That’d be very strange. From [0] NOTE: ExecuteScript will perform a session.commit() at the end of each execution to ensure the operations have been committed. You do not need to (and should not) perform a session.commit() within the script.
If that’s not the case, it sounds like a bug. @alessio: you have a strange way of writing Python code. [0]: https://community.hortonworks.com/articles/75032/executescript-cookbook-part-1.html From: Wes Lawrence [mailto:[email protected]] Sent: Monday, February 20, 2017 4:56 PM To: [email protected] Subject: Re: Rebel flowfile and python scripts. Hey Alessio, I've only used the ExecuteScript processor once before, but the one thing I noticed was the session wasn't being commited after the transfer (session.commit()). If you add that after the transfers on lines 23 and 26, does that make a difference? Also, what's the timer strategy for this processor? CRON or regular timer? --Wes On Mon, Feb 20, 2017 at 4:00 AM, Alessio Palma <[email protected]<mailto:[email protected]>> wrote: Hello all, there is a strange behavior on our NIFI installation. I wrote a script in python and for some reason and sometimes the flowfile does not always moves into it, like in this case: [cid:[email protected]] I see that Queued Duration for the above flowfile is 07:53:04.557! What is it waiting for ? The executeScript processor is running on each node with concurrent Task set to 1. In attachment there is the flowfile source code. Any help is appreciated. Thanks in advance. AP
