Determining ORACLE_HOME of a dead Instance

Db version 10g Release 2
OS: AIX
Lets say a machine have 3 dead instances(oraprod45, oraprod46, oraprod47) and now i want to bring up oraprod45. But this machine doesn't have an oratab file and oraenv due to a slightly flawed installation. How can i know the ORACLE_HOME of oraprod45?
Message was edited by:
GarryB

Thanks everyone.
Two questions:
1)So which is the best way to determine ORACLE_HOME
of a dead instance?
Is this something you inherited and now have to get back on its feet with no good documentation?
If there is no oratab, I'd say the installation was more than 'slightly' flawed!
There's no simple formula. You'll have to do some detective work.
1) Do you have any reason to believe there is more than one Oracle home?
2) Where is your listener.ora and sqlnet.ora? They should be in $ORACLE_HOME/network/admin, so if you find them in a <something>/network/admin, the <something> should be the value of ORACLE_HOME.
3) While there are no guarantees, (see my reply to Jaffy) I'd still look at the .profile for the oracle os account and see if there are any clues there. If there are, and they match what you found in step 2, I'd be comfortable in going with that.
4) report back what you find and we'll go from there.
2) How can i make the listener automatically attach
to a newly created instance? ie. How to enable
Self-registration for a listener?Listeners don't attach to instances. They listen for and service connection requests. In 10g, instances self-register by default.

Similar Messages

How do we determine if a Human Task instance in BPM Workspace is open?

Hi All,
An interesting one. How do we programatically determine if a BPM Workspace instance is opened for processing? I would like to ensure that a form is opened only by one user at a time, and any attempt to open the form concurrently will result in the form presenting an error message and then closing itself.
Any ideas?
I am posting this on the SOA form as well, as it is the BPM Workspace. Kindly overlook if this is inappropriate.
JDev : 11g.
Thank you.

posting it here will help you for sure
Business Process Management Suite
SOA Suite

Error when determining attribute 'OBJECTTOAPPROVE' of object instance

Hi,
I'm using Item-based level approval shopping cart. I've written code in BADI for approval level as per our requirement. When I create shopping cart, it is triggering workflow and going to particular user, but when I saw the step history for the approval activity it shows, Error when determining attribute 'OBJECTTOAPPROVE' of object instance BO.BUS30003.
And when I execute the workitem, the status changes to error state and it has one more message in step history which is Work item 000000006030: Object FORM method HTMLPROCESS cannot be executed.
Any one knows about this errors ?
Regards,
JMB

