Re: Re[4]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread Matt Burgess
Ugh I remember now, I think you're running into [1], it still needs to be done. As a workaround there might be an approach as you have with specifying the controller service ID as a property and then use "getControllerServiceLookup" to find the CS by ID. Regards, Matt [1] https://issues.apache.or

Re[4]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread David Early
Matt, I apologize for returning to the well, but we are truly stuck. Here is what I have: Controller configuration with the dynamic property with the ID of the controller service: Full code as it stands right now: /* * Licensed to the Apache Software Foundation (ASF) under one or more * c

Re: Re[2]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread Matt Burgess
The formatting got a bit wonky on the code snippet you provided, but if your GroovyRecordSetWriterFactory extends AbstractControllerService, it should have access to the getProperty() method. Try without the context, just "getProperty(CACHE_CLIENT).asControllerService(DistributedMapCacheClient)" O

Re: Re[2]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread davide
Matt, So we learned somebut not enough there. This snippet of code: @Override def init(context) { mapCacheClient = context.getProperty(CACHE_CLIENT).asControllerService(DistributedMapCacheClient) The problem seems to get that the "getProperty().asController

Re: Re[2]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread Matt Burgess
Dave, You can't override the initialize() method as it is final, but note that it calls the init() method at the bottom, that's the method you can override, its signature is: protected void init(final ControllerServiceInitializationContext config) throws InitializationException so instead of "de

Re: Re[2]: access property in ScriptedRecordSetWriter?

2020-08-24 Thread davide
All, I am drafting a co-worker to go try and understand where this is getting hung up, but if anyone has any additional information/thoughts on how to get this working, I would greatly appreciate it. Matt, thanks so much for your help, if you have an additional epiphany, let me know! Dave --

Re: Nifi takes too long to start(~ 30 minutes)

2020-08-24 Thread Mohit Jain
Hi Andy, Apologies for the late reply. Yes there are around 25000 processors (1000 flows with around 25 processors each, I mistyped.) I started the nifi in the debug mode to get something but nothing concrete. Following logs gets printed for all the processors: 2020-08-24 22:56:04,416 INFO org.a