Each validation request sent from the client has a sequence number associated with it. Multiple events can be sent to the server at the same time without waiting for the response from the server, and there is no guarantee that the first request sent from the client will be the first to reach the server. Therefore, if a request is received out of order, Maximo will wait twenty seconds (by default) for the previous sequenced request to reach the server. If the that event does not get to the server after twenty seconds, the message is shown on the client's browser.

The error is displayed when the client has changes that the server does not, due to a network problem, a server error, user lost internet/network connection, server refuses connection, etc. To ensure the state of the client and the server are the same, the framework forces the user to reload.

• The following steps can help you get started with diagnosing the problem before contacting IBM Maximo Support. Expand the item for more details on troubleshooting steps.
• Steps to troubleshoot
  1. First thing to do is set the system property mxe.webclient.lostconnectionwarningonly to true. Enabling this property does not fully remove the message, but it replaces the error message with warning box to allow users to recover from the error.
  2.  In Maximo logging application, add logger webclient.UIRequestManager and set it to debug, save and apply settings. Ensure that Maximo Root logger is set to INFO. This will log information about XMLHttpRequest from a client (browser) to the web server,  and you can determine if the request at the time of the occurrence actually made it to the server or not.
  3. Capture HTTP Archive (.HAR) file using the browser developer tool while reproducing the issue. See How to gather HTTP Archive (HAR) files for Maximo Troubleshooting
  4. Review the HAR and the SystemOut.log generated at the time of the occurrence. In the HAR, if you see a request with a response code HTTP/0.0 with status code 0 and empty status message indication communication error between the browser and the server. Then review the SystemOut.log and search for the UI request with the same UISessionID, PageSeq and XhrSeq identified in the HAR. This error doesn't often reflect in the JVM log, indicating that the request probably never made it to the server. This can be due to a number of reasons, such as Maximo application timed out while waiting for the client request, an interruption in the network, etc., and the message is displayed to notify the user of problems with server connectivity. If you do see the UI request with the same UISessionID, PageSeq and XhrSeq, it will be printed with a message such as "Passed the wait time for handling the request...", meaning that the request sent from the browser did not make it to the Maximo application on the server. You can also verify this by checking the /IBM/HTTPServer/logs/access.logs to see if the request is logged.
  5. Check if users are opening multiple tabs in a browser. When opening multiple tabs, if the user logs out of one tab or is timed out of one session, all tabs are affected, and the error is displayed. You may have the property webclient.multibrowsersuppor enabled so users can work with multiple browser tabs. This property only allows the use of multiple tabs that don't share the same state, but the session cookie is still shared. Therefore, when one browser tab logs out or is timed out, the session cookie is deleted, and all other tabs or windows sessions are immediately ended and the error is displayed.
  6. These UI System Properties that can be tweaked for improvement (these do not stop the message from displaying when there is a communication interruption) 
     • mxe.webclient.lostconnectionwarningonly, Default value is false
     • mxe.webclient.lostconnectionrecheckinterval
     • mxe.webclient.lostconnectionwarninginterval
     • mxe.webclient.asyncrendertimelimit - Maximum number of seconds between the responses that are sent to the user interface when processing multiple async validation requests. Default is 15 seconds, but can be set to as low as 1 second. The lower the value the higher the frequency of the refresh process.
     • mxe.webclient.outOfOrderReqTimeout. Default value is 20 minutes and can be changed to a higher value. Recommendation is not to go beyond 30 minutes
     Review User interface system properties for more web client system properties that define the behavior and characteristics of the Maximo user interface
  7. When application server security is enabled, in some cases the error can be associated with LTPA token timeout settings in WebSphere. In the SystemOut.log, if you see warnings like: SECJ0371W: Validation of the LTPA token failed because the token expired with the following info: Token expiration Date: Fri Mar 01 00:04:55 UTC 2024, current Date: Fri Mar 01 20:34:46 UTC 2024...
     May be followed by the error: SECJ0053E: Authorization failed for /UNAUTHENTICATED while invoking (Bean)MAXIMO#mboejb.jar#accesstokenprovider getAccessToken::3 is not granted any of the required roles: maximouser
     
     
     The LTPA Token timeout value may need to be increased in WebSphere. Default value is 120 minutes.
  8. In IBM Maximo Manage with SAML enabled, this can happen when the SameSite cookie is not Lax. In the 'Additional server config' for the UI bundle, if no secret exist, add:

     <server description="new server">
     <httpSession cookieName="JSESSIONIDUI" cookieSecure="true" invalidateOnUnauthorizedSessionRequestException="true" invalidationTimeout="12h" cookieSameSite="Lax"/>
     </server>

     If a secret already exists, add the following entry into the server xml reference for the secret right before the </server> end tag.

     <httpSession cookieName="JSESSIONIDUI" cookieSecure="true" invalidateOnUnauthorizedSessionRequestException="true" invalidationTimeout="12h" cookieSameSite="Lax"/>
     

     Reference document: Users are getting logged out of MAS Manage when SAML is enabled, error code 0
  9. If application server security is enabled, you may also see random error message "CWWIM4520E javax.naming.CommunicationException" in the SystemOut.log. If it shows up around the time the reload error box is displayed, you can work with your network administrators to check whether there is a network connectivity problem. You can also tune the LDAP performance settings by adjusting the context pool and network connection timeOut in WebSphere.
  10. • Log in to WebSphere console and go to Security > Global Security > Federated Repository > Configure
      • Click your Repository identifier
      • In the Additional properties section, click Performance
      Refer to Random error messages CWWIM4520E javax.naming.NamingException or javax.naming.CommunicationException and LDAP performance issues for more details
  11. Review the Maximo Best Practices for System Performance guide and refer to section 4: IBM HTTP Server and Plugin tuning, compare the settings in your httpd.conf file with the recommended parameter values in the guide. An important one is the MaxKeepAliveRequests which should be set to 0 to allow unlimited request. This can help with the error. Maximo 7.6 Performance Best Practice Guide
  12. This message is often related to the network (VPN, firewall, router, connection drop, latency, etc) or something in the load balancer.
      • Set up a TCP/IP trace using WireShark or any other network monitoring tool on both the server and client, and work with your network team to review any network issues.
      • Review and modify (if necessary) the load balancing option that the plug-in uses when sending requests to the application servers associated with that web server. Review the section 4.2: Load balancing in the Maximo Best Practices for System Performance guide for more details.