Hi..
In simulating the workflow is it working.
Regards,
Surjith

Dead instances in corba mode on Mac OSX

Hello,
Since our transition to IDS CS5 we often experience that corba instances run into a dead-lock randomly.
With a dead instance, i mean that the process is still running, the ior file still exists and seems to be valid (the contentstring is the same as on startup), but the application object is lost. It occurs randomly somewhere during execution of our code and almost always on a different part of our code.
I can simulate this state of IndesignServer by running a corba instance on my MacBook Pro, closing the MBP and wait for about an hour. Then reopen it and trying to connect. Then i get:
org.omg.CORBA.COMM_FAILURE: vmcid: SUN minor code: 201 completed: No
       at com.sun.corba.se.impl.logging.ORBUtilSystemException.connectFailure(ORBUtilSystemExceptio n.java:2200)
       at com.sun.corba.se.impl.logging.ORBUtilSystemException.connectFailure(ORBUtilSystemExceptio n.java:2221)
       at com.sun.corba.se.impl.transport.SocketOrChannelConnectionImpl.<init>(SocketOrChannelConne ctionImpl.java:205)
       at com.sun.corba.se.impl.transport.SocketOrChannelConnectionImpl.<init>(SocketOrChannelConne ctionImpl.java:218)
       at com.sun.corba.se.impl.transport.SocketOrChannelContactInfoImpl.createConnection(SocketOrC hannelContactInfoImpl.java:101)
       at com.sun.corba.se.impl.protocol.CorbaClientRequestDispatcherImpl.beginRequest(CorbaClientR equestDispatcherImpl.java:152)
       at com.sun.corba.se.impl.protocol.CorbaClientDelegateImpl.request(CorbaClientDelegateImpl.ja va:118)
       at org.omg.CORBA.portable.ObjectImpl._request(ObjectImpl.java:431)
       at com.adobe.ids.basics._ApplicationStub.updateFonts(Unknown Source)
       at be.nss.ids.advit.AdvitSession.<init>(AdvitSession.java:98)
       at be.nss.documentserver.controller.action.OpenSubDocumentAction.execute(OpenSubDocumentActi on.java:50)
       at be.nss.documentserver.controller.ActionQueue$ActionThread.run(ActionQueue.java:231)
Caused by: java.lang.NullPointerException
       at com.sun.corba.se.impl.transport.DefaultSocketFactoryImpl.createSocket(DefaultSocketFactor yImpl.java:59)
       at com.sun.corba.se.impl.transport.SocketOrChannelConnectionImpl.<init>(SocketOrChannelConne ctionImpl.java:188)
       ... 9 more
This is exactly what i get on our customers servers (which don't "go to sleep" obviously). Just setting my MacBook Pro to a very short sleep time (1 min.) doesn't invoke this error. This makes it hard to simulate the problem. Apparently the sleep invoked by closing the MBP and waiting an hour isn't the same as putting it asleep, all parameters on, after 1 min.
All OSX versions are 10.6
So i have two questions:
1) What are the possible causes of those dead-locks? Errors in our code (both java and our own IDS plugin) normally cause a real instance crash which is traceable.
2) Is there a way to re-establish a connection to a dead instance?
Thanks in advance for any input!
Bart Devos.

No one any input? It becomes a real problem reported by several of our customers since we changed to CS5.
Can anyone point me into possible directions we should investigate?
What we checked already:
1) If no more than 1 thread accesses an instance at a time.
2) We checked out what the error log really meant. We found out that this problem was reported since java 1.4.20. We compile all code for 1.5, but customers work java 1.6.24/26. No idea if this is an issue.
3) We checked out the ORB string and found out that the connection was made with an ipaddress and not with a servername, excluding dns problems.
We are really stuck here.
Thanks in advance for any usefull input.
Bart Devos.

Determining ORACLE_HOME location from DBA_ or V$ views

DB Version:10gR2
Is there a DBA_ or V$ view which i can query to determine ORACLE_HOME location?
Edited by: james_p on Oct 13, 2008 3:18 AM

SELECT substr(file_spec,1,instr(file_spec,'lib')-2) ORACLE_HOME FROM dba_libraries
WHERE library_name='DBMS_SUMADV_LIB';Interesting idea. Unfortunately when I tried the query on 10g NULL was returned :(
The oracle home might be extractable from V$PARAMTER name = 'user_dump_dest'
SQL> c/name/value
1* select value from v$parameter where name = 'user_dump_dest'
SQL> /
VALUE
C:\ORACLE\PRODUCT\10.2.0\ADMIN\ORCL\UDUMP

How to determine Oracle_Home default from multiple home installation?

Does anyone know how to determine the current Oracle Home default in the registry when multiple Oracle Home exists?
Both 8i and multiple 9i Version X's exist on a client machine. We need to determine the current default oracle home to determine path for various drivers.
In analyzing the Oracle Home registry keys I have been unable to find a means for identifying the current Oracle Home default.

Thank you for you reply, your advice is good. Unfortunately our application has to deal with all kinds of Oracle configurations and one of those is the multi-home issue. We have looked throughout the registry particular in the HKLM\SOFTWARE\ORACLE|ALL_HOMES\
but have not found any entry that would indicate a default home. One two different system installation of multiple client installations there is no DEFAULT_HOME. We do have the HOME0, HOME1, ... HOME4 but, these are just sequencial based on when they where installed not on the order of preference or any default setting. Problem is our application needs to know the default home to locate the appropriate drives to use and appropriate connects service to look-up(tnsname) also.
Thanks.

