Server Usage Report - fails in multi node config

[npsmith]

Hello,

Our server usage report fails often now that we added 2 worker nodes in addition to the controller node.  The app seems to require that it run only via the controller machine.  Is that correct? If so, I assume i need to use the worker assignment setup to ensure it runs on the controller machine?

Thanks!

[MichalM · Accepted solution]

Hi @npsmith 

The server usage workflow queries the underlying embedded database (MongoDB) which by defaults resides on the controller machine. I assume the workflow fails because it's configured to talk to the database on port 27018 on localhost which is the default setting. When executed on one of the workers, it attempts to access the database on that machine and it's not there. You will be able to fix this by using the hostname or IP address of the controller instead of localhost. 

[npsmith]

Oh wow- thanks so much for that.  I completely overlooked that  :)

Once I referenced the machine name instead of localhost, it worked.

Thanks!!!

[Madhavikishtipati]

Hi @MichalM,

I have used controller IP and port of database, but still is failing on other nodes, only working on gallery node.

[PanPP]

Hi @Madhavikishtipati 

Can you ping the IP on the node where it is failing?

Can you open the port on the node where it is failing?

Hope this helps, if it does please like this post and if it helps resolve your problem, mark it as a solution. If you have any other questions, please let us know.

[MichalM]

@Madhavikishtipati 

The process is only intended to be run on a server node with the database on (I assume this is the Controller node in your case) so there's no need to re-run on the Gallery or Worker nodes.

[Madhavikishtipati]

Hi PanPP

Thank you for your reply.

Yes, you said right, the MongoDB server, port is not working on other nodes, so we need to open firewall on other nodes as well.