WLS Clustering issues wrt to Multicasting.
Hello,
I am having a heck of a time trying to get the Clustering to work. THe problem seems
to be multicasting. We have two clusters, one using a multicast address of 224.0.0.1
and another with a multicast address of 224.0.0.2. Both clusters are in the same
network: 192.168.100.0/24 All four machines are hooked up to a Cisco Catalyst 4006
via fast ethernet. The cluster that is using the multicast address of 224.0.0.1 works
fine. However, the second cluster that uses the 224.0.0.2 address does not work at
all. The two machines that are in the 224.0.0.2 cluster can't see each other, we
found this out by using the Multicast tool that comes with WLS. We tried changing
the IP address to some other valid multicast address and it still does not work.
Here is the kicker, we take the Catalyst out of the picture and put in a hub and
both clusters come up fine and with any valid multicast address we issue. Does anyone
with some Catalyst knowledge have any suggestions on what the problem could be on
the 4006? I have a feeling it might be the Catalyst's fault since it works fine on
a hub. What do we need to turn on the Catalyst to make this work?
Thanks,
-Daniel-
If you have any switches in between, you can try setting the weblogic.cluster.multicastTTL=2
or some higher value.
"Tao Xie" <[email protected]> wrote:
>The address 224.0.0.1 is assigned to all systems on a subnet. The address
>224.0.0.2 is assigned to all routers on a subnet. Not sure if Catalyst
>is
>well configured.
>
>"Daniel Serna" <[email protected]> wrote in message
>news:[email protected]...
>>
>> Hello,
>>
>> I am having a heck of a time trying to get the Clustering to work.
>THe
>problem seems
>> to be multicasting. We have two clusters, one using a multicast address
>of
>224.0.0.1
>> and another with a multicast address of 224.0.0.2. Both clusters are
>in
>the same
>> network: 192.168.100.0/24 All four machines are hooked up to a Cisco
>Catalyst 4006
>> via fast ethernet. The cluster that is using the multicast address
>of
>224.0.0.1 works
>> fine. However, the second cluster that uses the 224.0.0.2 address does
>not
>work at
>> all. The two machines that are in the 224.0.0.2 cluster can't see each
>other, we
>> found this out by using the Multicast tool that comes with WLS. We
>tried
>changing
>> the IP address to some other valid multicast address and it still does
>not
>work.
>> Here is the kicker, we take the Catalyst out of the picture and put
>in a
>hub and
>> both clusters come up fine and with any valid multicast address we
>issue.
>Does anyone
>> with some Catalyst knowledge have any suggestions on what the problem
>could be on
>> the 4006? I have a feeling it might be the Catalyst's fault since it
>works
>fine on
>> a hub. What do we need to turn on the Catalyst to make this work?
>>
>> Thanks,
>>
>> -Daniel-
>
>
Similar Messages
-
WLS Clustering: beginner's question..
Hi,
Im just about to start get my hands dirty with WLS clustering. As I understand,
it is not available in the WLS6.0/6.1 eval version downloadable from the web.
[1] Do I need to purchase a web Logic licence to experiment with clustering..?
[2] If not, how and where could I obtain the cluster option for the WLS..?
Thanks in advance,
-hari
thanks very much for the info. So I shall attempt to use the eval version on experimenting
cluster support.
cheers,
-hari
"Lynch" <[email protected]> wrote:
>Yes that's true.
>
>We try our clustering on the basis of the downloaded eval version.
>
>"Mike Reiche" <[email protected]> ¼¶¼g©ó¶l¥ó
>news:3b950be7$[email protected]..
>>
>> I believe that clustering is included with the demo license.
>>
>> There is a 'cluster' entry in the demo license file.
>>
>> Mike
>
>
>
-
Error in posting Goods Issue wrt Reservation
Hi experts,
System gives following error while posting Goods issue wrt reservation
Message No - M7093
Fld selectn for mvmt type 261 / acct xxxxxx differs for Fund (xxx)
Then, I checked following things
1. RM07CUFA (SE38) report shows that
"fund" field in Mvt type 261 field selection control is OPTIONAL ENTRY
"fund" field in G/L acct xxxxxx field selection control is HIDDEN
2. In OMJJ, i did not find field " fund " for Mvt type 261
In OBC4, i did not find field " fund " for a/c xxxxxx (group Z012)
How to resolve the problem?
rgds,
jayThanks for your reply.
In SE38 , I run Program RM07CUFA with Movement type 261 and Company code and GL account.
Here it shows Fund field selection controls are different in Mvt type & G/L account
But,there " fund" field is not displayed to change its field selection control , neither in Mvt type nor in GL a/c.
pls help me out
Regards.
jay -
We are using Weblogic 4.5.1 with sp7 and have heard that there are
issues with Weblogic clustering. Specifically, the IIS plugin having
memory leaks, and 4.5.1 with sp7 has issues when implementing clustering
(the app servers in the cluster sometimes do not respond and sometimes
contend for who owns the next request coming from the proxy). I've also
seen that when a HTTP 1.1 request comes to the Web server, the proxy
sends a HTTP 1.0 request to the App server. I've also heard that sp10,
due out in July will fix these issues. Can anyone verify this?