Unable to delete dead instances

Due to somethin g wrong happened (shuch as an abnormal end or crash), it seems that the oidmon tables contain wrong information.
Now trying to stop oidmon also tries to stop non-existing processes. And so oidmon hangs.
Here are messages displayed in oidmon log:
2001/06/29::11:01:24: Could not delete the Instance
2001/06/29::11:01:24: OIDLDAPD Instance 2 deletion failed
2001/06/29::11:01:24: Cannot Stop server instances with PID 0
2001/06/29::11:01:24: Deleting OIDLDAPD instance 4 from Process table
2001/06/29::11:01:24: Instance cn=instance4,cn=osdldapd,cn=subregistrysubentry not found.
I'm looking for any method to refreshing info for oidmon, thus seeting its internal tables into consitent state.

hi,
Try restarting your oidctl processes with
the argument flags=" -f" on the oidctl command line. Use other arguments to oidctl as you did previously, before the processes were killed abnormally.
eg,
oidctl connect=xxxxxx server=oidldapd instance=4 flags=" -f" start
The -f flag means "force" this command to overwrite oid info in the database.

How determine repository for existing IDM instance

Hello all! Here is an interesting one. I've inherited an IDM instance and have to find out about its repository (nobody seems to know as the consultants who set it up are long gone). I looked through the idm.war file but don't see anything regarding where its database instance is (I'm sure its not in LDAP or a flat file). Where is this data persisted and how can it be changed if need be? :-)
Thanks!

Sorry - didn't see the last part of this.
If you want to change the assigned REPO, you can either use the correct flags / options with the "lh setRepo" command or you can run "lh setup" instead. The "lh setup" will launch a wizard-based window to take you through the steps to connect. This is where you can specify (drop-list, text entry) the type of REPO (files, DB) and how-to connect to it. Just be careful that when the REPO was originally created using the provided scripts, it would likely have been modified so that there's a specific account (ID, Password) that would be enabled for access. So net -- if you don't know the account credentials for remotely connecting to the existing or alternate REPO, I wouldn't go changing which REPO the instance points to.
Make sense?

Satellite A350D-203: How To Determine if Motherboard is Dead

My partner has had her Satellite for just over 2 years original Vista operating system & it has just died.
Symptoms are that when power button is pressed the button lights up & a click is heard which suggests that HD is attempting to boot/spin up but then no further sounds ... cooling fan does not start ... nothing displayed at all.
The power button can be switched off by pressing & holding it which suggests that unit must be monitoring ...
My partner recalls that prior to unit finally failing it used to get hot & shut itself down [again suggesting that overheat detection & shut down was functioning].
I don't believe the HD is faulty since a friend of my partner was able to back up data onto disks for her.
Can anyone suggest a logical process to follow?
If it subsequently transpires that the Motherboard is faulty, has anyone any experience of any basic visual examination/tasks that can be carried out ... i.e. re-soldering of known 'weak spots'/recognised component failures?
My apologies for asking such basic questions but the net is awash with 'experts' & conflicting information suggesting a range of faults from BIOS .... Motherboard.
PS: Although I am no PC expert I hold a degree in Electronic Engineering & am simply seeking some logical & credible advice before shelling out on a replacement Motherboard.
KR
Martin

> Symptoms are that when power button is pressed the button lights up & a click is heard which suggests that HD is attempting to boot/spin up but then no further sounds ... cooling fan does not start ... nothing displayed at all.
This does not mean that HDD is faulty.
If HDD would be faulty, you should see the Toshiba Splash Screen following by the message like PXE error or something like Bootmgr is missing, etc
But you should definitely see something on the screen
If this is not the case, an motherboard or an part on the motherboard like memory modules might be faulty.
You could try to test the notebook using new tested RAM modules but I think the best for you would be a short trip to an technician downtown who could run some hardware tests.

