Glassfish password problem - can 'configure' but not 'admin' - ORDS, SODA & JSON in the Database

Hi,
Using the release 1.10.179 with Glassfish Community 3.0.1 on Windows XP - I can deploy and configure my APEX listener, but once it's there I cannot get to the 'listenerAdmin', 'listenerStatus' screens - it doesn't accept the passwords that I created for the 'adminlistener' and 'managelistener' users.
Any ideas on how to get past this? 

I installed this successfully in Windows 7, just tonight.
I would suggest first double checking the PDF instructions for the configuration - I found it easy to miss a small step.
Double check your port numbers you're using, too. Mine defaulted to 4848 for the admin, but 8080 for listener admin.
Scott 

scott.wesley wrote:
I installed this successfully in Windows 7, just tonight.
I would suggest first double checking the PDF instructions for the configuration - I found it easy to miss a small step.
Double check your port numbers you're using, too. Mine defaulted to 4848 for the admin, but 8080 for listener admin.I don't think that it's a ports problem as 'wrong' ports result in getting a page not found error. I've also found that I'm having problems running APEX, seems to be on the wwv_flow.accept...
I'll have another look at the documentation! 

Hi,
check in apex-conf.xml (or during configuration in Security->Allowed procedures )
in Inclusion List you should have:
wwv_flow*, apex*, f ...
Best Regards,
Andrzej 

I just double checked then, I visited
http://localhost:4848/apex40/listenerConfigure/
And used "admin" with the password I set when installing glassfish
Although this time when I logged in, I got
/apex40/listenerConfigure//index.jsf not found
Helpful, but not helpful.
Scott 

Try without the trailing slash.
-kris 

Without trying to commandeer the post, yes, I tried that after my lesson with needing the slash for the standard URL.
Instead of seeing the login page, I get 404
The requested resource () is not available
I also tried manually amending the path it can't find, to remove the duplicate slash, even tried putting domain1 or i in there to see what happened. 

Scott,
Here's what I think. Glassfish's internal admin is on port 4848 and that is all that is on that port. The normal webserver is on 8080 by default. So you should try
http://localhost:8080/apex40/listenerConfigure
-kris 

Kris,
I tried those permutations,
This gets me
http://localhost:8080/apex40/listenerConfigure
The APEX Listener is already configured.  Please login as Administrator to access APEX Listener Administration.
Hence why I tried the other port.
This gets me to the Apex Builder login.
http://localhost:8080/apex40/listenerConfigure/f?p=4550:1:4103662282177325
Scott 

This is a one-only-use url : http://localhost:8080/apex40/listenerConfigure
Once the listener is configured this url is disabled and you have to use
http://localhost:8080/apex40/listenerAdmin
-kris 

araczkowski wrote:
Hi,
check in apex-conf.xml (or during configuration in Security->Allowed procedures )
in Inclusion List you should have:
wwv_flow*, apex*, f ...
Best Regards,
AndrzejThe APEX side seemed to settle down after a reboot, that works fine now.
I still can't get into the listenerAdmin... it doesn't like my passwords! 

You can change the passwords in the apex-config.xml. It's in the install/config doc how to do this.
-kris 

Kris Rice wrote:
You can change the passwords in the apex-config.xml. It's in the install/config doc how to do this.Glassfish doesn't use the apex-config.xml file.
However, once I reset the passwords for the users from with the Glassfish admin' pages, I regained my access to the 'admin' page.

