Xsan 2 MDC server restart hangs

Xsan 2 MDC server restart/shutdown hangs and I am forced to do a manual shutdown everytime. Here is the log when trying to do a restart or shutdown
10/16/08 2:58:43 PM loginwindow[58] ERROR | -[ApplicationManager(AppleEventHandling) _tryQuitAllForRequestor:replyEvent:] | _tryQuitAllForRequestor - vproc_set_global_ondemand returned error -1787534507
Running:
Xserve 2x3GHz Multi-Core 4GB RAM
Running OS X S 10.5.5
Xsan 2.1 (184)
Also if I try to do a terminal reboot command the server will panic.
Message was edited by: spraguga

Do you have more than one controller? Does the volume failover? Have you tried reinstalling the OS on this particular Xserve?

Similar Messages

How to upgrade OS on XSAN MDCs?

I have looked all around, but cannot seem to find a guide to help me upgrade my XSAN MDCs to the latest version of MAC OSX server. What do I back up? Do I do the primary or the secondary first? Do I have to disconnect anything? Do I have to do anything special with Open Directory too?
--bryan

For Xsan, you want to backup the /Library/Filesystems/xsan directory.
I would recommend a full system backup if possible.
Then you need to make an archive of your OD Master in Server Admin. I usually also export my users and groups out of Workgroup Manager as well, just in case.
You will then want to stop any Xsan volumes, and apply the Xsan 2.2 update. (Xsan 2.1.1 is not compatible with 10.6) Primary MDC first.
Then upgrade your primary to 10.6 and start the volume. If you have any problems, you can check the logs or revert to your backup. Worst case, you will have a backup MDC and (hopefully) an OD replica to fail to while you figure out what went wrong.

JDBC Connection pool recovery after DB server restart

I am finding that Kodo is throwing the following exception after I restart
my database server (mysql). I am doing the database server restart while my
application server is idle, so it is not during a transaction.
Communication link failure: java.net.SocketException [code=0;state=08S01]
NestedThrowables:
com.solarmetric.kodo.impl.jdbc.sql.SQLExceptionWrapper:
This is presumably due to the JDBC connection pooling. Is there a
configuration setting that allows Kodo to detect such failures and reconnect
to the database server without exposing this problem to the application
code? For example, WebLogic Server's JDBC connection pool has a setting that
enables testing a connection and recovering from such failures before
allocating it from the pool.
Ben

This is presumably due to the JDBC connection pooling. Is there a
configuration setting that allows Kodo to detect such failures and reconnect
to the database server without exposing this problem to the application
code?Not right now. You can log an enhancement request with our bug database:
http://bugzilla.solarmetric.com/

Updates to 10.5 Cause Restart Hang

Hi,
I have a mid-2006 Macbook, the warranty of which expired 7/15/09, a week before the trouble started. The internal drive died, so I removed it, booted from the Install DVD, and installed Leopard to my external drive, which worked fine. The problem is that when I try to install operating system updates the system hangs indefinitely during boot. I've tried
10.5.7 update through software update
10.5.7 standalone combo update
10.5.6 standalone combo update
10.5.1 standalone update
Each time the computer installs the update, restarts, and hangs indefinitely on boot at this point:
Typically I gave the system 25 minutes to boot each time, with at least two attempts on each install of the update. The external drive spins down and eats lunch about 5 minutes into boot. I once let it run all night and it never booted. The only fix has been to archive and reinstall the OS each time. You can imagine how time consuming it is to keep doing this.
Is this because I'm using an external boot drive??
Is this because I'm using an external monitor?? I've tried with two separate external displays. I've also tried using the Macbook screen but it's broken so I can't see what's going on during the boot sequence. Actually one of the external displays, a Gateway Monitor, doesn't display the boot sequence either, but I can see the boot sequence when I startup the laptop attached to an old APEX 12" television with a composite video cable and Apple Mini-DVI to Video Adapter. I mention this because some other folks have had a similar restart hang problem when using external displays: http://discussions.apple.com/thread.jspa?messageID=9412370&#9412370
However, I have allowed the computer ample time to boot using it's own display. If I give it 25 minutes and plug the TV in, I get the same screen.
Before and after each update, I've also tried:
repairing disk permissions
repairing the file system
reset PRAM
reset SMC
reset PMU
It is getting pesky to have to keep reinstalling the operating system after updates crash it.
I haven't tried
safe mode
single user mode
verbose mode
startup in open firmware
running Apple Hardware Test--my next step
This was originally posted in the OS X Leopard forum: http://discussions.apple.com/thread.jspa?messageID=9902392#9902392
Users there asked me to confirm I was starting up from the right disk, which isn't an issue. I've seen other users post their startup issues in the forum for their computer, rather than the Leopard forum, so I thought I'd try to post here.
Thanks all,
Anthony

