Github user aweisberg commented on a diff in the pull request:

    https://github.com/apache/cassandra/pull/191#discussion_r168321619
  
    --- Diff: src/java/org/apache/cassandra/net/MessagingService.java ---
    @@ -1664,4 +1676,113 @@ public static boolean 
isEncryptedConnection(InetAddressAndPort address)
             }
             return true;
         }
    +
    +    public void blockForPeers()
    +    {
    +        // TODO make these yaml props?
    --- End diff --
    
    If we are going that route it seems like well then it should just be in the 
YAML. I would say jvm.options should be for things we don't control because 
they aren't directly part of Cassandra.
    
    Another option is as a YAML option but undocumented if you want to not 
clutter the YAML, but then operators get annoyed and confused and ask why it's 
not a yaml option and not documented.
    
    I never really know how to balance the tension of not putting out thousands 
of config options into a file with not having undocumented but important knobs. 
I think this one doesn't need to be in the documented category yet because 
setting it is really as an escape hatch not something people should have to 
worry about. This is just some feature that should silently work and people 
don't have to think about. Then later if we find out we are wrong we can add 
the option and document. This stems the tide of config options people need to 
think about which has value.


---

---------------------------------------------------------------------
To unsubscribe, e-mail: pr-unsubscr...@cassandra.apache.org
For additional commands, e-mail: pr-h...@cassandra.apache.org

Reply via email to