modify the latest Jrocket JVM in the admin console.
JAVA HOME
D:Oraclewls11gjrockit_160_14_R27.6.5-32
Java Vendor
Oracle
BEA Home
D:Oraclewls11g
Root Directory
D:Oraclewls11guser_projectsdomainsmy_domain
Classpath is empty as the script values are sufficient
Arguments
-Xnohup
Security Policy File
D:Oraclewls11gwlserver_10.3serverlibweblogic.policy
User Name
weblogic
Password
welcome1
Confirm Password
welcome1
If i mentioned all those things in the serverstart column in the weblogic server admin console. Those
values are not taken during the weblogic startup.
Could you please provide the solution to resolve the issue.
Regards,
S.Vinoth Babu
December 11th, 2012 on 7:35 pm
Hi René van Wijk,
I have update the heap size in the setDomainEnv.sh it’s not taken during the weblogic starup. Then i have tried to update the heap size in commEnv.sh then the weblogic server has taken the updated heap size and also use the lastest Jrockit JVM during the weblogic starup by using the Java Based Nodemanager.
Regards,
S.Vinoth Babu
December 12th, 2012 on 1:34 pm
Open the setDomainEnv file located in the ${DOMAIN_HOME}/bin directory. Search for the following entry
if [ “${SERVER_NAME}” = “” ] ; then
SERVER_NAME=”AdminServer”
export SERVER_NAME
fi
And add below it the following
if [ “${SERVER_NAME}” = “AdminServer” ] ; then
USER_MEM_ARGS=”-Xms512m -Xmx512m -Xgc:throughput”
export USER_MEM_ARGS
fi
Note that the USER_MEM_ARGS variable overrides the JVM parameters. To make the changes effective the admin
server must be restarted.
December 12th, 2012 on 4:44 pm
Hi René van Wijk,
Thanks for your update.
If i update the heap size in the setDomainEnv.sh script for Adminserver only. I need update the heap size for all managed servers. In our environment we have four managed servers.
Adminserver –> 512MB
MS1 –> 768 MB
MS2 –> 1024 MB
MS3 –> 1536 MB
MS4 –> 1280 MB
Then we need to create five setDomainEnv.sh script or we have to add the corresponding managed servers name in the setDomainEnv.sh script.
setDomainEnv-admin.sh
setDomainEnv-MS1.sh
setDomainEnv-MS2.sh
setDomainEnv-MS3.sh
setDomainEnv-MS4.sh
(or)
if [ “${SERVER_NAME}” = “AdminServer” ] ; then
USER_MEM_ARGS=”-Xms512m -Xmx512m -Xgc:throughput”
export USER_MEM_ARGS
fi
if [ “${SERVER_NAME}” = “MS1” ] ; then
USER_MEM_ARGS=”-Xms768m -Xmx768m -Xgc:throughput”
export USER_MEM_ARGS
fi
if [ “${SERVER_NAME}” = “MS2” ] ; then
USER_MEM_ARGS=”-Xms1024m -Xmx1024m -Xgc:throughput”
export USER_MEM_ARGS
fi
if [ “${SERVER_NAME}” = “MS3” ] ; then
USER_MEM_ARGS=”-Xms1536m -Xmx1536m -Xgc:throughput”
export USER_MEM_ARGS
fi
if [ “${SERVER_NAME}” = “MS4” ] ; then
USER_MEM_ARGS=”-Xms1280m -Xmx1280m -Xgc:throughput”
export USER_MEM_ARGS
fi
Which method we have to use in our environment.
Regards,
S.Vinoth Babu
December 13th, 2012 on 1:36 pm
The second one. If all the managed servers were to be sized equally, you can also use something like
if [ "${ADMIN_URL}" = "" ] ; then # The then part of this block is telling us we are either starting an admin server OR we are non-clustered USER_MEM_ARGS=”-Xms512m -Xmx512m -Xgc:throughput” export USER_MEM_ARGS else USER_MEM_ARGS=”-Xms1024m -Xmx1024m -Xgc:throughput” export USER_MEM_ARGS fibut if they are all different you have to code it as in your example above.
