Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/22796#discussion_r227046597
--- Diff: dev/deps/spark-deps-hadoop-2.7 ---
@@ -148,7 +148,7 @@ metrics-graphite-3.1.5.jar
metrics-json-3.1.5.jar
metrics-jvm-3.1.5.jar
minlog-1.3.0.jar
-netty-3.9.9.Final.jar
+netty-3.7.0.Final.jar
--- End diff --
I see, I think what's happening is that we still have a netty 3.x
dependency from Hadoop. I guess we should simply leave that in place.
I think that previously we had tried to manually exclude netty dependencies
because it helped SBT resolve the version in the same way as Maven. I don't
know if we actually want to take netty 3.x away from ZK and Hadoop if it wants
them.
At least, this change doesn't actually get rid of netty 3.x from the
distribution, but we don't need to do that necessarily. I just wanted to make
sure Spark itself has no reference to it nor is trying to manage it or depend
on it.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