ORACLE_HOME direcory of instance

Hello!
How to now ORACLE_HOME direcory of started instance on Linux ?
I'm usung 11.2.
Thanks and regards,
Pavel

Pavel wrote:
Hello!
How to now ORACLE_HOME direcory of started instance on Linux ?
I'm usung 11.2.
Thanks and regards,
PavelCheck this link:
http://tardate.blogspot.com/2007/05/getting-environment-variables-on-oracle.html

1 server, multiple oracle_homes for each instance (dev, test, uat, etc)...

Hello,
We have one beefy server to hold our dev and test and maybe more non-production databases. My plan was to install separate oracle_homes for each so that I can patch and only affect one instance at a time. Is anybody else doing this? Does this sound right? Any gotchas that you have run into with this type of environment?
I read something about you should only have one active listener...
Same concept, but with the ASM instance. Can I also install multiple oracle_homes for each asm instance on the same box? Is anybody else doing this? Again any gotchas people have run into to? For ASM, should I also have only one active listener?
Am I right that the only thing that affects all instances is at the OS, hardware level such as Linux patches, anything to do with users, groups, shared folders, nics, etc?
Thank you for your input!

For ASM, should I also have only one active listener?ASM does NOT require any listener & none should exist for ASM.

How does Azure Compute Emulator (or the Azure one) determine if a role is web project or something else ("The Web Role in question doesn't seem to be a web application type project")?

I'm not sure if this is F# specific or something else, but what could cause the following error message when trying to debug locally an Azure cloud service:
The Web Role in question doesn't seem to be a web application type project.
I added an empty F# web api Project to a solution (which adds Global.asax etc., I added an OWIN startup class Startup etc.) and then from an existing
cloud service project I picked Roles and
chose Add
-> Web Role Project in solution, which finds the F# web project (its project type guids are 349C5851-65DF-11DA-9384-00065B846F21 and F2A71F9B-5D33-465A-A702-920D77279786),
of which the first one seem to be exactly the GUID that defines a web application type.
However, when I try to start the cloud project locally, I get the aforementioned error message. I have a C# Web Role project that will start when I remove the F# project. I also have F# worker
role projects that start with the C# web role project if I remove this F# web role project. If I set the F# web project as a startup project,
it starts and runs as one would expect, normally.
Now, it makes me wonder if this is something with F# or could this error message appears in C# too, but I didn't find anything on Google. What kind of checks are there when starting the emulator and which one needs
failing to prompt the aforementioned message? Can anyone shed light into this?
Sudet ulvovat -- karavaani kulkee

Sudet,
Yeah you are right, the GUID mentioned seems to be correct and the first one i.e. {349C5851-65DF-11DA-9384-00065B846F21} means the web application project which compute emulator uses to determine while spawning up role instances.
You might want to compare the csproj of your C# and F# web projects which might give some pointers.
Are you able to run your F# web project locally in IIS? If yes then you will definitely be able to run it on azure so I will recommend to test it in IIS Express first.
Here are some other tips which you can refer or see If you are yet to do those settings
1. Turn on the IIS Express - You can do it by navigating to project properties
2. Install Dependent ASP.NET NuGets / Web Api dependencies (If there are any missing), Reference System.Web assembly
Also I will suggest to refer this nice article about how to create a F# web Api project
http://blog.ploeh.dk/2013/08/23/how-to-create-a-pure-f-aspnet-web-api-project/
Hope this helps you.
Bhushan | http://www.passionatetechie.blogspot.com | http://twitter.com/BhushanGawale

Indesign CS6 Server: Multiple instance configured, just one random instance come up?!