Also, I want to make sure that our proposed configuration can work in a
clustered environment. I have two IIS Web servers, WebA and WebB, and
also two Application servers running 4.5.1 sp7, AppA and AppB. Both Web
servers are identical and both App servers are identical. Our Weblogic
ListenPort is 7005
If I set up WebA's iisproxy.ini file to contain the line
"WebLogicCluster=AppA:7005,AppB:7005"
I then copy WebA's iisproxy.ini file to WebB so that the two
iisproxy.ini files are identical.
The App servers have an identical configuration and point to 237.0.0.1
as their multicast IP address.
Is this configuration ok? Can two web servers point to the same two App
servers in the cluster, or does each Web server need it's own "cluster"
of App servers.
Thanks in advance...
What was that issue with deadlock and where is it described?
Prasad Peddada wrote:
> Also there was a dead lock issue with sp7 which was fixed in the later service
> packs.
>
> Vinod Mehra wrote:
>
> > >> Specifically, the IIS plugin having memory leaks,
> > These have already been fixed with SP7.
> >
> > >>and 4.5.1 with sp7 has issues when implementing clustering
> > >>(the app servers in the cluster sometimes do not respond and sometimes
> > >>contend for who owns the next request coming from the proxy).
> >
> > WebLogic server not responding can be because of many reasons. Most of the
> > time it is because of bad configurations and sometimes because of
> > Application
> > problems itself. Now about recovering from such a hung server SP8 onwards
> > the plugin-ins have a configurable parameter "HungServerRecoverySecs",
> > using which the plug-ins mark that server as bad (temporarily) and
> > the requets failover to the SECONDARY.
> >
> > >> I've also seen that when a HTTP 1.1 request comes to the Web server, the
> > proxy
> > >>sends a HTTP 1.0 request to the App server.
> >
> > 4.5.1 does noes not support HTTP1.1 yet. So they will still be HTTP1.0.
> >
> > >> I've also heard that sp10, due out in July will fix these issues. Can
> > anyone verify this?
> > Yes. Except the last one.
> >
> > >> Is this configuration ok?
> > Yes the configuration you have described is a valid one.
> >
> > --Vinod.
> >
> > lynch wrote:
> >
> > > We are using Weblogic 4.5.1 with sp7 and have heard that there are
> > > issues with Weblogic clustering. Specifically, the IIS plugin having
> > > memory leaks, and 4.5.1 with sp7 has issues when implementing clustering
> > > (the app servers in the cluster sometimes do not respond and sometimes
> > > contend for who owns the next request coming from the proxy). I've also
> > > seen that when a HTTP 1.1 request comes to the Web server, the proxy
> > > sends a HTTP 1.0 request to the App server. I've also heard that sp10,
> > > due out in July will fix these issues. Can anyone verify this?
> > >
> > > Also, I want to make sure that our proposed configuration can work in a
> > > clustered environment. I have two IIS Web servers, WebA and WebB, and
> > > also two Application servers running 4.5.1 sp7, AppA and AppB. Both Web
> > > servers are identical and both App servers are identical. Our Weblogic
> > > ListenPort is 7005
> > >
> > > If I set up WebA's iisproxy.ini file to contain the line
> > > "WebLogicCluster=AppA:7005,AppB:7005"
> > >
> > > I then copy WebA's iisproxy.ini file to WebB so that the two
> > > iisproxy.ini files are identical.
> > >
> > > The App servers have an identical configuration and point to 237.0.0.1
> > > as their multicast IP address.
> > >
> > > Is this configuration ok? Can two web servers point to the same two App
> > > servers in the cluster, or does each Web server need it's own "cluster"
> > > of App servers.
> > >
> > > Thanks in advance...
-
Hello,
the Weblogic Tuxedo Connector Administration Guide (v10.0) states on page
1-4 "Known Limitations":
"Does not support inbound TGIOP in clustered environments."
This seems to be a major limitation. Or am I missing something? Are there
any more details about this limitation available?
We want to integrate an existing Tuxedo Corba application with a clustered
WLS application. The Corba application will invoke EJB services.
RogerRoger Fischer <[email protected]> writes:
The docs are correct. There is an issue with the way objects are
addressed in a clustered environment that Tux does not yet understand.
You might want to contact WLS PM for an update since supporting this
has been considered for some time now.
andy
> Hello,
>
> the Weblogic Tuxedo Connector Administration Guide (v10.0) states on page
> 1-4 "Known Limitations":
>
> "Does not support inbound TGIOP in clustered environments."
>
> This seems to be a major limitation. Or am I missing something? Are there
> any more details about this limitation available?
>
> We want to integrate an existing Tuxedo Corba application with a clustered
> WLS application. The Corba application will invoke EJB services.
