1. Deploying JD Edwards EnterpriseOne on Linux Learning Path
  2. Troubleshooting Your One-Click Deployment

Troubleshooting Your One-Click Deployment

This section shows you the list of troubleshooting topics and resolutions for your One-Click deployment.

One-Click Provisioning Console Troubleshooting Tips

This section contains the following troubleshooting tips for the EnterpriseOne One-Click Provisioning Console:

  • Accessing a Previously Saved Provisioning Console Configuration
  • Console Administration
  • Console Service Status
  • Start the Console
  • Stop the Console
  • View the Console Logs
Accessing a Previously Saved Provisioning Console Configuration
Symptom: You cannot access a previously created Quick Start deployment plan.
Cause: After you created a Quick Start deployment plan (either complete or partial), you exited the browser used to connect to the One-Click Provisioning Console.
Resolution: When you next access the Provisioning Console, you are prompted to enter the same credentials that you entered for the first access to the console. Those credentials are the password for the Server Manager for JD Edwards EnterpriseOne. After the credentials are validated, you can access the previously saved configuration.

Console Administration

The One-Click Provisioning Console process is configured as a systemd service in the One-Click Provisioning Server image. The configuration file (E1CloudConsole.service) is located in this directory:

C:\JDE\bin\E1CloudConsoleWin.ps1

Console Service Status

Access the Task Manager and verify that node.exe is running in the Provisioning Server machine:

Start the Console

Run the following command in the power shell to start the One-Click Provisioning Console service.

PS C:\JDE> E1CloudConsoleWin.ps1

Outside the machine, you can access the One-Click Provisioning Console application using the below URL:

https://<#PUBLIC_IP#>:3000/

Stop the Console

Access the Task Manager and terminate the node.exe process if it is running in the Provisioning Server machine.

View the Console Logs

See the following One-Click Provisioning Console log files in C:\JDE\PP\E1CloudConsole\logs:

E1ProvisionConsole.err

E1ProvisionConsole.out

The input validation logs files are located in this path: C:\JDE\PP\E1ProvisionPrime\InputValidation\log

Press Page-Down to scroll through the logs.

Press Shift + G to scroll to the end of the log.

Note: The One-Click Provisioning Console logs are completely regenerated each time the console is run.

Enterprise Server Provisioning Fails - Communication Failure

Symptom: Provisioning fails to deploy the Enterprise Server with an error returned indicating a communication failure.
Cause: A timeout may have occurred during communications between the One-Click Provisioning processes and the on-premises machine.
Resolution: Restart the deployment of One-Click Provisioning.

Enterprise Server Provisioning Fails - Health Check Fails

Symptom: PORTTEST runs remotely, but fails.
Cause: The kernels are not starting up fast enough to run a PORTTEST successfully possibly because of a timing issue.
Resolution: Rerun the deployment. It should quickly skip to the Ent Health Check and pass.

Enterprise Server Not Able to Run PORTTEST Manually

Symptom: Not able to run PORTTEST manually on a Microsoft Windows Enterprise Server.
Cause: The Service user for the Enterprise Server is not set to opc.
Resolution:

You must set the Windows Service user on the Enterprise Server Service User to opc using the procedure documented here:

JD Edwards EnterpriseOne Administration Guide

4. Administering the Windows Server

4.3 Working with Network Services

4..3.2 Setting Up the Network Service

https://docs.oracle.com/cd/E53430_01/EOTSW/windows_server.htm#EOTSW00222

HTML Server Intermittently Unavailable

Symptom: The HTML Server provisioned on the WebLogic Server is intermittently going down and the users are not being able to access the EnterpriseOne HTML login page. Generally, when this issue occurs, the below error is displayed in the WebLogic Server logs or in the HTML Server logs:

java.lang.OutOfMemoryError: PermGen space
Resolution:

To bring up the EnterpriseOne HTML Server when the above error happens, try one of the following steps:

  1. Restart the EnterpriseOne HTML Server instance from the Server Manager Console and then check if the EnterpriseOne HTML login page is accessible.
  2. If the JD Edwards EnterpriseOne HTML Server login page is not accessible even after restart, then reset the below JVM arguments in the HTML Server from the WebLogic Server console and then restart the HTML Server.
    • Log in to the WebLogic Server console.
    • Go to Servers on the Environment tab and then select the HTML Server.
    • Go to the Server Start tab and click the Lock & Edit button in the left upper corner.
    • In the Arguments section, change the values for the following arguments from their existing value to 1024m:

      -XX:PermSize=1024m

      -XX:MaxPermSize=1024m

    • In the Arguments section, go to the end and enter one line space and paste the below arguments:

      -XX:+CMSClassUnloadingEnabled

      -XX:+CMSPermGenSweepingEnabled

      -XX:+UseConcMarkSweepGC

    • Click the Save button.

    • Click the Activate Changes button in the left upper corner.

    • Restart the EnterpriseOne HTML Server.

  3. If the JD Edwards EnterpriseOne HTML Server login page is still not accessible then, free up some resources (OCPU and memory) from the cloud domain and restart the EnterpriseOne HTML Server from the Server Manager Console.

Server Manager Console Fails to View Log File and Download Log for Enterprise Server and HTML Server

Symptom: Server Manager Console fails to view the log file and download the log for Enterprise Server and HTML Server.
Resolution: You should always use HTTP to access the Server Manager Console for normal operations. Although the Server Manager Console is also enabled for the HTTPS or SSL protocol, such access is restricted for internal use as part of the REST/API functionality, and should not be used for normal operations.

