Thread status and disk % full

This example shows using the Alert Configurator to define alerts that notify the operator if any thread goes down. It then runs a script if disk space is more than 90% full for at least 30 seconds.

Note: If your process configuration uses Disk-based Message Queuing, then use an alert for 90% full disk space.
  1. Open the Alert Configurator and click New Append.
    Use the default thread status alert.
  2. In the Trigger pane, select the thread names to monitor in Source. Click the button to select the thread names from a list of all threads that are configured for the system.
  3. Click Apply.
  4. Set Source Count to At Least and specify 1 in the text box to the right. Do this to see if any threads enter the down state.
  5. In Comparing, select == from the menu. This is the default.
  6. Then select down in the Comparing .
  7. Set Duration to Once. This is the default. The example does not specify that the alert condition has to exist for a certain period of time. The triggers as soon as a thread enters the down state.
  8. Click Apply.
  9. In the Actions pane, select notify. This is the default. This sets the Alert Action to notify the operator when the alert triggers.
  10. Click Update and then click Apply. The updated alert displays on the Alerts List.
  11. On the Alert Configurator, click the New Append tool.
  12. In the Trigger pane, select disk % full from the Alert Type menu.
  13. Clear Delta, if required, to check the actual percentage of full disk space.
  14. In Source, specify the path of the root directory to monitor after the drive letter.
    For example, path names should be in the form of: C:\cloverleaf\cis6.2\integrator
    Or, you can click the button to open the Select File System dialog box, where you select the drive. Click Apply and the selected drive displays in the Source field, where you can finish specifying the path.
  15. Click Apply.
  16. Set Source Count to All to indicate when the file system approaches capacity.
  17. Set Comparing to > and specify 90 in the adjacent text box. This triggers the alert when the disk space is more than 90% full.
  18. Set Duration to N Seconds and specify 30 in the adjacent text box. This triggers the alert when disk space remains more than 90% full for 30 seconds.
  19. Click Apply.
    The next alert triggers two actions. First, a shell script runs. In this example, it is UserCleanup. This notifies users to clean up the directories. Then, a message is sent to root at the specified trigger.
  20. In the Actions pane, select exec from the Alert Action menu. The pane reconfigures for command input.
  21. Specify the name of the script (UserCleanup) in the Command text box.
  22. Click Update.
  23. Click Apply.
  24. On the Alerts list of the Alert Configurator, click the alert configured, if necessary.
  25. Click the Append tool in the Actions pane.
  26. Select the exec alert action.
  27. In Command, specify mailx -s ’%A’ root < \dev\null (UNIX).
    This sends a message to root with the "disk % full" subject line and an empty message body.
  28. Click Update.
  29. Click Apply. The new alert configuration displays on the Alerts List.
  30. Click Save on the Alert Configurator toolbar after configuring all alerts. Or, select File > Save As to open a file selection dialog box.
  31. Specify a file name, or select from the listed files and click Save.