So I tried again, this time with the 10.5.6 combo update, and I experienced the same restart hang and had to archive and install. But this time I tried a couple of more tricks:
1) I tried to start up in safe mode (shift), which still resulted in hang on boot.
2) I started up in single user mode (cmd+s) and ran a file system check with the command "fsck -y". I ran this twice in a row. The first result was "file system was modified" and the second result "the volume appears to be OK."
3) I started up in verbose mode (cmd+v). The computer stalled on the following line:
Applications: Ethernet Address 00:16:cb:cf:00:05

BPEL Server Restarts intermittently

Hi All,
I have got a strange problem with my BPEL Server. Every once in a while the BPEL Server restarts on its own. I do not know if it has to do something with memory leaks or performance or some other reason.
I have made my logs to roll at every 2 GB. I also noticed that the BPEL Domains are reloaded everytime the logs are rolled. If during this restart there is some process on the go then i get erroneous results.
I am not able to reproduce the issue but it does happen intermittently. Has anyone else faced this issue?

Also check logs in $OAS_HOME/opmn/logs/ for "Restart" messages, then check the timeouts for the BPEL instance in $OAS_HOME/opmn/config/opmn.xml. These are maybe too low.
--olaf

PI 7.0 Production server restarts automatically

Hi guys,
We have installed and configured correctly a PI 7.0 system on Windows 2003 Standard Edition X64-bit with SQL Server 2005. We are on SP12. Everything, SLD, RFC connections work properly.
The problem is that 4-5 days after the installation the Windows Server restarts automatically in unexpected time during the day (unexpected shutdown), even when anybody doesn't do anything in the system. In the Event Viewer we get the following message/error "System Error: Category 102 EventID 1003". There are no other messages/errors.
This is a production system and we are supposed to go live within the next days! So I need your help ASAP.
Any ideas?
Thank you.
Best Regards.
Evaggelos

1) update your network card drivers to the latest and for all devices including your motherboard chipset drivers.
2) Are you using ethernet instead of USB?
3) This URL might help you
http://www.pcreview.co.uk/forums/thread-1885463.php
Regards
Sangeetha

Canceling timers after server restart

hi,
i create a couple of timers. while the application server is up i can cancel them by getting the TimerService object of the bean it's associated with, calling getTimers() and cancelling the Timer objects returned from it.
if i restart the server (let's say it crashed), the timers created in the previous session are back to run - they are persistent by definition. but now i can't cancel them anymore: same getTimers() returns an empty Collection. i think this happens because the old timers from the previous session were not rebound to the new ejbcontext, or something like this.
my question is: how can i cancel timers, that were re-invoked after server restart.
thank you.
Message was edited by:
cyclid

