KM Portal 8.6.2

After selecting the KM Search Article in the portal header close button not working, Once click the close button browser was hanging and unable to close the screen. kindly help up on how to resolve the issue and unable to trace the issue.

@RajasekharN1860

Can you replicate this on all types of browser? Did this used to work? When did the issue first start to appear?

I’m aware of the following issues in some versions of Pega KM:

When the user attempts to close a tab in the main harness, the screen can hang and become unactionable until a page refresh. The root cause of the issue is caused by code that was added in pzpega_ui_doc_dirtyhandling sometime between 8.3 and 8.5.2. A console error was also visible when the screen would hang ( A console error was also visible “Cannot read property ‘innerHTML’ of null”)

Workaround:

Add the following JS local change, either in UserWorkForm or through a js file added to the harness.

pega.u.d.bUseNativeDirtyConfirm = true;

(The workaround is to move only the “pega.u.d.bUseNativeDirtyConfirm = true;” JS code into this text file, and then remove the line from UserWorkForm. This text file would be included in the main harness Scripts & Styles tab. Steps below:

  1. Create a new text file within Pega
  2. Name it, and mark the extension as js (for JavaScript)
  3. Copy the same User Work Form code we added previously, into this new JS file.
  4. Include this js file within the “Scripts & Styles” tab of your Harness, and then Save
  5. Now test the behavior to ensure that you are still able to close tabs in the end-user portal without the previous script being included in User Work Form."

This local change is effective up to Pega 8.7 where the code was fixed.

If you’ve followed the workaround steps and this issue is still persisting, then please check the browser trace and share the Pega ALERTSECURITY logs so that we may analyze this warning message.

@MarijeSchillern Thanks for ur response i raised pega SR so they fix the issue.

@RajasekharN1860 That’s good to hear! Would you be able to provide the incident number so that we might learn from the issue?