Related

Getting ORA-06508 in APEX Listener Configuration

Hi,
I'm trying to configure the Apex Listener. After entering all required fields and clicking on "Apply" I'm getting the error messages listed below. This happens when trying both Basic and TNS Connection types.
ORA-06508: PL/SQL: could not find program unit being called
Unable to run page sentry in application 4500.
Thanks in advance for your help.
Edited by: user3654649 on Jun 14, 2011 11:32 AM
Edited by: user3654649 on Jun 14, 2011 11:33 AM 
Hi,
it seems the APEX Listener can connect to your database, so the error must be somewhere in your APEX configuration.
Could you please provide some additional information on your setup, e.g.
- Which database user did you configure your APEX Listener to use? Usually you would choose APEX_PUBLIC_USER.
- Which database version are you using?
- Which APEX version does your database have?
- Did you use some other web server before using APEX Listener? (Embedded PL/SQL Gateway or Oracle HTTP Server)
-Udo 
Hi,
Thanks for your answer.
The database user that I'm truing to configure the listener to use is APEX_PUBLIC_USER. the database version is 10.2.0.4. The APEX version is 4.0.2.00.07 and for this database I did not use Embedded PL/SQL Gateway or Oracle HTTP Server.
thanks. 
So, everything up to now seems to be correct. Now, second line questions:
- Did you encounter any problem in your APEX installation?
- Could you check if there is any invalid object in your APEX schema? (run apexvalidate.sql from your APEX installation package connected to that instance as sys)
- Which JEE container do you use?
- Did you deploy the images for APEX yet? And did you deploy it to the context you've configured as images prefix during your APEX installation?
-Udo 
I did not encounter errors in the APEX installation. I don't have any invalid objects in apex_040000 or flows_files schemas (I did select * from dba_objects where owner='APEX_040000' and status='INVALID').
I ran the apexvalidate.sql procedure and now when doing "select * from dba_registry;" I see APEX status as INVALID - however there are no invalid objects in apex_040000 or flows_files schemas.
When doing the installation I followed the Application Express installation guide Release 4 - specifically "Scenario 1: Downloading from OTN and Configuring the Oracle Application Express
Listener" and I ran the installation script as "#apexins APEXTBS APEXTBS TEMP /i/"
The images directory is under apex (apex/images) while apex is the root directory of the installation.
Is there anything that I should have done differently with the /i/ ? Should it be the physical location for the images (something like /oradata/apex/images) ? Another thing - I just noticed that APEX_040000 and FLOWS_FILES are both locked - can this have to do anything with the error I'm getting ?
Waht does it mean "configured as images prefix " ? Is there anything that should be done that is not listed in the installation guide ? Regarding the "JEE container " I don't know.
thanks. 
One more thing - When doing the configuration it prompts me to enter the images directory and to this I reply with the exact physical location.
Enter the path to the directory containing the APEX static resources
Example: /Users/myuser/apex/images
or press Enter to skip:
thanks. 
Hello Sue,
I ran the apexvalidate.sql procedure and now when doing "select * from dba_registry;" I see APEX status as INVALID - however there are no invalid objects in apex_040000 or flows_files schemas.I'm sorry, I forgot to mention that you have to add some parameters to the apexvalidate-Call. In the APEX 4.0.x-case, this would be
#apexvalidate.sql x x APEX_040000I guess this will either bring up errors or put the APEX entry in the dba_registry back to VALID.
In addition to checking for invalid objects and trying to recompile/... them in case there are hits, the validation routine also checks if all vital objects are really there. I'd expect some error should have been raised if you didn't call the procedure as noted above.
I ran the installation script as "#apexins APEXTBS APEXTBS TEMP /i/"That's correct. Are the tablespace APEXTBS and TEMP present and below their size limit?
Is there anything that I should have done differently with the /i/ ?No, that's the default value and fits perfectly in most cases.
Another thing - I just noticed that APEX_040000 and FLOWS_FILES are both locked - can this have to do anything with the error I'm getting ?No, that's the way it should be. These two schemas should never be active database users. When using APEX Listener, you use APEX_PUBLIC_USER to connect to the database.
Waht does it mean "configured as images prefix " ?+/i/+ is the images prefix in your case. As I see in your following post, you are running APEX Listener in standalone mode, so you don't have to care - it has unchangeable contexts +/apex/+ and +/i/+ which fits to your APEX installation.
To explain my question:
You can deploy the APEX Listener to a "full blown" application server (which I referred to as JEE-Container) instead of running in standalone mode.
In that case you have to provide a context containing your images that fits to that prefix.
- When doing the configuration it prompts me to enter the images directory and to this I reply with the exact physical location.That's where the context +/i/+ is mapped to in standalone mode. Everything's correct here as well.
So, after all, everything looks great. If there really hasn't been any error during the installation and the validation routine doesn't throw any error as well, I'd try a clean uninstall (drop FLOWS_FILES, APEX_040000 and APEX_PUBLIC_USER) and rerun the installation routine.
If that doesn't help, could you try to enable the EPG once just to check the APEX installation itself is fine? I really doubt this error is really related to the APEX Listener at the moment...
-Udo 
Hi Udo,
Thanks for your reply.
Running apexvalidate.sql with the suggested parameters did not produce any errors.
SYS#cwsrvsit> #apexvalidate.sql x x APEX_040000
Procedure created.
No errors.
SYS#cwsrvsit> SYS#cwsrvsit> SYS#cwsrvsit>
However the status is still invalid :