Below are my tests that show canceling of timers working as expected:
Session bean methods:
public void createTimers()
{      System.out.println("### CREATE - a single-action timer that expires in 4s");
ctx.getTimerService().createTimer(4000, "single-action timer that expires in 4s");
System.out.println("### CREATE - an interval timer that expires in every 2s");
ctx.getTimerService().createTimer(2000, 2000, "interval timer that expires in every 2s");
public void cancelAllTimers() {
     System.out.println("### CANCEL - timers:");
     for (java.util.Iterator iter = ctx.getTimerService().getTimers().iterator(); iter.hasNext();) {
          Timer timer = (Timer) iter.next();
          System.out.println(" canceling ... " + timer.getInfo());
          timer.cancel();
public void logCurrentTime() {
     System.out.println("### CURRENT TIME - " + java.util.Calendar.getInstance().getTime());
public void ejbTimeout(Timer timer)
{     System.out.println("### EJB TIMEOUT - " + timer.getInfo());
Java Client test:
ejb.cancelAllTimers();
Thread.sleep(4000);
ejb.logCurrentTime();
ejb.createTimers();
Thread.sleep(8000);
ejb.cancelAllTimers();
ejb.logCurrentTime();
Thread.sleep(8000);
ejb.logCurrentTime();
Logs:
1. Shutdown servers while there is an interval timer
The log shows the interval timer came back and then was cancel after server restart
07/02/06 13:46:23 ### CANCEL - timers:
07/02/06 13:46:27 ### CURRENT TIME - Tue Feb 06 13:46:27 EST 2007
07/02/06 13:46:27 ### CREATE - a single-action timer that expires in 4s
07/02/06 13:46:27 ### CREATE - an interval timer that expires in every 2s
07/02/06 13:46:29 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:46:31 ### EJB TIMEOUT - single-action timer that expires in 4s
07/02/06 13:46:31 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:46:32 Shutting down OC4J...
C:\oc4j\10131_061009\j2ee\home>java -jar oc4j.jar
07/02/06 13:48:12 Oracle Containers for J2EE 10g (10.1.3.1.0) initialized
07/02/06 13:48:13 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:48:15 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:48:17 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:48:18 ### CANCEL - timers:
07/02/06 13:48:18 canceling ... interval timer that expires in every 2s
07/02/06 13:48:22 ### CURRENT TIME - Tue Feb 06 13:48:22 EST 2007
2. Cancel timers right after they were created
07/02/06 13:50:11 ### CANCEL - timers:
07/02/06 13:50:15 ### CURRENT TIME - Tue Feb 06 13:50:15 EST 2007
07/02/06 13:50:15 ### CREATE - a single-action timer that expires in 4s
07/02/06 13:50:15 ### CREATE - an interval timer that expires in every 2s
07/02/06 13:50:17 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:50:19 ### EJB TIMEOUT - single-action timer that expires in 4s
07/02/06 13:50:19 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:50:21 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:50:23 ### EJB TIMEOUT - interval timer that expires in every 2s
07/02/06 13:50:23 ### CANCEL - timers:
07/02/06 13:50:23 canceling ... interval timer that expires in every 2s
07/02/06 13:50:23 ### CURRENT TIME - Tue Feb 06 13:50:23 EST 2007
07/02/06 13:50:31 ### CURRENT TIME - Tue Feb 06 13:50:31 EST 2007

Deployed Files getting overwritten every time when I do a server restart

Hi,
I have deployed a web application in weblogic server. For testing purpose, instead of placing the changed file inside the war our team will change the files directly in the deployed area. Once when we restart the application server, the new files are getting overwritten with the old one. Is there any setting with which we can stop the weblogic server deploying the old war file again?

Hi,
I am stopping and restarting the weblogic server through startManagedWeblogic.sh script. I am not using the weblogic console. Some of our application developers directly drag and drop the Units in the Deployed Area for testing Purpose. Even with nostage mode the war file in the upload folder is getting redeployed at the time of server restart.

How so we schedule a reports server restart

We want to schedule a reports server restart each night that is scheduled in windows. Does anyone know the command to do this?
The reason we want to do this is because our RDF reports often error out and a simple restart of the reports server solves the issue. Has anyone else encountered this issue and if so how did you get round it?

Hi
This is really a reports question. You can auto schedule with the queue manager to run reports on the 27th of the month in background, or have a batch script to initiate them or write a new report that initiates each ot the reports so that it can be user initiated
Brian

Web Service from stateful Session Bean deleted after server restart

Hi,
i created a stateful session bean and from this a web service.
I am able to deploy it and it works really fine.
My problem is: after a server restart the web service is gone and i
have to deploy it again.
I have some other web services (but these are web services from stateless session beans)
but they "survive" the server restart.
I don't want to deploy the web service after every server restart, did I forget something or
is it a general problem of stateful session beans or web services?
Thanks in advance
Jens

Hello Björn,
I am not quite sure what your problem is but did you have a look at these articles:
http://help.sap.com/saphelp_nw04/helpdata/en/4d/993441c73cef23e10000000a155106/frameset.htm
http://help.sap.com/saphelp_nw04/helpdata/en/be/2e2c4142aef623e10000000a155106/frameset.htm
I hope they can give you some idea.
Regads
Vyara