Hello everyone,
we have a problem with our Indesign server. It is configured to run as service on a Windows 2008 R2. We configured one instance with the management snap - that worked well. So we decided to add more instances and now the problems start: If I start the indesign server service all configured instances come up as process. But only a random one comes up able to process jobs. The other instances are dead - they don't even open a port. And I could see that the dead instances take hardly memory (ca. 50MB). The working instance takes ca. 260MB RAM directly after launching the service. What we are doing wrong?
As you can see the instance with the PID 7832 takes a lot more RAM - thats the working instance. I checked with "netstat -an" for the ports 4711-4714 - just port 4713 was open.
Thanks for your help and Regards from Germany
Benjamin Gannott

Hello everyone,
we have a problem with our Indesign server. It is configured to run as service on a Windows 2008 R2. We configured one instance with the management snap - that worked well. So we decided to add more instances and now the problems start: If I start the indesign server service all configured instances come up as process. But only a random one comes up able to process jobs. The other instances are dead - they don't even open a port. And I could see that the dead instances take hardly memory (ca. 50MB). The working instance takes ca. 260MB RAM directly after launching the service. What we are doing wrong?
As you can see the instance with the PID 7832 takes a lot more RAM - thats the working instance. I checked with "netstat -an" for the ports 4711-4714 - just port 4713 was open.
Thanks for your help and Regards from Germany
Benjamin Gannott

Filter using a variable instance by Papi

Hi I'm implementing a filter to get the instances in determinated activity by some variable instance attributes. Now I'm receiving all the instances of my activity and save them with a loop that validates by a variable instance, but I want to establish a property of the filter before my method receives all the instances. How can I achieve this?
+public List<Object> getInstanciasByIdFraccionamiento(Long idFraccionamiento){+
List<Object> listaObjetos = new ArrayList<Object>();
System.out.println("inicio");
+if (sesion == null) {+
+try {+
acceso = new AccesoPAPI();
acceso.setUp();
processService = ProcessService.create(acceso.getConfiguracion());
sesion = processService.createSession(acceso.getUserBPM(), acceso.getPasswordBPM(), null);
String idProceso = "/Promociones";
String idActividad = "AdministrarPromociones";
String tipoDato = "informacionPromocion";
Filter filtro;*
InstanceInfoList instancias;*
ProcessIdSet idProcesos;*
+//Establecer filtro de busqueda+
filtro = ProcessService.createFilter();*
SearchScope searchScope = new SearchScope(fuego.papi.ParticipantScope.ALL,fuego.papi.StatusScope.ALL, null);*
searchScope.setActivityScope(idActividad);*
searchScope.setStatusScope(StatusScope.ONLY_INPROCESS);*
filtro.setSearchScope(searchScope);*
filtro.setMatchAll(true);*
+// Busca instancias por actividad+
idProcesos = ProcessIdSet.create();*
idProcesos.add(idProceso);*
instancias = sesion.processesGetInstancesByFilter(idProcesos, filtro);*
+for (InstanceInfo informacionInstancia : instancias.toArray()) {+
informacionInstancia.getId();
DynamicObject objeto = (DynamicObject)informacionInstancia.getVariable("idFraccionamiento");
+if(objeto.longValue() == idFraccionamiento){+
System.out.println("long: " objeto.longValue());+
System.out.println("id is: " informacionInstancia.getId());+
System.out.println("nombre actividad es : " informacionInstancia.getActivityName());+
System.out.println("XML del Objeto BPM: " informacionInstancia.getVariable(tipoDato));+
+//Obtiene objeto BPM+
listaObjetos.add(informacionInstancia.getVariable(tipoDato));
+}+
+}+
sesion.close();
processService.close();
+}catch (AuthenticationException e) {+
System.out.println("No se pudo autenticar al participante " e.getParticipantId());+
e.printStackTrace();
+} catch (CommunicationException e) {+
System.out.println("No se pudo conectar al Servicio de Directorio");
e.printStackTrace();
+} catch (OperationException e) {+
System.out.println("No se pudo completar la operacion requerida");
e.printStackTrace();
+}catch (BatchOperationException e) {+
System.out.println("Could not perform the batch operation");
e.printStackTrace();
+}+

