Troubleshooting application configuration

Use these resources to resolve the problems that you might encounter with configuration of the IBM Process Mining application.

Error 404 on port 80

This issue occurs when you install IBM Process Mining on the RedHat Linux 8 operating system.

Symptom

When you install IBM Process Mining, you get 404 error on Port 80.

Workaround

Check for the following command lines for server on port 80 in the (/etc/nginx/nginx.conf) file:

location / {
}

If you see the above-mentioned command lines in the nginx.conf file, then the undefined redirection to Port 443 causes the issue.

To define the redirection to Port 443, do the following steps:

  1. Remove the following command line:
location / {
}
  1. Add the following command line for the redirection:
return 301 https://$host$request_uri;
  1. Restart the NGINX server after making the changes.

Application launch error

This issue occurs when your IBM Process Mining failed to start.

Symptom

When you start your IBM Process Mining application, you get HTTP Error 503.

Workaround

You get this error because your database can be uncorrectly configured.

To see whether the database is configured correctly, check the following log file:

<PM_HOME>/jetty-web/logs/yyyy_MM_dd.jetty.log

Process app launch fail

This issue occurs when the process application fails to start and displays an error related to the Python version.

Symptom

If your process app failed to start, you get the following information in the log file:

./pm-accelerators.sh start
Using tmp: /opt/processmining/repository/temp
Starting accelerator server...
output to [/opt/processmining/repository/logs/pm-accelerator-sap-p2p.log] started.
Using tmp: /opt/processmining/repository/temp
Starting custom accelerator service...
output to [/opt/processmining/repository/logs/pm-accelerator-custom-std.log]
initiating python venv..
    first time running....
'Python3.12' is WAS NOT found as specified in 'environment.conf'
    '/usr/bin/python3.12'
    custom accelerator requires python3.12 installed.
    please check python3.12 is installed and available.

Workaround

If you already have a custom deployment of Python 3.12 or later and you want to use it instead of a standard Python 3.12, then ensure that your configuration is available in the <PM_HOME>/bin/environment.conf file.

CMD_PYTHON=/usr/bin/python3.12

In addition, commit an override check on Ubuntu.

ACF_CUSTOM_UBUNTU_CHECK=1

If you have an existing deployment of standard Python 3.12, then in the <PM_HOME>/bin/environment.conf file run the following command:

CMD_PYTHON=/usr/bin/python3.12

You must also confirm if your original deployment matches with the details that are listed in <PM_HOME>/bin/environment.conf

which python3.12
/usr/bin/python3.12

For more information about installing Python 3.12 for process app, see Installing Python for Custom process app.

Error while running update.sh

This issue occurs when you attemt to upgrade IBM Process Mining to version 1.14.2.

Symptom

When you attempt to upgrade the IBM Process Mining to version 1.14.2 from version 1.13.1 or earlier, the system throws the following error after running ./update shell script:

./update.sh
current deployment is older than version '1.13.1'
this upgrade is ONLY available for upgrading from '1.14.1' to '1.14.2'
please follow documentation on how to upgrade to '1.14.1' before attempting this upgrade

When you attempt to upgrade the IBM Process Mining to version 1.14.2 from any version other than 1.14.1, the system throws the following error after running ./update shell script: :

./update.sh
current deployment is not applicable for this upgrade.
current release: [<current release>]
current version: [<current version>]
this upgrade is ONLY available for upgrading from '1.14.1' to '1.14.2'
please follow documentation on how to upgrade to '1.14.1' before attempting this upgrade

If you do not set the environment variable when you upgrade the IBM Process Mining to version 1.14.2, the update.sh throws the following error message:

./update.sh
Error: PM_HOME environment variable is not set. Set it to the processmining home.

Application service startup might fail on SELinux

This issue occurs when you start the application service on Red Hat with Security-Enhanced Linux (SELinux).

Symptom

After you start the application, you get the 203/exec error.

Workaround

You get this error, because you need to allow the application script to be run by the Linux service. To fix the issue, do the following steps:

  1. Run the following command to check whether SELinux is enabled. For more information, see Changing SELinux states and modes.
getenforce

The getenforce command must display an output Enforcing.

  1. To allow the script to run, execute the following command:
chcon -t bin_t /opt/processmining/bin/<SCRIPT NAME>

This command is a temporary option that you can use for testing.

  1. To keep the configuration after system restart, run the following command:
semanage fcontext -a -t bin_t '/opt/processmining/bin/<SCRIPT NAME>'
restorecon -Fv /opt/processmining/bin/<SCRIPT NAME>

Configure the following scripts for the <SCRIPT NAME> value:

  • pm-accelerators.sh,
  • pm-analytics.sh pm-engine.sh,
  • pm-eureka.sh,
  • pm-monitoring.sh,
  • pm-web.sh.

You can find them in the directory /opt/processmining/bin/.

  1. Optional: Assign the execute rights to shell scripts. For example, use the following commands:
  • chmod +x /opt/taskminer/tomcat/bin/*.sh
  • chmod +x /opt/taskminer/bin/*.sh
  • chmod +x /opt/processmining/bin/*.sh