Missing data after server restart

Hi,
We restarted our MDM server at the OS level last night. Everything seemed okay after the restart. However, when our users came in this morning, they found that some data was missing from a table. We have a Material repository with 3 main tables. Some data which have previously been populated was not appearing in our Products Detail table, the largest of the three. A large number of rows in the table had columns that failed to load with data on the repository auto-start that occured after the server restart. The auto start value in the mds.ini file was set at 300 (5 minutes) as we had seen this iisue happen once before, and thought that may help.
A stop and restart of the repository corrected the issue. Is there anything we could have done to prevent this? Is it worthwhile to try increasing the auto start delay again? Validating detail data after every server restart is not an effective option.
MDM version: 7.1 SP08
OS: AIX DB: Oracle 11g
Thanks!
Dave

Hi Dave,
I meant that when you load repository using Immediate data is loaded from .acl files (accelerator files) in a async manner.
Whereas when you load by Update Indices new acl files are generated.
Below men6tioned is a comparision between the 2 ways of laoding repository:
Load Repository Immediate
o Loads indexes from accelerator files instead of being regenerated by reading the data them from the DB and creating the index.
o Automatically builds indexes stored in out-of-date accelerator files.
o Is usually fast.
 Load Repository with Update Indices
o Rebuilds all MDM indexes files by reading and indexing data stored in the DB. The results are flushed on disk as accelerator files.
o Necessary by repository schema changes or if any error occurs.
o Might take significant amount of time to complete.
So when you observe some visual data impact in your repository it is good idea to rebuild indexes.
I guess we cannot schedule this as default behavior is to load by immediate even through the inin parameter.
as a best practice one can do this by fixing a time frame eg monthly depending on data/schema changes.
Thanks,
Ravi

Best Practice to troubleshoot a Crawler Issue after SQL Server restarts

Hi Everyone
I am after some general advice or a suggested place to start when troubleshooting crawler issues after an unplanned, out-of-sequence server restart.
Specifically, the SQL Database in the SharePoint 2010 farm. Better yet, is there a standard practice way of resolving such issues?<o:p></o:p>
So far articles I have found suggest options from reviewing the crawl logs, creating a new crawl component, right through to using fiddler to monitor the crawler.
Are these sufficient places to start / methodologies to follow, what else should be considered?
Any advice greatly appreciated.
Thanks,
Mike

Well, as I said before, there are lots of different potential issues & resolutions for crawlers. It really depends on the issue. I would say that the base troubleshooting steps start the same no matter which service/feature you are looking
at. So, I'll try to keep this sort of generic, but beyond finding the details of the problem, the SOP or process will vary greatly based on what the error is. I hope this helps, and sorry if it's not specific enough.
1 - check the ULS logs
2 - check the windows application logs
3 - verify related services are running (get-spserviceinstance), possibly stop/start them to reprovision the instance on the affected server
4 - clear the config cache (this alone will clear about 30% of your basic problems)
6 - verify disk space & resource consumption on affected server (& SQL, SQL is always the potential to be the true "affected" server)
7 - iisreset
8 - verify connectivity between all servers in the farm and SP
9 - verify requir3ed features activated
10- check if any changes were made to environment recently (new hardware, updates to OS or apps, updates to GPOs, new solutions, etc)
11- check if the issue is reproducible in another environment (only reliable if this is a similar environment, i.e. same patch level test farm or dr farm). see if it occurs in other site collections within the web app, different web apps, different
servers/browsers, etc, basically just try to nail down the scope of the problem
There is a whole slew of thiings you could check from verifying certificates & perms, to rerunning psconfig, to checking registry keys, again I go back to it depends. hopefully this can get you started though. in the end ULS is where all
the real info on the issue is going to be, so make sure you check there. don't go in with tunnel vision either. if you see other errors in ULS, check them out, they may or may not be related; SharePoint is an intricate product with way more moving
parts than most systems. fix the little quick ones that you know you can handle, this will help to keep the farm clean and healthy, as well as crossing them off the list of potential suspects for your root cause.
Christopher Webb | MCM: SharePoint 2010 | MCSM: SharePoint Charter | MCT | http://christophermichaelwebb.com