SYS#cwsrvsit> select comp_id,comp_name,status from dba_registry where comp_id='APEX';
COMP_ID COMP_NAME STATUS
------------------------------ ----------------------------------- ----------
APEX Oracle Application Express INVALID
Should I try a new installation ? Is it possible that something in the Java insrtallation is wrong ?
thanks again,
Sue 
Hi Sue,
I'm sorry, I just didn't have my APEX troubleshooting notes at hand when posting the previous comments. Now I found them I see I missed some other steps to restart the validation, but I tested the following to work on my testing environment:
set define '^'
set concat on
set concat .
set verify off
#apexvalidate x x APEX_040000
exec validate_apex;Result:
SQL> exec validate_apex
...Database user "SYS", database schema "APEX_040000", user# "46" 23:39:36
...Compiled 0 out of 2721 objects considered, 0 failed compilation 23:39:36
...234 packages
...227 package bodies
...426 tables
...12 functions
...19 procedures
...3 sequences
...439 triggers
...1177 indexes
...176 views
...0 libraries
...4 types
...0 type bodies
...0 operators
...0 index types
...Begin key object existence check 23:39:36
...Completed key object existence check 23:39:36
...Setting DBMS Registry 23:39:36
...Setting DBMS Registry Complete 23:39:36
...Exiting validate 23:39:36
PL/SQL procedure successfully completed.At the end of that block you see that the dbms_registry is set according to the result. So if your query
select comp_id,comp_name,status from dba_registry where comp_id='APEX';still returns "INVALID", I'd recommend to redo the installation.
Is it possible that something in the Java insrtallation is wrong ?I don't think so, as your APEX Listener is able to connect to the database, but the database is not able to produce the APEX workspace login page.
Sorry once more for not providing the correct troubleshooting script at once.
-Udo 
Hi Udo,
Thanks for your reply - it was very helpful. I dropped the AOPEX environment and did a reinstall - now I'm able to do the listener configuration without issues. My next question is - is there a way to do it in the background ? When doing it as "java [options] -jar apex.war" it depends on my session being connected. If I'm doing a CTRL+C the listener goes down. Is there a way to do the configuration in the background and keep it up and running ?
thanks again,
Sue 
Hi Sue,
When doing it as "java [options] -jar apex.war" it depends on my session being connected. If I'm doing a CTRL+C the listener goes down. Is there a way to do the configuration in the background and keep it up and running ?That's actually two different problems of the APEX Listener in standalone mode: One concerning the configuration to be persistent and one to let the APEX Listener run in background.
You can solve both problems by using a "full blown" application server to deploy the APEX Listener. But you can also do this in standalone mode.
Lets start with easier one - the configuration:
The APEX Listener stores its configuration in the "apex.home". By default, this is located in your TEMP-directory. So, depending on your environment, it might be gone after you logged out. You can configure the "apex.home" as a startup parameter and let it point to some directory that is persistent. You could just configure your APEX Listener once, shut it down, and copy the apex-directory to that directory and you'll never have to configure it again.
java -Dapex.home=<Path_to_your_non-temporary_directory> -jar apex.warThe second one might be a little more difficult, depending on your operating system.
If you use a Windows system, you'll find a solution description in the following message: {message:id=9606494}
On Linux systems, a simple solution might be to use a "screen" session. That doesn't really start the APEX Listener as system service, but lives as long as the screen session lives in the background and you can easily do this with a less privleged user. For a neat solution that even start/stop cleanly with your OS, I need to know which one you run your APEX Listener.
-Udo 
Hi Udo,
Thank you very much for your help - it worked.
After doing :
java -Dapex.home=/orabackup_iris2/export/apex_listener/home -jar apex.war and answereing all the prompts,
I did CTRL+C to stop the listener and then I copied the content (cp -R) of /orabackup_iris2/export/apex_listener/home to /orabackup_iris2/export/apex_listener
and the I ran :
nohup java -Dapex.port=8080 -Dapex.home=/orabackup_iris2/export/apex_listener -Dapex.images=/orabackup_iris2/export/apex/images -jar /orabackup_iris2/export/apex_listener/apex.war &
If you can please go over the steps above and see if it makes sense - I hope it's not just pure luck...
thanks again,
Sue 
Hi Sue,
it's not just luck, it definetly works that way.
Just some final remarks:
- There's no need to change/copy the contents from +/orabackup_iris2/export/apex_listener/home+ to +/orabackup_iris2/export/apex_listener+ in your case, as the first directory doesn't seem to be a temporary one.
- I'd recommend to put the java-call including parameters into a shell script and call that instead.
- nohup works fine for on-demand background processes. As you probably know it's not providing a "system service"-like functionality.
- nohup creates a default file for console output of the job it started. I'd recommend to redirect it to a named file and append all output.
So, your final setup would look as follows
start_apex_listener.sh
#!/bin/sh
java -Dapex.port=8080 -Dapex.home=/orabackup_iris2/export/apex_listener -Dapex.images=/orabackup_iris2/export/apex/images -jar /orabackup_iris2/export/apex_listener/apex.warnohup call
nohup /orabackup_iris2/export/apex_listener/start_apex_listener.sh >> /orabackup_iris2/export/apex_listener/apex_listener.log 2>&1 &If your question is answered, please mark this thread and any helpful or correct answer post accordingly.
-Udo 
Udo,
Thank you very much for all your help - my question is definiteley answer
All the best,
Sue
Edited by: Sue on Jun 19, 2011 6:25 AM 
Hi Sue,
I'm glad I could help you.
If you don't mind, you can mark "helpful" and "correct" answers in this thread. Especially in longer threads, this makes it easier for other people to find relevant posts.
Thanks,
Udo