>
> Roger -
Hyper-V 2012 r2 Clustering issue
Hi All
We have the following issue and I just thought of why this could be happening
We are using a 3 node cluster with a witness disk
The problem is that the nodes in the cluster fail daily and its not a specific node it happens to any of them, but there is always 1 online thou
Here is the specs we are running :
*3 identical high end desktop hardware
*Server 2012 std R2
*ISCSI using free nas OS on a separate physical machine
*Teamed NIC's for Management, Clustering, ISCSI and Migration with weights defined
*9 VM's shared on this cluster between the 3
*The server have the exact same config except for the one server which is also running hyper-v with 1 vm which has our AD on
*Currently there is no backup software running for these VM's
*All the update have been done but no hot-fixes have applied
My theory why the system is crashing is because the 1 Host is running the AD VM on the HyperV which is different setup compared to the other servers.
Could this be the issue ?
ThanksWhat does the validation wizard tell you? The validation wizard should always be the first thing to run when trying to troubleshoot, as it does hundreds of tests on the environment and will alert you to anything it finds wrong or questionable.
Out of the box, it sounds like you are running an unsupported configuration since you state it is configured with "3 identical high end desktop hardware". One of the basic requirements of clustering is that you use hardware that the manufacturer
has tested for supporting Windows Server 2012 R2.
www.windowsservercatalog.com lists all that hardware. Since you are using desktops, I feel comfortable in saying that those servers are most likely not on the list. That is not to say that clustering won't work, but it does mean that you are
running an unsupported configuration. Since clustering is usually done to provide highly available access to business critical applications, building the environment on unsupported hardware is not a good way to start. This also means that you might
be using NICs that are not of server quality, meaning that you may be running into networking issues, as suggested by Darren. Then there is your iSCSI configuration. The validation wizard will tell you if you have issues with the storage - a free
NAS OS may not provide the features necessary for clustering. Suggestion is to get proper hardware. What you are describing might be okay for a learning lab, but not for production.
. : | : . : | : . tim -
Using Basic or Multi-tier configuration for WLS Clusters...
Ok,
The scenario is: I'm using WLS6.1 (hopefully, version doesn't matter here)
with three instances clustered. Each instance provides all Services:
Servlets / JSPs/ EJBs / Database communication... I'm using the Basic
configuration described on the BEA documentation....
Now, I want to move forward to a Multi-tier configuration, where the first
layer will serve Servlets and JSPs and the second layer will serve EJBs.
The problem: On the first scenario, I was looking EJBs with the Default
InitialContext() constructor without the environment, since everybody was on
the same JVM...
What do I have to do, so my Servlets start looking for EJBs on the second
cluster now? Do I have to re-write every single lookup? Can I use a
jndi.properties file on the first layer - on the presentation tier?
Thanks in Advance,
Marcelo.
Thanks Patrick!
But my actual problem starts here. The Servlets aren't mine. We have to support
applications which run atop our product(which is a bunch of beans+RMI Objects
running on the top of WLS) in a backward compatible way with a standalone.
Will try and make a post if I can find a 'configurable' way out.
Harish
Robert Patrick <[email protected]> wrote:
>Put the other server's PROVIDER_URL in a system property, property file,
>whatever and
>have the servlets explicitly use it when creating the InitialContext
>to talk with the
>other server...
>
>Harish Vajha wrote:
>
>> Hi Cameron,
>>
>> A -Djava.naming.provider.url nor jndi.properties would work. The problem
>is like
>> this:
>> Say I have two WL servers hosting just EJBs: E1, E2. And I have a single
>WL Server
>> hosting just servlets: S1. Now I give the -Djava.naming.provider.url=E1(using
>> http or t3) (or I provide the same thing using my own jndi.properties).
>>
>> The problem with this setup is that while starting up, the Server will
>try binding
>> every object(including the System bindings like wl.transaction.TransactionManager,
>> wl.common.T3Services) to the other NS (in this case E1's naming service)
>which
>> is bound to fail, as they're already bound in that Server.
>>
>> The solution has to be a way to specify the java.naming.provider.url
>only for
>> the servlets (or the web app).
>>
>> Any ideas on how to provide a Webapp specific lookup address?
>>
>> Marcelo, if you've found a way out to separate, pl. make a post.. Thanks!
>>
>> Harish
>>
>> "Cameron Purdy" <[email protected]> wrote:
>> >Hi Marcelo,
>> >
>> >With Weblogic, your performance will typically decrease if you go
>the
>> >route
>> >that you are describing.
>> >
>> >Yes, you can do jndi.properties, however I believe you will find it
>inside
>> >one of the WL jars.
>> >
>> >Note that you can use java -D instead to define some of the JNDI properties
>> >to point to the other cluster, but I can't remember them off-hand.
>> >
>> >Peace,
>> >
>> >--
>> >Cameron Purdy
>> >Tangosol Inc.
>> ><< Tangosol Server: How Weblogic applications are customized >>
>> ><< Download now from http://www.tangosol.com/download.jsp >>
>> >
>> >
>> >"Marcelo Caldas" <[email protected]> wrote in message
>> >news:[email protected]...