Composites in "Unknown" state after managed soa server restart

Hi,
I am using SOA suite 11.1.1.6 version on 64 bit OEL Linux. I am getting following error in managed server output log files:
Feb 20, 2013 5:59:32 AM IST> <Error> <oracle.integration.platform> <SOA-20003> <Unable to register service.
oracle.fabric.common.FabricException: Error in getting XML input stream: oramds:/deployed-composites/WirelessEnterprise/svcEnrichSEARequest_rev1.0/svcEnrichSEARequest.wsdl: oracle.mds.exception.MDSException: MDS-00054: The file to be loaded oramds:/deployed-composites/WirelessEnterprise/svcEnrichSEARequest_rev1.0/svcEnrichSEARequest.wsdl does not exist.
Though this composites gets loaded, the composites dependent on this composite goes into "Unknown" state. Other composites are properly loaded on server restart.
Please let us know if anybody has any solution to this problem.

Hi,
Thanks for your help.
we have narrowed down to problem now.
The actual problem occurs while redeploying same composite on same revision.
E.g 3 composites A, B and C are deployed on and in "Up" state.
B And C composites depends on A.
B composite composite.xml has following entry
<import namespace="http://xmlns.oracle.com/A"
location="oramds:/deployed-composites/Enterprise/A_rev1.0/A.wsdl"
importType="wsdl"/>
same entry exist for C composite.xml as well.
Now whenever composite A is redeployed with same revision (1.0 in our case), and managed servers are restarted, B and C composites goes in to "Unknown" state.
Moving A.wsdl to /apps/wsdl/* in MDS works but it duplicates the A.wsdl file beacause it also exists in MDS @ /deployed-composites/Enterprise/A_rev1.0/A.wsdl
Please let us know if anyone has any solution.
Thanks,
Rupesh.

Composite with dependency not working after soa server restart

Hello,
I have composite application that invokes another composite application. After deploying those work fine. But after restarting soa server the one having dependency does not work any more. To soa server log I get following:
<Jul 1, 2010 11:45:29 AM EEST> <Error> <oracle.integration.platform> <SOA-20003> <Unable to register service.
oracle.fabric.common.FabricException: Error in getting XML input stream: http://Yacico:8001/soa-infra/services/default/validationForCC/getStatusByCC?WSDL: Response: '503: Service Unavailable' for url: 'http://Yacico:8001/soa-infra/services/default/validationForCC/getStatusByCC?WSDL'
at oracle.fabric.common.metadata.MetadataManagerImpl.getInputStreamFromAbsoluteURL(MetadataManagerImpl.java:276)
Caused By: java.io.FileNotFoundException: Response: '503: Service Unavailable' for url: 'http://Yacico:8001/soa-infra/services/default/validationForCC/getStatusByCC?WSDL'
<Jul 1, 2010 11:45:30 AM EEST> <Error> <oracle.integration.platform> <SOA-20020> <Deployment of composite "POProcessing" failed: Unable to find a WSDL that has a definition for service {http://oracle.com/sca/soapservice/POProcessing/POProcessing/receivePO}receivePO and port execute_pt. Please make sure that the port attribute for the binding defined in the composite file is correct by checking the namespace, service name, and port name. In addition, check that the WSDL associated with the binding namespace is imported and currently reachable (check the import nodes at the top of the composite file). Finally, validate the HTTP proxy settings for the server..>
So POProcessing does not work any more after server restart. validationForCC works fine also after server restart. Url http://...validationForCC/getStatusByCC?WSDL points to wsdl file and is visible by browser.
I use soa suite 11g patch set 2 (11.1.1.3) running on redhat enterprise linux 5.
Any idea what is the problem?
Is it somehow possible to configure which composites are started first during server startup?
regards, Matti
Edited by: user10197965 on Jul 1, 2010 2:28 AM