January 25th, 2013 on 3:50 pm
Hi,
Am having a doubt.
We are having a Clustered production environment.
We are following a method for restarting the managed server as first to disable the FrontEndHTTP configured in channels, then we will go for a restart.
Once the server turns back we will enable it, hence it will start serving the request.
My question is : Is we need to disable all the time or if we go for a direct restart w/o disabling FrontendHTTP it will behave normally when turns into running?
Also i want to know about the heart beats how it is working internally in clustered env.
Thanks in Advance,
Srinvasan
February 23rd, 2013 on 7:24 pm
HI Team,
I have a workspace with 3 nodes( in(3 servers)and engine with 2 nodes)in 2 servers). will there be any time lag for the workspace nodes to update the details from engine since its residing in another server where there is no direct connection to engine?
Please clarify.
thanks,
Amrita
May 17th, 2013 on 5:55 pm
Hi team,
facing a crirical issue
May 17, 2013 5:06:49 AM weblogic.nodemanager.server.Handler run
WARNING: Uncaught exception in server handlerjava.io.InterruptedIOException
java.io.InterruptedIOException
at com.certicom.tls.record.ReadHandler.getProtocolVersion(Unknown Source)
at com.certicom.tls.record.ReadHandler.checkVersion(Unknown Source)
at com.certicom.tls.record.ReadHandler.readRecord(Unknown Source)
at com.certicom.tls.record.ReadHandler.readUntilHandshakeComplete(Unknown Source)
at com.certicom.tls.interfaceimpl.TLSConnectionImpl.completeHandshake(Unknown Source)
at com.certicom.tls.record.ReadHandler.read(Unknown Source)
at com.certicom.io.InputSSLIOStreamWrapper.read(Unknown Source)
at sun.nio.cs.StreamDecoder.readBytes(StreamDecoder.java:264)
at sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:306)
at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:158)
at java.io.InputStreamReader.read(InputStreamReader.java:167)
at java.io.BufferedReader.fill(BufferedReader.java:136)
at java.io.BufferedReader.readLine(BufferedReader.java:299)
at java.io.BufferedReader.readLine(BufferedReader.java:362)
at weblogic.nodemanager.server.Handler.run(Handler.java:71)
at java.lang.Thread.run(Thread.java:662)
Plaintext data for protocol T3 was received from peer
September 24th, 2013 on 12:16 am
Hello,
I am getting an SSLKeyException BAD_CERTIFICATE alert when I create a new domain on the same server. The other domain is running fine, but the created machine is not reachable. Here is some output:
javax.net.ssl.SSLKeyException: [Security:090482]BAD_CERTIFICATE alert was received from hwvalad2021 – 172.21.80.118. Check the peer to determine why it rejected the certificate chain (trusted CA configuration, hostname verification). SSL debug tracing may be required to determine the exact reason the certificate was rejected.
