I'm trying to build some configuration distribution with gRPC for All is working, but I could use a second opinion on whether 
I'm implementing error handling correctly. 
My clients would like to setup a connection and then listen forever. So far 
I have found that they successfully reconnect after network blips, but when 
I deploy the server my clients will hang unless I throw away the 
connection. Is that expected? 

My approach has been to treat all errors as semi-fatal, back off myself, 
nuke the connection and start over.

I have some open source client libraries and 

Relevant code is in

Does that seem like the right approach? 

(For bonus points you can see my kinda weird "stub" cache. That felt... 
weird...   but I was trying to avoid creating new stubs everywhere. Am I 
being silly?)



You received this message because you are subscribed to the Google Groups 
"" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
To post to this group, send email to
Visit this group at
To view this discussion on the web visit
For more options, visit

Reply via email to