Hello

When seeing HTTP 504 outcome in ISAM WRP http logs, can we always presume that the 504 originates from the backend, or could it also be caused by WebSEAL generating a 504 if the backend takes too much time to complete HTTP transaction ? With "%F" being added to 'request-log-format' setting, it seems that HTTP transactions we see in HTTP logs returning 504 do complete with duration below the default https-timeout value of 120s.  

Or the other way around: if a back-end JCT does not reply (or is too slow) to an HTTP transaction sent by WebSEAL, is an entry always logged in its HTTP logs ?

Other way would be to active pdweb.debug tracing but that is the last resort we are considering especially since the behavior occurs most often during the day and we cannot scope the pdweb.debug tracing to only a particular JCT to keep the logs manageable (maybe that could be the subject of another discussion/RFE).