Deploytool

Hello,
For some reason, when I click on my "localhost" server in my deploy tool, it's not accepting my admin username/password.
Is there somewhere that I configure the Deploytool username/password for my localhost server in the Admin Tool?
What am I doing wrong? 
Hi Buck,
Which version are you using? If you are using the reference implementation 1.3x then you shouldn't have that problem, but there is a menu option TOOLS => SERVER CONFIGURATION where you can add a user, then set the default user to yourself. But this is version specific of course, which brings me back to, what version are you using?
See ya
Michael 
Thanks for the reply Michael.
I'm using Sun Appserver PE 8. I tried seeing if it had the options that you mentioned exists in other versions of appserver, but it does not seem to exist in version 8.
Try your user name & password in the browser hosted admin interface (http://localhost:4848/asadmin/index.html - adjust name & port to match your install, if necessary) to ensure you are using the correct ones.
The default username is admin, default password is adminadmin. Other than that, I think you are doing it correctly. 
Peter,
I tried my the admin/password combination I established at install and successfully used to enter the admin tool. That didn't work.
I also tried the default username/password you mentioned in your post, and that still didn't work either.
I'm baffled as to where in the installation process or otherwise that username/password is configured. Where can I go to edit/view/change the https username/password? 
Hi,
Check this page http://java.sun.com/j2ee/1.4/docs/relnotes/troubleshooting.html#1026910
One question before last.
Hope this helps. 
I tried my the admin/password combination I
established at install and successfully used to enter
the admin tool. That didn't work.I'm confused. By this, do you mean
(a) "I tried to login to the admin tool with my user name and password and it didn't work" -- e.g. you failed to login to the admin tool as well as deploytool.
or
(b) "I tried to login to the admin tool with my username and password and it was successful" -- e.g. you were successful in logging into the admin tool, but you still fail to login via deploy tool.
To further investigate and possibly reset your admin password, see the section of the troubleshooting document on this: http://docs.sun.com/source/817-6085/troubleshooting.html#invalidPassword
You should also read the portion of the document that is slightly above that link as it may also relate to this issue. 
Hi Buck,
I found a potential problem. The default port for the admin area when installing is 4848, I changed it to 8000, personal preference. When I tried to login with the deploytool I got the login prompt with the default port, and subsequently couldn't find the server of course, resulting in a password/userid error.
Solution: Change the port number that you are logging in to.
Instructions: Highlight the server in the left panel, then FILE => REMOVE SERVER . Then FILE => ADD SERVER, and put in the details that you used for the install port and userid/password combo, and violla.
I hope this is the problem that you were having. Otherwise I suggest a clean install, leave the ports and everything else default.
See ya
Michael 
I'm confused. By this, do you mean
(a) "I tried to login to the admin tool with my user
name and password and it didn't work" -- e.g. you
failed to login to the admin tool as well as
deploytool.
or
(b) "I tried to login to the admin tool with my
username and password and it was successful" -- e.g.
you were successful in logging into the admin tool,
but you still fail to login via deploy tool.I said that I try to log into the "localhost" server configured in my deploytool. It then asks me for a username and password. My admin/password combination, which works successfully when logging into the "admin tool", does not work when logging into the deploytool's localhost server on the https port I configured at installation.
Michael,
I found a potential problem. The default port for the
admin area when installing is 4848, I changed it to
8000, personal preference. When I tried to login with
the deploytool I got the login prompt with the default
port, and subsequently couldn't find the server of
course, resulting in a password/userid error.That was the problem. In my 3rd post, buried in the last sentence, I ask where I can edit my https port's username/password. I was trying to access the localhost with the deploytool through my https port, instead of my admin port.
That was my problem the whole time. Thanks Michael and everyone else who contributed to this thread. I've really learned a lot from my mistake. 
No worries Buck, happy that you are sotred out.
So a little humor wouldn't go amiss:
A project manager, a computer programmer and a computer operator are driving down the road when the
car they are in gets a flat tire. The three men try to solve the problem.
The project manager said: "Let's catch a cab and in ten minutes we'll reach our destination."
The computer programmer said: "We have here the driver's guide. I can easily replace the flat tire and
continue our drive."
The computer operator said: "First of all, let's turn off the engine and turn it on again. Maybe it will fix the
problem."
Suddenly a Microsoft software engineer passed by and said: "Try to close all windows, get off the car, and
then get in and try again."See ya
Michael

