Support Questions

Find answers, ask questions, and share your expertise

Nifi - RPG - does not have any input port error. HTTPS enabled.

Contributor

Hi Team,

How to enable RPG on https enabled cluster because we are facing "does not have any input ports error" but in http it was working.

Please guide me to solve this thread.

6 REPLIES 6

Master Guru

@Anishkumar Valsalam

NiFi Site-To-Site between two secured NiFi instances will have several authorization policies that need to be set.

There is one Global policy:
--> Retrieve Site-To-Site Details: This one must be set for an RPG to be able to connect to the target NiFi and retrieve details like load and number of nodes (If target is a NiFi cluster).

There are two other component level policies:

--> receive data via site-to-site: This policy is set on all remote input ports and designates what clients (Source NiFi instance running Remote Process Group (RPG)) is authorized to send to this port.

--> send data via site-to-site: This policy is set on all remote output ports and designates what clients (Source NiFi instance running Remote Process Group (RPG)) is authorized to pull data from this port.

You are seeing "does not have any input ports error" because you are trying to drag a connection to your RPG; however. the target NiFi does not have any remote input ports for which your client has been authorized to send to.

Thanks,

Matt

Hi @Anishkumar Valsalam

As Matt explained, this is a matter of policies between the two clusters.

Look at option 2 of this article to understand what policies you should put in place : https://community.hortonworks.com/articles/88473/site-to-site-communication-between-secured-https-a....

If you need 2 ways communication then you need to configure the two clusters to accept S2S communication from each other.

I hope this helps

Contributor

Hi @Matt Clarke @Abdelkrim Hadjidj

Thanks for the detailed note but stll the below 2 options are not enabled eventhough we careated polocies (Retrieve Site-To-Site Details) for nodes.

-> receive data via site-to-site

--> send data via site-to-site

42389-capture.png

And please guide now in standalone cluster we are facing some load(mem / cpu) , In cluster do we face this kind of issue?

or it will distribute the data and process so that the above mem/cpur load issue we can avoid.

Is my understanding correct?

Hi @Anishkumar Valsalam

Is the capture from your input port policies? can you better explain what are the configuration you did?

Contributor

@Abdelkrim Hadjidj

Thanks i am able to create policy now , I was checked into default flow policy so its was not enabled.

In Input port policy it is enabled

And In your flow i can see this is between cluster to cluster.

https://community.hortonworks.com/articles/88473/site-to-site-communication-between-secured-https-a....

How we can test the same between 3 instances ( 3 node cluster) ?

I am looking for the sample flows to test RPG in the cluster.

Thanks Matt and Abdelkrim.

Hi @Anishkumar Valsalam

Perfect, thanks for the feedback. If you found that this answer addressed your question, please take a moment to click "Accept" below and close the thread.

Take a Tour of the Community
Don't have an account?
Your experience may be limited. Sign in to explore more.