Note this only pertains to WFA workflows driven by the "OnCommand Workflow Automation package for VMware vCenter Orchestrator".
Upon upgrading to WFA 2.2, I would sometimes see vCO workflows prematurely fail, complaining with "The requested job is in state 'PLANNING'. The job was unsuccessful due to the current state.". The backend WFA job would complete normally.
After a little digging it appears this is related to a new job status the vCO plugin wasn't expecting. All of our jobs post-WFA-2.2 appear to start with this 'planning' state, and the plugin only considers 'pending' and 'scheduled' as valid non-running states - so if it happens to poll during that phase it will abort in vCO. I sent a note over to goodrum with more particulars, but in short we've worked around this by modifying the job status scriptblock within the vCO-side workflow "Wait for WFA job to complete". I can provide more detail if anyone is interested, just let me know.