clusters and logins and cookies oh my

This note is specifically about version 8.2 of the appserver and the 7.1 access manager in the JES5 update 1 release. All on Solaris 10 Sparc.
I've seen enough to be pretty sure that each version will have its own quirks. Here's my list of discovered magic... probably not complete but useful to know.
The only guide to appserver installation on a cluster is a tech note for the 2005Q4 version. It contains an error relating to realm mode (server ports have to MATCH)
You must not try to do the AM cluster installation on appservers that have been turned into a service.
Firstly - If you want to install on a cluster you have to use amconfig and the silent config file. No way to do it with "configure now"
Second - If you install an appserver with this installer it WILL install the node agent even though it doesn't say it is going to and you didn't tick the box. It just won't tell you about it and it will NOT configure it if you've told it to configure now.
Third - You can't put in a node-agent name like hostname.mydomain.net.ac in the gui installer if you did tick that box. You CAN configure it that way if you use the CLI to create the node agent after the fact. ( Note that I am not certain this is the correct way to create a node agent name but it is part of a way that worked)
Fourth - If when you've installed your access manager your login of amAdmin is rejected right away. Try the login using "uid=amAdmin, ou=people,ds=mydomain,ds=net,ds=ac" or your local equivalent. If that WORKS you may well get returned to the login prompt, but without the rejection message. Check your access logs and the directory service.
If that was the case, go to your etc/hosts file and force it to have the fully RFC compliant version of an entry for your host. That is - "192.168.xxx.xxx hostname.mydomain.net.ac hostname " ,,, (other aliases as desired)
Then reinstall. (Including a wipe of the directory down to bare metal)
Fifth - The redirection back to the login prompt can still happen and this is related to cookies as well. ( One of the clues is that the login for the AM offers up the certificate error message that the certificates named hostname and hostname.mydomain.net.ac don't match. )
Check your cookie domain to be starting with the '.' in the silent config file as in ".mydomain.net.ac" instead of "mydomain.net.ac" .
Check your browser cookie store for a "net.ac" entry which contains cookies meant for mydomain.net.ac. Find and delete them.
Install node agents with the fqdn version of a node name hostname.mydomain.net.ac ( ? not sure of this one )
Reinstall your AM. ( wipe the directory if you do )
After this it works... (well, so FAR it works)....
I hope this helps someone. I certainly could've used it a while back.
Oh yeah... beware of the cluster management in the appserver. If you tell it to start it will go to "starting" but it will display "stopped". Temptation to stop and start again is large.. it takes quite a long time for me ( 3-5 minutes) and if you stop and start a few times it can get quite confusing.
respectfully
BJ 
If the note four login attempt actually works and you get in you may wish to refer to this:
http://forum.java.sun.com/thread.jspa?threadID=5204436&tstart=45
As a way of fixing the problem with the domain.
respectfully
BJ 
Ouch... my bad...
Try the login using "uid=amAdmin, ou=people,dc=mydomain,dc=net,dc=ac"
Hurry up and err...
respectfully
BJ