>> >> Ok,
>> >> The scenario is: I'm using WLS6.1 (hopefully, version doesn't matter
>> >here)
>> >> with three instances clustered. Each instance provides all Services:
>> >> Servlets / JSPs/ EJBs / Database communication... I'm using the
>Basic
>> >> configuration described on the BEA documentation....
>> >>
>> >> Now, I want to move forward to a Multi-tier configuration, where
>the
>> >first
>> >> layer will serve Servlets and JSPs and the second layer will serve
>> >EJBs.
>> >>
>> >> The problem: On the first scenario, I was looking EJBs with the
>Default
>> >> InitialContext() constructor without the environment, since everybody
>> >was
>> >on
>> >> the same JVM...
>> >> What do I have to do, so my Servlets start looking for EJBs on the
>> >second
>> >> cluster now? Do I have to re-write every single lookup? Can I use
>a
>> >> jndi.properties file on the first layer - on the presentation tier?
>> >>
>> >> Thanks in Advance,
>> >> Marcelo.
>> >>
>> >>
>> >
>> >
>
-
HFM server configuration - Application Server Clusters Issue
Hi,
When I am tryin to get into the second tab (Application Server Clusters)of the HFM configuration utility(server and webconfigurtion):
I am getting the below ERROR
+" There must be a valid Database Connection in order to set Application Sever Clustering. Please check and apply a working UDL file."+
Database and server I am using:_
I have Oracle database: Oracle Enterprise Manager Version 9.2.1.0
on Windown 2003 SP1 server
I created a UDL file and also tested it. The test connection was successful. I am using the provider for udl: "Microsoft OLE DB Provider for Oracle)
+PS: I am through with the first tab "Application Server"..+
Belolw is the log.. may be this will be useful.
Error*11*Administrator*11/09/2009 10:47:08*CHsxHFMConnection.cpp*Line 909*<?xml version="1.0"?>
<EStr><Ref>{DFCB5E4F-0940-4993-84A1-1F7568F670CB}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxHFMConnection.cpp</File><Line>909</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
Error*11*Administrator*11/09/2009 10:47:08*HfmADOConnection.cpp*Line 262*<?xml version="1.0"?>
<EStr><Ref>{1290DF08-9EC5-45B5-B846-38EA0D58E2CE}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>HfmADOConnection.cpp</File><Line>262</Line><Ver>9.3.1.0.1502</Ver><DStr><?xml version="1.0"?>
<EStr><Ref>{DFCB5E4F-0940-4993-84A1-1F7568F670CB}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxHFMConnection.cpp</File><Line>909</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
</DStr><PSec><Param>C:\Hyperion\FinancialManagement\Server Working Folder\hfmlink.udl</Param></PSec></ESec></EStr>
Error*11*Administrator*11/09/2009 10:47:08*CHsxSQLHelper.cpp*Line 102*<?xml version="1.0"?>
<EStr><Ref>{1290DF08-9EC5-45B5-B846-38EA0D58E2CE}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>HfmADOConnection.cpp</File><Line>262</Line><Ver>9.3.1.0.1502</Ver><DStr><?xml version="1.0"?>
<EStr><Ref>{DFCB5E4F-0940-4993-84A1-1F7568F670CB}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxHFMConnection.cpp</File><Line>909</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
</DStr><PSec><Param>C:\Hyperion\FinancialManagement\Server Working Folder\hfmlink.udl</Param></PSec></ESec><ESec><Num>-2147220956</Num><Type>0</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxSQLHelper.cpp</File><Line>102</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
Error*11*Administrator*11/09/2009 10:47:08*CHsxSystemConfig.cpp*Line 2508*<?xml version="1.0"?>
<EStr><Ref>{1290DF08-9EC5-45B5-B846-38EA0D58E2CE}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>HfmADOConnection.cpp</File><Line>262</Line><Ver>9.3.1.0.1502</Ver><DStr><?xml version="1.0"?>
<EStr><Ref>{DFCB5E4F-0940-4993-84A1-1F7568F670CB}</Ref><User/><DBUpdate>1</DBUpdate><ESec><Num>-2147220956</Num><Type>1</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxHFMConnection.cpp</File><Line>909</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
</DStr><PSec><Param>C:\Hyperion\FinancialManagement\Server Working Folder\hfmlink.udl</Param></PSec></ESec><ESec><Num>-2147220956</Num><Type>0</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxSQLHelper.cpp</File><Line>102</Line><Ver>9.3.1.0.1502</Ver></ESec><ESec><Num>-2147220956</Num><Type>0</Type><DTime>11/9/2009 10:47:08 AM</DTime><Svr>HFMSERVER</Svr><File>CHsxSystemConfig.cpp</File><Line>2508</Line><Ver>9.3.1.0.1502</Ver></ESec></EStr>
Can someone help me on this?This issue has been resolved.
I was using the wrong proveder in the UDL file.
The correct provider to be used is " Oracle Provider for OLE DB" - when we are running on Oracle database.