at com.certicom.tls.interfaceimpl.TLSConnectionImpl.fireException(Unknown Source)
at com.certicom.tls.interfaceimpl.TLSConnectionImpl.fireAlertReceived(Unknown Source)
at com.certicom.tls.record.alert.AlertHandler.handle(Unknown Source)
at com.certicom.tls.record.alert.AlertHandler.handleAlertMessages(Unknown Source)
at com.certicom.tls.record.MessageInterpreter.interpretContent(Unknown Source)
at com.certicom.tls.record.MessageInterpreter.decryptMessage(Unknown Source)
at com.certicom.tls.record.ReadHandler.processRecord(Unknown Source)
at com.certicom.tls.record.ReadHandler.readRecord(Unknown Source)
at com.certicom.tls.record.ReadHandler.readUntilHandshakeComplete(Unknown Source)
at com.certicom.tls.interfaceimpl.TLSConnectionImpl.completeHandshake(Unknown Source)
at com.certicom.tls.record.ReadHandler.read(Unknown Source)
at com.certicom.io.InputSSLIOStreamWrapper.read(Unknown Source)
at sun.nio.cs.StreamDecoder.readBytes(StreamDecoder.java:264)
at sun.nio.cs.StreamDecoder.implRead(StreamDecoder.java:306)
at sun.nio.cs.StreamDecoder.read(StreamDecoder.java:158)
at java.io.InputStreamReader.read(InputStreamReader.java:167)
at java.io.BufferedReader.fill(BufferedReader.java:136)
at java.io.BufferedReader.readLine(BufferedReader.java:299)
at java.io.BufferedReader.readLine(BufferedReader.java:362)
at weblogic.nodemanager.server.Handler.run(Handler.java:71)
at java.lang.Thread.run(Thread.java:662)
I have added the flags to turn off some validation in my startWebLogic.sh and startNodeManager.sh:
JAVA_OPTIONS=”-Dweblogic.security.allowCryptoJDefaultPRNG=true ${JAVA_OPTIONS}”
JAVA_OPTIONS=”-Dssl.debug=true ${JAVA_OPTIONS}”
JAVA_OPTIONS=”-Dweblogic.nodemanager.sslHostNameVerificationEnabled=false ${JAVA_OPTIONS}”
JAVA_OPTIONS=”-Dweblogic.security.SSL.enforceConstraints=off ${JAVA_OPTIONS}”
The AdminServer on port 7001 is working correctly, but the newly created AdminServer in the new domain on port 7006 gives the error. Please assist.
Thanks,
Grant
November 25th, 2013 on 6:15 pm
Hi Jay / René,
Frequently the managed server is goes down in production environment. We didn’t find any memory releated issue and stuck thread issue. We have configured the single sign on configuration in weblogic server by using SAML.
This weblogic domain have only one managed server. We didn’t configure the node manager in the domain.
We only find the error message like “BEA-000449 <Closing socket as no data read from it on" frequently
How to analyze the output by using the oracle RDA ( Remote Diagnostic Agent )
How to resolve the issue for frequently restarting the Production managed server.
Regards,
S.Vinoth Babu
November 26th, 2013 on 5:19 pm
Hi Jay/René,
If the Production managed server goes UNKNOWN then we have found the error message in nohup.out file.
com.gm.mobile.webservice.SharedWebService – createAreaPayLoad;userId=[sz8r1h]
java: res_query.c:251: __libc_res_nquery: Assertion `hp != hp2′ failed.
The above error message is related with OS Level?
Regards,
S.Vinothbabu
December 16th, 2013 on 4:55 pm
Hi Jay/René,
I have find the solution for frequently restart managed server issue in production Environment.
The issue is related with glibc DNS package issue in Suse Linux OS 11SP1.
Oracle provide the solution to upgrade the glibc package or upgrade the OS kernal 11SP2 ( Suse Linux OS )
Oracle Document is Doc ID 1595084.1
Regards,
S.Vinoth Babu
March 3rd, 2014 on 5:25 am
Hi,
Can i have 2 apps deployed on a single cluster, any security or session replication issues? Or is it recommended to have 2 apps on 2 different clusters ? what are the pros n cons. Please suggest. Im using just the jdbc service for these apps.
March 4th, 2014 on 9:24 pm
Hi Team,
Is it possible to setup a cluster of two managed server where each managed server resides in different subnet.
I’ll use the unicast messaging model for communication.
Actually I like to have one managed server in primary DC and the other in DR DC. Both will be up and running and serving the requests as usual.
The advantage of having this is that my MS1 will be DR for MS2 and vice versa.
Could you please let me know is it recommended and can it be done.
Thank you very much.
Regards,
Sohel.
March 5th, 2014 on 8:07 am
can some one suggest on this please?
March 7th, 2014 on 5:03 am
im getting a doubt if this blog still active!!! can someone please advise on the above…