Using password.conf

Hello,
I'm running App Server 7 on a V240 with a Crypto Accelerator 500 installed. I've created two security domains on my CA500, thus when I start my AS admin server (which has encryption enabled), I am prompted for three passwords:
139# asadmin start-appserv
password:admin-server:internal>
password:admin-server:Sun Crypto Accelerator:admin#qisc.com:admin#qisc.com>
password:admin-server:Sun Crypto Accelerator:nobody#qisc.com:nobody#qisc.com>
Instance domain1:admin-server startedIf I manually enter these three passwords, the server starts up and everything works as expected.
Following the instructions in the manuals, I have created a "password.conf" file in the admin server's config directory (/var/opt/SUNWappserver7/domains/domain1/admin-server/config) so that I won't be prompted for passwords and the server can be restarted in unattended mode.
The contents of my password.conf file looks something like this:
internal:myPassword
admin#qisc.com:myPassword
nobody#qisc.com:myPasswordHowever, when I start my admin server, I am still prompted for passwords.
What am I missing?
Thanks,
Bill 
On the face of it, you don't seem to be missing anything. It should have worked.
Can you please try starting the server using the startserv script in /var/opt/SUNWappserver7/domains/domain1/admin-server/bin directory?
Does it also prompt for passwords? If yes, what are the tokens that it prompts for? 
Hmmm...I tried that, it started without prompting me, and it is working as it should. So what do I have to do to get asadmin to use the password.conf?
Thanks,
Bill 
Try using something like this...
create a file autostart with excecute permission
#!/bin/sh
<path to asadmin>/asadmin startserv<<END
password1
password2
password3
END
I have not tried this one on my server. But used start the secured servers in a similar fashion. Let me know how this goes.
Srivatsan. 
Srivatsan,
Unfortunately, it appears as though the passwords are being read from the tty directly, not stdin, so this doesn't work.
However, the fact that my password.conf file appears to be correct (by virtue of the fact that invoking the admin-server/bin/startserv script directly starts the admin server w/o prompting for a password) begs the question is this a bug?
Ah, in fact it is a bug (I searched sunsolve, and found bug 4926581). The bug readme says it is fixed, but I find no reference to this bugid in any of the AS7 patches found on SunSolve (118031-118036 which were just released on Thursday, 10/7). Does this mean I need to wait until the next Orion release to get this fix?
Thanks,
Bill 
I think it is fixed in Sun Java� System Application Server 7 2004Q2 Update 1
Cheers,
VM 
Yes, I can confirm that it is working with Update1.
But the file must be in the config directory of the instance, not of the admin-server!
Which makes sense because you could have multiple instances, with multiple truststore databases and therefore multiple passwords.
Best regards

