The failure is sporadic (in more than 50% of the cases).

The failure occures if the parallel jobs are executed and one of them failed. With the option failFast: true the complete pipeline jobs shall be aborted. The job is logging the jobs which were not yet started but already in the queue. At this point the logging stops.

Normally I would expect in the logging:

"
Failed in branch XXX[Pipeline] }
...
[Pipeline] // parallel[Pipeline] }[Pipeline] // scriptPost stage[Pipeline] script
..."

But it stops with
Failed in branch XXX[Pipeline] }
 

Further mails which are initiated by the post stage are still sent. This gives the indication that the script is still running but the logging is stopped.