Cheers.. -
Cluster issue with Sun Message Queue 3.5
We are developing a basic ESB (Enterprise Service Bus) system with MQ3 as the JMS
provider. Naturally we want to cluster multiple brokers on different boxes incase one
fails. For development and testing we�ve setup two MQ3 brokers on two different
Windows 2000 machines and both are configured the same and have created the same
destinations on both. We have one producer client running on a RS6000 AIX machine
sending messages to a message queue, and one consumer client running on a iSeries
reading the messages. (Note: There are no problems when we�re not clustering and using
just one broker). The problem is after the consumer processes about 1k-3k messages it
like drops its active connection to the message queue. Looking at the Administration
Console it appears the active consumer rolls down to backup consumer. Here is what the
console shows.
When working ok:
Current Number of Messages: 65231
Current Total Message Bytes: 501033112 bytes
Current Number of Producers: 1
Current Number of Active Consumers: 1
Current Number of Backup Consumers: 0
When it stops receiving messages:
Current Number of Messages: 72652
Current Total Message Bytes: 515987678 bytes
Current Number of Producers: 1
Current Number of Active Consumers: 0
Current Number of Backup Consumers: 1
At this point the consumer does not pick up anymore messages, it waits on onMessage().
However, the producer keeps on running just fine. Also we�re using persistent queues,
not topics.
Following are the configuration settings.
Here is the startup screen shot, both brokers are the same except for the name, this one is
CFERRIS the other is CFERRIS2K so I just show one here.
C:\Program Files\Sun\MessageQueue3\bin>imqbrokerd -license try
[12/Apr/2005:11:35:40 CDT]
================================================================================
Sun Java(tm) System Message Queue
Sun Microsystems, Inc.
Version: 3.5 SP2 (Build 48-G)
Compile: Fri 08/27/2004
Copyright ⌐ 2004 Sun Microsystems, Inc. All rights reserved.
Use is subject to license terms.
This product includes code licensed from RSA Data Security.
================================================================================
Java Runtime: 1.4.2_05 Sun Microsystems Inc. c:\j2sdk1.4.2_05\jre
***** Trial license enabled. This license will expire on *****
***** Jun 28, 2005. This license is for evaluation purposes *****
***** only. You are in violation of the license agreement if *****
***** you use this product with this license enabled in a *****
***** production environment. *****
[12/Apr/2005:11:35:40 CDT] IMQ_HOME=C:\Program Files\Sun\MessageQueue3
[12/Apr/2005:11:35:40 CDT] IMQ_VARHOME=C:\Program Files\Sun\MessageQueue3\var
[12/Apr/2005:11:35:41 CDT] Windows 2000 5.0 x86 CFERRIS (1 cpu) cferris
[12/Apr/2005:11:35:41 CDT] Java Heap Size: max=194432k, current=16256k
[12/Apr/2005:11:35:41 CDT] Arguments: -license try
[12/Apr/2005:11:35:41 CDT] [B1060]: Loading persistent data...
[12/Apr/2005:11:35:41 CDT] Using built-in file-based persistent store: C:\Progra
m Files\Sun\MessageQueue3\var\instances\imqbroker\
[12/Apr/2005:11:35:45 CDT] [B1039]: Broker "imqbroker@CFERRIS:7676" ready.From the config.properties for both brokers, they are the same:
#Last Update:
#Fri Apr 08 15:19:45 CDT 2005
imq.jms.min_threads=10
imq.instanceconfig.version=300
imq.autocreate.queue=false
imq.jms.tcp.port=0
imq.cluster.url=file\:\\H\:\\MIS\\928DataBrokerInterfaces\\MOM\\cluster.properties
imq.autocreate.topic=false
imq.jms.max_threads=1000
From the cluster.properties file:
#Last Update:
#Fri Mar 25 15:47:49 CST 2005
imq.cluster.brokerlist=cferris2k:7676,cferris:7676
imq.cluster.masterbroker=cferris2k
From imqcmd query bkr which are the same on both boxes.