Just for context, here is what I'm trying to do:
I have jpegs in my library and I want to add them to the stage when they're needed, so just to add one image, here is the code I have:
var pic_1=new pic1(0,0);
var image_1:Bitmap=new Bitmap(pic_1);
material_mc.addChild(image_1);
I want to put the above into a loop so that I dont have to repeat those three lines for every image in my library like so:
var pic_1=new pic1(0,0);
var image_1:Bitmap=new Bitmap(pic_1);
var pic_2=new pic2(0,0);
var image_2:Bitmap=new Bitmap(pic_2);
var pic_3=new pic3(0,0);
var image_3:Bitmap=new Bitmap(pic_3);
var pic_4=new pic4(0,0);
var image_4:Bitmap=new Bitmap(pic_4);
var pic_5=new pic5(0,0);
var image_5:Bitmap=new Bitmap(pic_5);
var pic_6=new pic6(0,0);
var image_6:Bitmap=new Bitmap(pic_6);
var pic_7=new pic7(0,0);
var image_7:Bitmap=new Bitmap(pic_7);
the variable "picNum" is the total amount of images that in the library, each one exported as "pic1", "pic2", "pic3" respectively.
var picNum:uint=7;
var picCaller:uint=1;
var picMC:MovieClip = new MovieClip();
picMC=this["pic_"+picCaller];
for (var i:int = 1; i <= picNum; i=i+1)
var "pic_"+i = new image_i(0, 0);
var image:Bitmap = new Bitmap("pic_"+i);
Thanks so much for your help.

Dead-code elimination--what are all the rules?

This article http://www-128.ibm.com/developerworks/library/j-jtp12214/ has an entire section on dead-code elimination (DCE).
Does anyone know what all the criteria are by which the hotspot JIT compiler can determine that code is "dead"?
One obvious criteria is if the code will never get executed. I think that the canonical example is code like this:
       static final boolean debug = false;
       static void someMethod() {
               if (debug) {
       }Here, the compiler (javac as well as hotspot's JIT) can determine that someMethod has nothing inside it that could ever get executed, so it can be turned into an empty method.
If you look at that developerworks link above, he gives an even more interesting example for how hotspot's JIT can make sophisticated DCE based on inlining.
What I want to know is what are all the criteria that hotspot can use to decide that code is dead. Code that is never executed is always correct. But can it make more liberal decisions as to what counts as dead code?
Consider this code:
     int[] foo = new int[1000];
     int bar = foo.length;
     // don't use foo again, but do use barA smart compiler could recognize that the array declaration and allocation is actually useless, and simply do
     int bar = 1000;(Thank you Brian for this example.)

What are the rules? Naturally, the only definitive answer can be given can be given by the HotSpot JIT.
That's not a very satifying answer, but it's true. The server JIT, like most compilers, has been taught hundreds and hundreds simplifying transformations that, in part, enable dead code elimination. Enumerating the criteria is a nearly impossible task.
But, your specific example is an interesting case.
int[] foo = new int[1000];
int bar = foo.length;For simplicity, I assume foo and bar are locals to a method, and in this case, the 6.0 HotSpot Server JIT will recognize that foo is unchanged before the initialization of bar, and effectively change it to:
int[] foo = new int[1000];
int bar = 1000;The JIT can't get rid of the allocation of the new array without a more advanced optimization called Escape Analysis, which attempts to determine the dynamic scope of objects. Of course, if these two lines comprise the whole method, then the analysis is fairly trivial.
Any intervening code will make the analysis more complex:
int[] foo = new int[1000];
baz(foo);
int bar = foo.length;
qux(foo);The calls sites will preclude your desired optimization, unless the called methods baz and qux can be inlined and analyzed. The code in baz can neither modify foo nor allow it to escape, and qux can not allow it to escape.
HotSpot's implementation of escape analysis is not quite yet complete, but should be available relatively soon.

Determining ORACLE_HOME of a dead Instance

Similar Messages

Maybe you are looking for