Hey!
I'm implementing a system in which the services communicate with each other 
via gRPC. I figured that I could just spawn a client inside of the server 
to talk to another server. Unfortunately, this solution failed for multiple 
simultaneous connections.
What I mean is, I have: *client*, *service1*, *service2*.
*How it works*:
- *client* makes a connection to *service1* and sends data,
- *service1* makes a connection to *service2* and sends data,
- *service1* waits until *service2* finishes processing data,
- *service2* terminates the connection with *service1*
- *service1* terminates the connection with *client*

I'm using gRPC's reactors and the async API.
The *problem*: When I spawn 2 simultaneous connections from the client, the 
flow hangs in different places (it's not always the same place).

*QUESTIONS*
- Is the system properly designed (with the client spawned on the server)?
- Is there an example of something like that maybe somewhere?
- Has such a scenario (or similar) been considered by the authors of the 
gRPC?
- Are there any known issues/pitfalls when doing something like this? Maybe 
I missed setting something in the connection's config?
- Is it possible that the gRPC has not been designed to handle such 
scenarios? I'm having a hard time finding out whether the problem lies in 
my logic or in a library I use.

Thank you in advance!

-- 
You received this message because you are subscribed to the Google Groups 
"grpc.io" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/grpc-io/457d3569-1b9b-4203-ae91-8abeee53ff2fn%40googlegroups.com.

Reply via email to