Querying the broker specified by:
Host Primary Port
localhost 7676
Version 3.5 SP2
Instance Name imqbroker
Primary Port 7676
Current Number of Messages in System 0
Current Total Message Bytes in System 0
Max Number of Messages in System unlimited (-1)
Max Total Message Bytes in System unlimited (-1)
Max Message Size 70m
Auto Create Queues false
Auto Create Topics false
Auto Created Queue Max Number of Active Consumers 1
Auto Created Queue Max Number of Backup Consumers 0
Cluster Broker List (active) CFERRIS/10.3.xxx.xx:7676 (imqbroker), CFERRIS2K/10.3.xxx.xx:7676(imqbroker)
Cluster Broker List (configured) cferris2k:7676,cferris:7676
Cluster Master Broker cferris2k
Cluster URL file:\H:\MIS\Project Office
\928DataBrokerInterfaces\A-Request-Requirements\MOM\cluster.properties
Log Level INFO
Log Rollover Interval (seconds) 604800
Log Rollover Size (bytes) unlimited (-1)
Successfully queried the broker.From imqcmd query dst
C:\>imqcmd query dst -t q -n resultsFromBOS
Querying the destination where:
Destination Name Destination Type
resultsFromBOS Queue
On the broker specified by:
Host Primary Port
localhost 7676
Destination Name resultsFromBOS
Destination Type Queue
Destination State RUNNING
Created Administratively true
Current Number of Messages 0
Current Total Message Bytes 0
Current Number of Producers 0
Current Number of Active Consumers 0
Current Number of Backup Consumers 0
Max Number of Messages unlimited (-1)
Max Total Message Bytes unlimited (-1)
Max Bytes per Message unlimited (-1)
Max Number of Producers unlimited (-1)
Max Number of Active Consumers unlimited (-1)
Max Number of Backup Consumers 3
Limit Behavior REJECT_NEWEST
Consumer Flow Limit 1000
Is Local Destination false
Local Delivery is Preferred false
Successfully queried the destination.For development and testing I am not using JDNI. The following shows
how I configured the connection factory.
public ConsumerClient(int acknowledgementMode) {
// Initiate the factory and connection.
// Also directly instantiate and configure the object.
connectFactory = new com.sun.messaging.ConnectionFactory();
try {
connectFactory.setProperty("imqAddressList",
�mq\://cferris2k\:7676/jms,mq\://cferris\:7676/jms�);
connectFactory.setProperty("imqReconnectEnabled", �true�);
// Set client ID/user ID
connectFactory.setProperty("imqDefaultUsername", "admin");
connectFactory.setProperty("imqDefaultPassword", "admin");
connectFactory.setProperty("imqAddressListIterations", "-1");
connectFactory.setProperty("imqReconnectAttempts", "5");
connectFactory.setProperty("imqAddressListBehavior", "RANDOM");
// Next create the connection and session
connection = connectFactory.createConnection();
session = connection.createSession(false, acknowledgementMode);
logger.info("A connection and session have been established by ConsumerClient");
catch(JMSException jmsError){
reportError("Error creating the connection factory. Following is the message: ",
jmsError.getMessage());
} // of constructor I added the three properties imqAddressListIterations, imqReconnectAttempts
and imqAddressListBehavior as an attempt to fix it, but this did not work, so
previously they were the defaults. I am wondering if I needed to create destination
queues on all the brokers or just the master?
Any help with figuring this will greatly be appreciated!!
Thanks, ChrisChris,
This is fishy.
Since the queue is configured to have unlimited active consumers, it should never attach a consumer as a backup. It seems like when the client reconnects, it somehow is establishing the consumers as a backup.
The first thing I'd try is to change maxNumBackupConsumers on the queue to 0. Hopefully this will either keep the consumer a primary (active) consumer, or trigger another error that would shed more light.
The other curious thing is why is the client loosing it's connection in the first place? To narrow this down I would turn off reconnect (i.e. set imqReconnectEnabled to false), and register a connection exception listener (Connection.setExceptionListener) and see what exception is being generated. Also check the broker's log file and see if there are any interesting messages.
When you have brokers clustered, you can create a destination on any broker in the cluster, and the destination will be propogated across the cluster. The following SunSolve document contains more details:
79051 Sun Java[TM] System Message Queue: Understanding the Role of a Master Broker in an MQ Cluster
And one other thing about clusters. Messages are persisted on the broker the sender is connected to. Messages are then forwarded to consumers anywhere on the cluster. Acknowledgements flow back to the "producing" broker and are processed by it. If a broker goes down, the messages it holds in its persistent store are not available until that broker comes back up. We're looking at improving data availability in a future MQ release and you can get it today in conjuction with the SunCluster product.
Joe -
Same Certificate for WLS Clustering?
Hi all.
First of all, is it worth to have 2 way authentication SSL connections among
the weblogic servers under the same weblogic clustering?
I'm wondering if I can set up same certificate/private key pair into all
weblogics running on the same machine under
weblogic clustering? or do they require to have unique key pairs?
To make 2 way authentication SSL between IIS and weblogic server, does IIS
have to obtain certificate from
trusted CA such as Verisign? Can IIS and weblogic server running at the
same machine share same certificate and
private key?
Thanks in advance.First of all, is it worth to have 2 way authentication SSL connectionsamong
the weblogic servers under the same weblogic clustering?There really is no reason to do so.
>
I'm wondering if I can set up same certificate/private key pair into all
weblogics running on the same machine under
weblogic clustering? or do they require to have unique key pairs?Digital certificates are typically tied to machine name. Each digital
certificate is associated with a unique private key. So, unless all the
machines have the same machine name, (ala machine.com), you'll need a
different visual certificate.
>
To make 2 way authentication SSL between IIS and weblogic server, does IIS
have to obtain certificate from
trusted CA such as Verisign?You're trying to do with SSL connection from the plug-in to WLS?
Can IIS and weblogic server running at the
same machine share same certificate and
private key?I do not believe that this is possible. Each vendor uses their own
mechanism for storage of the private key.