Yes, I did that. I'm not all that happy about this as a solution either, but it's better than making multiple copies.
We have since found out that his is a known bug and that it is fixed in some, but not all, deployments.
-------- see below -------------
Composites With WSDL Dependencies Fail To Deploy Following SOA Server Restart [ID 1272070.1]
Modified:Jul 19, 2012 Type:PROBLEM Status:MODERATED Priority:3
In this Document
     Symptoms
     Cause
     Solution
     References
This document is being delivered to you via Oracle Support's Rapid Visibility (RaV) process and therefore has not been subject to an independent technical review.
Applies to:
Oracle SOA Platform - Version 11.1.1.3.0 and later
Information in this document applies to any platform.
Symptoms
A SOA Project has an external reference to a Web Service or a reference to another Composite.
When the SOA Server is started, the Composite is trying to access the WSDL of it's referenced Web Service in order to load data structures. If SOA can not find the WSDL then the Composite fails to load/deploy.
Once this happens the Composite can not be started, or shutdown, from Enterprise Manager Fusion Middleware Control Application.
Related Error Messages:
[ERROR] [SOA-20020] ... Unable to find a WSDL that has a definition for service ... Please make sure that the port attribute for the binding defined in the composite file is correct by checking the namespace, service name, and port name. In addition, check that the WSDL associated with the binding namespace is imported and currently reachable (check the import nodes at the top of the composite file). Finally, validate the HTTP proxy settings for the server.]
javax.wsdl.WSDLException: WSDLException: faultCode=INVALID_WSDL: Error reading import of oramds
Cause
When SOA Server is restarting the Composite can not access the WSDL of it's referenced Web Service (it is not available).
In the first scenario there are two Composites on the same server: CompositeA and CompositeB.
In SOA Suite 11g there is no possibility to specify the load order for the composites. If CompositeA references CompositeB and if CompositeA is firstly loaded then it can not access the WSDL from CompositeB and the issue occurs.
In the second situation there is one Composite on the server (CompositeA) which references an external WebService (ExternalWS). If the ExternalWS is not available when the SOA Server is starting then this issue occurs.
Solution
Solution 1
Redeploy the affected Composite into the SOA Server.
You can do that but it is not advisable in a production environment and in a development environment it will take a lot of time to redeploy the composite manually
Solution 2
Copy the abstract WSDL locally into the project.
This is documented in these articles in our Knowledge base:
•     Document:1155033.1 Node Restart Cause Composites To Become Unavailable. Response: '503: Service Unavailable'
•     Document:1151973.1 Boot Order Of The Composites Upon Soa Suite Restart
Steps:
o     a. Copy and use the WSDL file in the Project
o     b. Edit the WebService Adapter
o     c. Change the "WSDL URL" to point to the WSDL copied into the project
o     d. Redeploy the Project
o     e. Make this configuration with all the Projects that have a references with other Web Services
Solution 3
Use shared artifacts in Metadata Service (MDS). A WSDL used by more than one composite is a shared artifact per definition. If the WSDL structure is changed (that is not happening frequently in a production environment), you will normally deploy a composite with a new version because overwriting it will break your production environment. Moreover, in a development environment you will need a proper process in place where different developers access the same artifacts within MDS. In addition, when you use external WSDLs (owned by third parties), you will have a proper process/agreement in place to be notified about modifications affecting your applications.
1.     In order to use shared artifacts a MDS connection must be configured in jDeveloper:
http://download.oracle.com/docs/cd/E14571_01/integration.1111/e10224/sca_lifecycle.htm#SOASE85488
2.     Deploy the shared artifacts:
o     a. Create a JAR profile and include the artifacts to share
o     b. Create a SOA bundle that includes the JAR profile
o     c. Deploy the SOA bundle to the application server
http://download.oracle.com/docs/cd/E14571_01/integration.1111/e10224/sca_lifecycle.htm#SOASE85472
If the shared artifacts (WSDLs) needed are from other composites this step can be skipped
3.     Create a new WebService
o     a. When completing the "WSDL URL" click on "Find existing WSDL's"
o     b. Select "Resource Palette"
o     c. Go to the SOA-MDS
o     d. Select a WSDL from a Composite or the one deployed at step 2.
Known Restriction 1
The port and the location for the WSDL reference is not completed automatically by jDeveloper in the composite.xml file. This information must be inserted manually. In Bug:10287325 is raised this issue and it is available for SOA 11gPS2+.
Known Restriction 2
Another issue regarding MDS caching is raised in Bug:10218147 - the MDS cache is not refreshed when an artifact is deployed or deleted.
