JR61467: AFTER YOU RECEIVE AN ERROR MESSAGE THAT INCLUDES "STATUS:0", YOUR TASK ENDS UNEXPECTEDLY

workflow.19003.delta.repository

APAR status

• Closed as new function.

Error description

• In IBM Process Portal, you might see an error message briefly
  before a task closes. Outside Process Portal, you see an error
  message that includes a relative URL that is followed by
  "status: 0" and your task is no longer in your Work list in
  Process Portal; the task ended unexpectedly. The process remains
   in a failed state until an administrator makes the task
  available again.
  
  The error message will look similar to
  
  Error: The "Service" service failed with the following
  information: RequestError: Unable to load
  /rest/bpm/wle/v1/coachflow/service/1.dcab62ff-0e85-466c-8f04-8ec
  4c106da62?modelID=1.16c5d012-1933-4204-ac92-b534e8b1dc8f&callAct
  ivityID=2025.d6e79572-fc0c-46af-8d9d-87ceb8e9b712&branchId=2063.
  974ecaca-f166-4d64-aee2-62bcf8062274 status: 0
  
  PRODUCTS AFFECTED
  IBM Business Process Manager (BPM) Advanced
  IBM BPM Standard
  IBM BPM Express
  

Local fix

• In Process Inspector, click "Retry failed steps". The task
  resumes from the last save point in your work list. If an error
  occurs, move the token to an earlier step in the process.
  

Problem summary

• When an http request made by a client-side human service task
  ends, the request is often observed as having a status of 0. The
   resulting error ends the task and the error is reported to the
  process instance. If the process doesn't catch all errors from
  the task, the process instance goes into a failed state. If the
  task is running in Process Portal, the task view closes when the
   error occurs so you won't have time to read the error message,
  if you see it at all. An administrator must identify the failed
  process instance and retry the task to make it visible in your
  Work list again.
  
  The request might have ended unexpectedly because of a temporary
  inability to communicate with the server.
  

Problem conclusion

• This fix changes the behavior when an http request made by a
  client-side human service task is aborted.
  
  An error resulting from the aborted request will be displayed to
  the user with a new message. The error will not be reported to
  the process instance. The state of the process instance will
  remain unchanged and the task will remain in the Work list. The
  user can launch the task again to resume from the last save
  prior to the error without requiring assistance from an
  administrator. The new message is:
  
  No response was received from the server. Try the call again.
  
  When the task is launched from within Process Portal the error
  page will now remain open for users to see. This applies to all
  errors that end a client-side human service task.
  
  A pair of configuration options can be added as custom
  properties on the Mashups_ConfigService resource environment
  provider to revert to the previous behavior:
  1. failTaskOnAbortedRequest: set to "true" to have the error
  passed to the process instance to either be caught by the
  process or to put the process instance into a failed state
  2. closeTaskOnError: set to "true" to have Portal continue to
  close the task view when an error occurs in the task
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  BPM

• Fixed component ID

  5737A5700

Applicable component levels