PDB on Compute Not Open

The Console checks to ensure the Pluggable Database (PDB) on premises is open (set to READ WRITE) as you begin to provision it. If the PDB is not open, the Console scripts will issue commands to open it. However, if the PDB is still closed (not set to READ WRITE) when the deployment commences, the deployment will fail. Use the following procedure to determine if the PDB is open, and if not how to set it to open.

  1. Connect to the database using this command:

    sqlplus '/as sysdba';

  2. Verify the pluggable database JDEORCL is open using this command:

    select OPEN_MODE from v$pdbs where NAME = 'JDEORCL';

  3. If the OPEN_MODE is set to READ WRITE, then the PDB is ready to be provisioned by the JD Edwards EnterpriseOne One-Click Provisioning Server.

  4. If the OPEN_MODE is not READ WRITE, then the PDB is not open. Open the PDB using this command:

    alter pluggable database JDEORCL open;

Not Able to View and Download Server Manager Home Logs When Server Manager is SSL-Enabled

When the Server Manager is SSL-enabled, the users may not be able to view and download the Server Manager Home logs. The root cause may be because the proper security certificates are not available. For instructions on how to create and import security certificates, refer to these documents:

  • JD Edwards EnterpriseOne Tools Server Manager Guide
  • JD Edwards EnterpriseOne Tools Security Administration Guide

JD Edwards EnterpriseOne HTML Server Login Fails

If the JD Edwards EnterpriseOne HTML Server login fails, run PORTTEST on the JD Edwards EnterpriseOne Enterprise Server. If PORTTEST fails, check the host entries in the /etc/hosts file. See Edit the /etc/hosts File for Connectivity in this Learning Path.

JD Edwards Enterprise Server - PORTTEST Fails

If PORTTEST fails on the JD Edwards EnterpriseOne Enterprise Server, check the host entries in the /etc/hosts file. See Edit the /etc/hosts File for Connectivity section in this Learning Path.

Resolving Host Name Issues During Provisioning

If you encounter host name issues during an orchestration because the provisioning was run against machines with the long host names, FQDNs, or both, then use the following procedure to resolve the issue:

  1. On the operating system level, change the host names of all the servers in the on-premises architecture to a short host name without the FQDN.
  2. On the database level, change all the host names in the JDE tables to short host names without the FQDN.
  3. Ping all the machines with short host names and verify that the ping test is successful.

See Understanding Machine Names for more information.

Health Checks Fails During Final Step in One-Click Provisioning

If predefined ports for the Enterprise Server are either not enabled or not properly defined in the VCN, health checks can fail during the final step in One-Click Provisioning (as shown in the example below). For details on enabling this setting, For details on the required settings, refer to: Enterprise Server Predefined Ports.

Security Server JdeSocket Or SocketException Errors When Logging In To EnterpriseOne

If a JD Edwards EnterpriseOne Server is deployed inside a firewall-protected network, you must enable predefined ports so that all NET server ports are allocated within the predefined port range. For details on enabling this setting, refer to the section of this document entitled: Enterprise Server Predefined Ports.

Error ORA-00904: "NRUSER1": Invalid Identifier for Table F980051

The JAS log files display the following error after you install EnterpriseOne using the JD Edwards One-Click Provisioning Console with Tools 9.2.2.4:

[SEVERE] - [JDBJ] SQLException occured in the SQLPhysicalConnection.select():Auto: | Table or View Name = F980051 - Data Source[0] = System - 920 java.sql.SQLSyntaxErrorException: ORA-00904: "NRUSER1": invalid identifier

[SEVERE] - [BASE] com.jdedwards.database.base.JDBException: [SQL_EXCEPTION_OCCURRED] An SQL exception occurred: ORA-00904: "NRUSER1": invalid identifier . java.sql.SQLSyntaxErrorException: ORA-00904: "NRUSER1": invalid identifier

[SEVERE] - [BASE] com.jdedwards.database.base.JDBException: [SQL_EXCEPTION_OCCURRED] An SQL exception occurred: ORA-00904: "NRUSER1": invalid identifier . com.jdedwards.database.base.JDBException: [SQL_EXCEPTION_OCCURRED] An SQL exception occurred: ORA-00904: "NRUSER1": invalid identifier

Cause:

The delivered image for One-Click Provisioning has incorrect OCM mappings for the F980051 table. The OCM for this table is pointing to the System - 920 database data source where the structure of the table is not consistent with the table specs.

Solution:

  1. Check both the System and the Server Map data source OCM mappings and change the OCM mappings as shown in the following list:
    • F980051 (Bell Status) mapped to Control Tables - <Environment>
    • F980052 (Notification Execution History) mapped to Control Tables - <Environment>
    • F980053 (Subscription Metadata) mapped to Control Tables - <Environment>
    • F980054 (Notification Queue) mapped to System - 920
    • F980055 (Notifications Constants) mapped to System - 920
    • F980056 (Notifications Offline Repository) mapped to System - 920
    • F980057 (Notifications Offline Repository Archive) mapped to System – 920

  2. Restart the Enterprise service.

For more information, see My Oracle Support Doc ID 2408472.1.

Orchestrating JD Edwards EnterpriseOne Servers Return IPV6 Is Not Disabled

Use this procedure if One-Click Provisioning returns this message for any Linux server that you are provisioning:

  • IPV6 Is Not Disabled
    1. Ensure that the /etc/sysctl.conf file contains the below list of entries:

      net.ipv6.conf.default.disable_ipv6 = 1

      net.ipv6.conf.all.disable_ipv6 = 1

    2. If the above settings do not exist, you must update the /etc/sysctl.conf file to include the above settings.
      Important: A mandatory reboot is necessary for the settings to take effect.

    3. Verify the output of the below command to ensure there is no entry for the string "inet6"

      sudo ifconfig | grep inet6

      If the string "inet6" continues to be in the output of the above command output, verify your settings.