In order to refresh this cache the server must be restarted. This Bug resolves the issue but in a production environment the server restart will be the preferred option. The reason for that is because in a production environment you never deploy single composites under the same version after artifacts (WSDLs) have been modified.
The fix for the Bug:10218147 can be an acceptable solution in development environment where redeployment would be quicker than restarting the server.
Bug:10218147 is available for SOA 11gPS2+.
To find out more information about how to use Shared Metadata check the following documentation:
Oracle Fusion Middleware Developer's Guide for Oracle SOA Suite 11g
41 Deploying SOA Composite Applications
41.7.3 Deploying and Using Shared Metadata Across SOA Composite Applications in Oracle JDeveloper
URL:
http://download.oracle.com/docs/cd/E14571_01/integration.1111/e10224/sca_lifecycle.htm#CACFEAJJ
Solution 4
The last solution is to use an UDDI (Universal Description Discovery and Integration).
Oracle recommends Oracle Service Registry 11g (OSR). The advantage is that you can use OSR and SOA Suite in a Shared WebLogic Domain.
http://www.oracle.com/technetwork/middleware/registry/overview/index.html
http://blogs.oracle.com/governance/2010/05/oracle_service_registry_11gr1.html
To find out more information about Oracle Service Registry check the following documentation:
Oracle Fusion Middleware Administrator's Guide for Oracle SOA Suite 11g
33 Configuring Service and Reference Binding Components
33.1.3 Changing the Endpoint Reference and Service Key for Oracle Service Registry Integration
http://download.oracle.com/docs/cd/E15523_01/integration.1111/e10226/bc_config.htm#SOAAG37248
Oracle Fusion Middleware Developer's Guide for Oracle SOA Suite 11g
A BPEL Process Activities and Services
A.4 Publishing and Browsing the Oracle Service Registry
http://download.oracle.com/docs/cd/E15523_01/integration.1111/e10224/bp_appx_ref.htm#SOASE85561
Oracle Fusion Middleware Service Registry 11g
http://download.oracle.com/otndocs/tech/soa/OSR11gR1ProductDocumentation.pdf
Solution 5
In case you have BPM Components in the Composites deployed check the following note:
Document 1317803.1 Soa Suite Composite Fails To Deploy Upon Restart Of Managed Server
The issue is caused by the Bug:11822470 SOA SUITE COMPOSITE FAILS TO DEPLOY UPON RESTART OF MANAGED SERVER
References
BUG:10218147 - WSDL CHANGES NEED SOA SUITE SERVER RESTART
BUG:10278478 - WHEN SOA SERVER IS RESTARTED, SOME SOA COMPOSITES COULD NOT LOAD/BE DEPLOYED
BUG:10287325 - ABSTRACT WSDL NOT AVAILABLE WHEN COMPOSITE STARTED
BUG:10311698 - WHEN SOA SERVER IS RESTARTED, SOME SOA COMPOSITES CAN NOT BE STARTUP / SHUTDOWN
BUG:11822470 - SOA SUITE COMPOSITE FAILS TO DEPLOY UPON RESTART OF MANAGED SERVER
@ BUG:9267312 - MDS ARTIFACTS ARE STILL CACHED AFTER DELETING
@ BUG:9708488 - AFTER SOA RESTART, ALL PROCESSES HAVE TO BE REDEPLOYED
@ BUG:9749845 - SCHEMA CACHE STARTS EMPTY AFTER RE-START, BUT NOT AFTER DEPLOYMENT
NOTE:1151973.1 - Boot Order Of The Composites Upon Soa Suite Restart
NOTE:1155033.1 - Node Restart Cause Composites To Become Unavailable. Response: '503: Service Unavailable'
NOTE:1317803.1 - Soa Suite Composite Fails To Deploy Upon Restart Of Managed Server
Bug 11822470 : SOA SUITE COMPOSITE FAILS TO DEPLOY UPON RESTART OF MANAGED SERVER
               Bug Attributes
Type     B - Defect     Fixed in Product Version     11.1.1.6
Severity     2 - Severe Loss of Service     Product Version     11.1.1.4
Status     80 - Development to QA/Fix Delivered Internal     Platform     912 - Microsoft Windows (32-bit)
Created     Mar 1, 2011     Platform Version     2003
Updated     Oct 12, 2012     Base Bug     N/A
Database Version     N/A     Affects Platforms     Generic
Product Source     Oracle
Abstract: SOA SUITE COMPOSITE FAILS TO DEPLOY UPON RESTART OF MANAGED SERVER
*** 03/01/11 08:07 am ***
Customer has a project consists of two services for interaction with the
process and one reference to an external service.
On deployment of the process to an Enterprise Environment, the process works
as expected. However, upon a restart of the managed server, the process will
then fail to deploy.
Workaround used is to redeploy the project again.
Bug 10278478 : WHEN SOA SERVER IS RESTARTED, SOME SOA COMPOSITES COULD NOT LOAD/BE DEPLOYED
               Bug Attributes
Type     B - Defect     Fixed in Product Version
Severity     2 - Severe Loss of Service     Product Version     11.1.1.3.0
Status     92 - Closed, Not a Bug     Platform     226 - Linux x86-64
Created     Nov 10, 2010     Platform Version     RED HAT ENTERPRISE LINUX 5
Updated     Dec 10, 2010     Base Bug     N/A
Database Version     N/A     Affects Platforms     Generic
Product Source     Oracle
Abstract: WHEN SOA SERVER IS RESTARTED, SOME SOA COMPOSITES COULD NOT LOAD/BE DEPLOYED
Detailed Problem Description
====================
When a SOA composite get dependencies with other SOA composites located on the same server, if the dependency is still not loaded, the SOA composite will not be loaded, and there is no possibilities for Entreprise Manager console to restart it (EM crash). The only way to make it working is to redeployed the SOA composite from jDeveloper that is not acceptable in a production environment.
The composite that has the issue cannot be start-up or shut-down, although the Enterprise Manager do not show any errors with it.