Thanks,
Michael
Michael Girdley
BEA Systems Inc
"Won H. Cho" <[email protected]> wrote in message
news:[email protected]..
Hi all.
First of all, is it worth to have 2 way authentication SSL connectionsamong
the weblogic servers under the same weblogic clustering?
I'm wondering if I can set up same certificate/private key pair into all
weblogics running on the same machine under
weblogic clustering? or do they require to have unique key pairs?
To make 2 way authentication SSL between IIS and weblogic server, does IIS
have to obtain certificate from
trusted CA such as Verisign? Can IIS and weblogic server running at the
same machine share same certificate and
private key?
Thanks in advance. -
Clustering uses IP unicast, multicast or both?
Hi.
I am just getting started with Weblogic (6.1) and reading throught the
clustering setup docs, it seems that multicast is heavily used to
communicate between instances in a cluster. However, every once in a
while, the doc talks about unicast communication between individual
peers?
Now I'm kinda confused. Is Unicast used at all for clustering?
Thx,
- deepak
Yes, that looks correct.
Deepak Balakrishna wrote:
> Don,
> Thanks for the reply. I'll attempt to summarize the findings (and holler if I
> have something wrong):
>
> a) Multicast is used among instances in a cluster for heartbeat purposes and
> for cluster wide updates of the JNDI tree
> b) Unicast is used among servers which need to share session (HTTP session as
> well as the state for stateful session)
>
> Thx,
> - deepak
>
> Don Ferguson wrote:
>
> > Sure. For example, to handle in-memory replication of servlet sessions
> > (and for stateful session beans) point-to-point communication will be used
> > between the primary and the secondary.
> >
> > Deepak Balakrishna wrote:
> >
> > > Hi.
> > >
> > > I am just getting started with Weblogic (6.1) and reading throught the
> > > clustering setup docs, it seems that multicast is heavily used to
> > > communicate between instances in a cluster. However, every once in a
> > > while, the doc talks about unicast communication between individual
> > > peers?
> > >
> > > Now I'm kinda confused. Is Unicast used at all for clustering?
> > >
> > > Thx,
> > > - deepak
-
I am trying to get a new Windows 2008 R2 cluster to work on two VMware servers (VMware ESXi 5.0.3) and am experiencing some form of timeout issue during the creation of the cluster according to the logs.
All validation checks pass successfully, whether trying to create a single node or dual node cluster, but the same error message is seen when creating the cluster. I have tried various suggestions found but none of them have made a difference. In an effort
to further diagnose the issue I enabled the following diagnostic logs:
- FailoverClustering
- FailoverClustering-Manager
- FailoverClustering-Client
The only useful error I can find is in the FailoverClustering-Client log which reports an error 'Couldn't resolve tcp binding to cluster' for each physical node to be part of the cluster.
In order to further diagnose the cause of the issue, I am looking for any options for increasing logging when attempting to create the cluster. Is there a way to increase the logging to see what the actual timeout is occurring with?
Alex.Sorry, I should have been more clear ...The servers I am attempting to create a cluster with are Virtual Servers running on a VMware ESXi (5.0.3). They have access to two shared disks (connected to each VM using a separate SCSI Controller from that for the
disk used to install the OS) with Virtual SCSI Bus Sharing set within VMware.
Each Virtual Server has two NICs configured (with static IPs) - one for internal (domain connectivity) and the other for heartbeat connectivity. The servers can ping each other via both interfaces, as expected, but only domain connectivity works via the
internal NIC.
I have tried creating a single node or dual node cluster and get the same error each time. When the cluster is attempting to be created I can see the new computer object is created in Active Directory (which is then deleted when the cluster creation fails)
and I have also tried pre-creating the computer object and specifically setting permissions on it.
When creating the cluster I have tried via the GUI and via PowerShell (both with and without the option to attach the shared storage) but always get the same type of timeout error message. I am hoping there are further debugging options that can be used
to provide a better output as to what the timeout is occurring with. -
General install and deployment problem with WLS clustering on two boxes
Hi,
I've come across a general installation problem with WLS with a cluster spanning two different physical machines and two Managed Servers, one Managed Server per physical machine.
The problem is, starting from scratch, you install WLS on both physical machines. You can then use the config wizard to create the domain, cluster and managed servers, but this is a manual typing in process.
So we've now got a WLST Script which we run with "java weblogic.WLST domaincreate.py domain.properties". This is good as it creates the domain, cluster, managed servers and JMS modules.
Now the trouble is, it doesn't do anything to the 2nd physical machine which will be running only the 2nd Managed Server.
One dirty way to get around this in order to start the 2nd managed server was to copy the domain directory to the 2nd server, but I'm sure this is bad practice and what you can actually do is improve the WLST script to actually, but the same as the config wizard and dump the nessary information to the 2nd physical machine in order so you don't need to copy the domain directory across so that it will startup straight after running the WLST.
Any help appriecated.
