You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@nifi.apache.org by "Fields, Bradley E CTR (USA)" <br...@mail.mil.INVALID> on 2023/06/12 15:47:34 UTC

ExecuteStateless processor

We have been trying to use the ExecuteStatless processor with our kafka
dataflow but we are having major issues. We have followed the only youtube
video that we have found to make sure our setup is good. The issue we have
is as soon as this processor is turned on, we start seeing the errors of
"too many open files handles". We are running this on a RHEL 7 environment
and have set our limiteNOFILE to 100000 in our
/usr/lib/system/system/confluent-nifi.service. This seems to work perfect
with the dataflow without the executestateless processor. 

 

I will try to provide everything I can, but we are running this on a private
system so some things will be harder to provide. Any help will be
appreciated. 

 

Bradley Fields

JRSS IT Data Engineer, TN11 JRSS

Contractor, Leidos Inc.

 


Re: ExecuteStateless processor

Posted by Mark Payne <ma...@hotmail.com>.
Hello,

I recommend running:

lsof -p <nifi pid>

This will give you a listing of all file handles that NiFi has open. From that, you should be able to better understand what files (or sockets) are open.

Thanks
-Mark


On Jun 12, 2023, at 11:47 AM, Fields, Bradley E CTR (USA) <br...@mail.mil.INVALID> wrote:

We have been trying to use the ExecuteStatless processor with our kafka dataflow but we are having major issues. We have followed the only youtube video that we have found to make sure our setup is good. The issue we have is as soon as this processor is turned on, we start seeing the errors of “too many open files handles”. We are running this on a RHEL 7 environment and have set our limiteNOFILE to 100000 in our /usr/lib/system/system/confluent-nifi.service. This seems to work perfect with the dataflow without the executestateless processor.

I will try to provide everything I can, but we are running this on a private system so some things will be harder to provide. Any help will be appreciated.

Bradley Fields
JRSS IT Data Engineer, TN11 JRSS
Contractor, Leidos Inc.