simbadzina commented on code in PR #6510:
URL: https://github.com/apache/hadoop/pull/6510#discussion_r1475654924
##########
hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/server/namenode/ha/RouterObserverReadProxyProvider.java:
##########
@@ -84,7 +84,8 @@ public class RouterObserverReadProxyProvider<T> extends
AbstractNNFailoverProxyP
public RouterObserverReadProxyProvider(Configuration conf, URI uri, Class<T>
xface,
HAProxyFactory<T> factory) {
- this(conf, uri, xface, factory, new IPFailoverProxyProvider<>(conf, uri,
xface, factory));
+ this(conf, uri, xface, factory,
+ new ConfiguredFailoverProxyProvider<>(conf, uri, xface, factory));
Review Comment:
In the original code I assumed the routers would all sit behind a single
host name and traffic would be split via DNS. The IpFailoverProxyProvider is
then needed incase one router is down but the IP is still in the DNS record.
I can see how in other setups users may need to list all the routers
explicitly in the client configurations.
Can we make the proxy provider configurable? To support both usecases.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