Surfbum.Hi I think I've found the answer. You can build a managed server template using the pack command.
http://e-docs.bea.com/common/docs90/pack/tasks.html -
Clustering issue with WLCS 3.2 weblogic 5.1 and service pack 8
I'm having a problem with Portlets on a page with commerce server 3.2 on
Cluster A below. One Portal page has multiple portlets on it. When the
Portal Page executes the portlets are being scheduled across both SCCWLS00
and SCCWLS01 and they are not able to share the session. Thus they will
throw exceptions trying to read properties from the WLCS profile which is
shared.
In Cluster B, When the portal page excutes all portlets are "pinned" to that
server and they share the information perfectly. Both Cluster Groups have
their CLUSTER-wide weblogic.properties file set for
weblogic.httpd.clustering.enable=true
weblogic.httpd.session.persistence=true
weblogic.httpd.session.persistentStoreType=replicated
weblogic.httpd.register.cluster=weblogic.servlet.internal.HttpClusterServlet
Any ideas on why this is happening? And How I can correct it?
David Marshall
==============================================
Environmental Background
CLUSTER Group A (Production)
VLAN1
StrongHold 3 (Apache 1.3.19 w/SSL) (Solaris 8)
! mod_wl-ssl.so
! WebLogicCluster
sccwls00:nnnn,sccwls01:nnnn
V
<firewall translates IPs from 1 VLAN 1 VLAN 2>
VLAN2 !
V V
SCCWLS00 (Solaris 8) SCCWLS01 (Solaris 8)
CLUSTER Group B (Development)
StrongHold 3 (Apache 1.3.19 w/SSL) (Solaris 7)
! mod_wl-ssl.so
! WebLogicCluster
weblogic00:nnnn,weblogic01:nnnn
V V
WEBLOGIC00 (Soaris 8) WEBLOGIC01 (Solaris 8)
What kind of problem do you have?
If it's not connected with JDBC, you'd better post your question
in webservices newsgroup.
Regards,
Slava Imeshev
"leopld goess" <[email protected]> wrote in message
news:[email protected]..
hy out there.
i've been working with apache soap on wl 5.1 for a while now, and
everything seems to be working allright- as long as i don't try to
install a servicepack, namely sp8 or sp10. if i do that, the entire
soap service fails to deploy.
any ideas...
thanx
leopold -
Issue wrt LMS 4.0 on VMware ESX4.1
Hi;
I am facing issue during installation of Cisco Works LMS 4.0 on Windows 2008 Server R2 Enterprise 64 bit VMware ESX 4.1 instance. The installation is termicating and give me following error:
"You cannot install CiscoWorks LAN Management Solution 4.0 on an unsupported operating system or when Terminal Services is running on the supported Windows 2003 Server Standard Edition, Windows 2003 Server Enterprise Edition, Windows 2003 R2 Server platforms, and Windows 2008 Server Enterprise Edition.
The setup program will exit when you click OK
You must either upgrade the operating system on the server to a supported version or install CiscoWorks LAN Management Solution 4.0 on another server that runs a supported operating system."
As per Cisco document LMS 4.0 support VMware ESX 4.1, Windows 2008 Server R2 Enterprise 64 bit.
Please help me in this regards; Also found the screen shot in attachment.
Thanks
Shoaibthis is not quite correct, LMS 4.0 does not support Windows 2008 R2 per default:
http://www.cisco.com/en/US/partner/prod/collateral/netmgtsw/ps6504/ps6528/ps11200/ps11202/data_sheet_c78-610760.html
it is windows 2008 with service pack 1 and 2 which are supported - not R2
But, nevertheless, it is supported with a patch available on CCO. See this doc:
http://www.cisco.com/en/US/docs/net_mgmt/ciscoworks_lan_management_solution/4.0/install/guide/instl.html#wp1616783
Maybe you are looking for
-
Applescript & Outlook; make new contact
Hi all, I'm working on an Applescript that will allow me to create a new contact in Outlook. This is what I have thus far based on some old Entourage code and the Applescript Outlook dictionary: tell application "Microsoft Outlook" activate mak
-
Accessing the variables of a nested comp
Hi, I have a main comp and a shapes comp. The shapes comp is nested in the main comp. The shapes comp contains an inner rectangle and an outer rectangle and the outer rectangle is on layer 1. In the main comp I need to change the opacity of the oute
-
Genius Function - Never worked so far in ANY Itunes Version
I'm getting the common [Genius is unavailable for the song "______"]. However, this happens for EVERY song in my music library (708 songs). I've read almost 50 forum recommendations for troubleshooting and tried these: -Reinstalling iTunes v9 -Moving
-
Hello Experts, I have implemented a tabular form on a huge view (about 300 million rows) in APEX 4.0. I then created custom items in the tabular form page for users to filter records and update them. The search would typically return few rows only. I
-
Extracting employee photo to BW
Hi Experts, Is there a method avialble to extract the employee photo uploaded in SAP HR to BW? My requriement is to display employee photo against peronnel number in xcelsius. Regards, Arun.