SEVERE: Need a console to configure listener credentials

We have been using APEX 4.1 since it cam out and the latest patch set for several weeks now. Over the weekend, we had an unplanned power outage and now, the Apex Listener can not be restarted.
Attempts to restart the listener result in the following error stack:
INFO: Starting: /u01/oracle/product/apex_listener/apex.war
See: 'java -jar apex.war --help' for full range of configuration options
INFO: Extracting to: /tmp/apex
INFO: Using classpath: file:/tmp/apex/apex/____embedded/start.jar:file:/tmp/apex/apex/WEB-INF/lib/je-4.0.103.jar:file:/tmp/apex/apex/WEB-INF/lib/poi-3.6-20091214.jar:file:/tmp/apex/apex/WEB-INF/lib/xdb-11.2.0.jar:file:/tmp/apex/apex/WEB-INF/lib/commons-fileupload-1.2.1.jar:file:/tmp/apex/apex/WEB-INF/lib/apex.jar:file:/tmp/apex/apex/WEB-INF/lib/xmlparserv2-11.2.0.jar:file:/tmp/apex/apex/WEB-INF/lib/ojmisc.jar:file:/tmp/apex/apex/WEB-INF/lib/ojdbc6.jar:file:/tmp/apex/apex/WEB-INF/lib/ucp.jar:
INFO: Starting Embedded Web Container in: /tmp/apex
SEVERE: Need a console to configure listener credentials
java.lang.IllegalStateException: Need a console to configure listener credentials
at ____bootstrap.____Bootstrap._start(____Bootstrap.java:29)
at ____bootstrap.____Bootstrap.start(____Bootstrap.java:163)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at ____embedded.____EntryPoint.invoke(____EntryPoint.java:299)
at ____embedded.____EntryPoint.main(____EntryPoint.java:95)
Anyone have any ideax of something we could try?
-Joe 
Hi Joe,
it seems you are using the APEX Listener in Standalone Mode.
It would have been helpful to know how you started your APEX Listener. Probably you didn't configure the apex.home, so your configuration (apex-config.xml as well as your adminlistener/managerlistener credentials and the information where your static contents (images) resides) would have been stored in +/tmp/apex+ as well. Depending on your OS, it's likely that +/tmp+ will be "recycled" upon every reboot, so it's lost by now.
My next guess is that you use some kind of automatic startup with nohup or some similar tool. This doesn't offer the console needed to enter credentials and images location.
You can solve this issue by starting APEX Listener directly, i.e. using a regular terminal instead of some background service, so you'll be able to rerun the initial configuration.
Please note that running APEX Listener in Standalone Mode is not recommended for a production deployment. If you can't do anything about that right now, I'd at least recommend to include apex.home in your startup and let it point to a permanent file system location, e.g.
java -Dapex.home=/u01/oracle/product/apex_listener/conf -jar apex.warThere is also a parameter apex.images to provide the path to the images instead of using the value from first startup. See the APEX Listener Installation and Developer Guide for details and further options for deploying APEX Listener.
-Udo 
Thanks, Udo. What you have pointed to, the location of the config files on the /tmp directory was, for sure, the issue.
I have managed to reconfig the listener and now have it back up again. However, I would like to stop it form using the /tmp directory to prevent this from happening in the future.
I opened the apex.war file and editing the web.xml file such that I uncommented the section, config.dir and changed the value to a permanent location. I restarted apex but, the change doesn't seem to have any affect. That is, it is still using the /tmp/apex directory.
Does your command do the same thing, albeit form the command liine?
Do I have to call it like that everytime or will calling it once like that change the location from that point forward?
Thanks,
-Joe 
Hi Joe,
the config.dir from web.xml is just for the apex-config.xml , not for user credentials. I'm not sure if it is evaluated at all when running in Standalone Mode, perhaps I'll run a test case to be sure.
Anyway, you need to set that startup parameter, as mentioned before. Obviously, you need to do that upon every startup, as there is no way to know where the home would be without having any configuration data. However, once you've finished the initial configuration, you just need to pass the parameter for the home, you don't need to re-enter the configuration data if that home has been configured before.
-Udo 
Udo,
Not sure that I understand...it seemed like you were saying that I did need to pass the parameter every time but, then, later you said "you don't need to re-enter the configuration data if that home has been configured before"
Let me ask a more specific question to make sure I am doing this correctly.
Right now, this is the command line with which I start up the stand alone listener: /usr/java/jdk1.6.0_27/bin/java -jar /u01/oracle/product/apex_listener/apex.war
You are saying that I need to use: /usr/java/jdk1.6.0_27/bin/java -Dapex.home=/u01/oracle/product/apex_listener/conf -jar /u01/oracle/product/apex_listener/apex.war
Correct?
My question is, do I need to pass in the Dapex.home argument only when I initially reconfigure. Or do I need to permanently issue the command with the "Dapex.home" switch included?
Thanks for clarifying,
-Joe 
Hi Joe,
/usr/java/jdk1.6.0_27/bin/java -Dapex.home=/u01/oracle/product/apex_listener/conf -jar /u01/oracle/product/apex_listener/apex.warRight. You need to pass that parameter every time you start your APEX Listener, because, as I intended to clarify before, the application wouldn't have any chance to know where it's home would be if you didn't pass that parameter.
What I meant to say is that you need to enter the configuration data (images, usernames and passwords) just upon initializing that home directory, i.e. upon the first time you call APEX Listener using that path. Later, you can run APEX Listener in background (referencing a proper home directory).
-Udo
P.S.: If your question is answered, please mark this thread and any helpful or correct answer accordingly to help other users searching for a similar issue to spot relevant posts easily. Thanks!
Edited by: Udo on 02.04.2012 20:12 
Thanks, Udo.
All is working as desired now.
-Joe

Categories

Resources