Business Objects 4.0 Server Restart Order

Hi All,
SAP has recommended to us to do a monthly server restart for our servers to clean out temporary and orphaned jobs and such.
We currently have a clustered CMS environment with another clustered processing tier for the adaptive job servers.
My question is the order in which this restart should happen.
Should we just stop and restart the SIA? Will that accomplish the goal?
Or should we stop the job servers, then the CMS and then restart the CMS and next the job servers?
If anyone has any advice, please let me know.
Thanks!

I agree with SAP to restart the servers at least weekly once or monthly once to resolve the stuck processes, orphans or zombies in the system.
I have seen when the CMS db connections overload with SQL server happened it didn't resolve the issue with simple SIA restart then it forced us to restart the servers.
Either the servers are VM or Physical it is best to restart the servers weekly once in the following order.
If you are in clustered mode assuming 2 CMS nodes with multiple processing nodes.
I would recommend to setup some batch job to execute all the stops/restarts from one server instead of from each server.
1) Write the batch script and place it on last processing node ( Assume 2 CMS nodes and 4 Processing nodes then script will run from 4th processing node)
          Stop the SIA/Tomcat on Processing node 4
          wait for 3 minutes
          Stop the SIA/Tomcat on Processing Node 3
          Wait for 3 minutes
          Stop the SIA/Tomcat on processing Node 2
          Wait for 3 minutes
          Stop the SIA/Tomcat on Processing Node 1
          Wait for 3 minutes
          Stop the SIA/Tomcat on CMS node 2
          Wait for 5 minutes
          Stop the SIA/Tomcat on CMS Node 1
          Wait for 5 minutes
          Restart CMS Node 1
          Wait for 7 to 10 minutes
          Restart CMS Node 2
          Wait for 7 to 10 minutes
          Restart Processing Node 1
          Wait for 7 to 10 minutes
          Restart Processing Node 2
          Wait for 7 to 10 minutes
          Restart Processing Node 3
          Wait for 7 to 10 minutes
          Restart Processing Node 4
You have to make sure that your CMS nodes are starting prior to processing nodes to avoid any issues.
Make sure no network communication issues between servers other wise some times it leads to some services on processing nodes lead to either Running/Disabled or Initializing/Disabled which requires force termination and start of the service.
You have to do the validation testing after every week restart.

Xsan 2 MDC server restart hangs

Similar Messages

Maybe you are looking for