Dan Burkert has posted comments on this change.

Change subject: [java-client]: support for Kerberized RPC
......................................................................


Patch Set 2:

(12 comments)

http://gerrit.cloudera.org:8080/#/c/5150/1//COMMIT_MSG
Commit Message:

PS1, Line 10: ha
> redundant
Done


http://gerrit.cloudera.org:8080/#/c/5150/1/java/kudu-client/src/main/java/org/apache/kudu/client/AsyncKuduClient.java
File java/kudu-client/src/main/java/org/apache/kudu/client/AsyncKuduClient.java:

Line 950:    * The subject contains credentials necessary to authenticate to 
Kerberized Kudu clusters.
> Nit: "to connect to Kerberized..." maybe?
Done


http://gerrit.cloudera.org:8080/#/c/5150/1/java/kudu-client/src/test/java/org/apache/kudu/client/MiniKdc.java
File java/kudu-client/src/test/java/org/apache/kudu/client/MiniKdc.java:

PS1, Line 228: createKrb5Conf
> Would it be useful to extract this and kdc.conf into template files that ar
They have already diverged somewhat - see the additional restrictions that are 
being added on cipher types here due to Java restrictions.  Unless you feel 
strongly I'm going to skip this, I don't think it would simplify things overall.


Line 333:   }
> If you've removed redirectOutput() because PIPE is the default value, you c
Done


http://gerrit.cloudera.org:8080/#/c/5150/1/java/kudu-client/src/test/java/org/apache/kudu/client/MiniKuduCluster.java
File java/kudu-client/src/test/java/org/apache/kudu/client/MiniKuduCluster.java:

Line 107
> Technically this is still true...
It wasn't adding any information.


Line 111
> Why remove this precondition?
I briefly had a test that set the number of tservers to 0 for debugging (less 
log spew).  I kept the removal since I don't see the point in restricting this. 
 I use 0 tserver tests pretty heavily when testing the RPC layer for the rust 
client, for instance.


Line 75:   private final List<String> pathsToDelete = new ArrayList<>();
> Sorry to pick on you for this (if it helps, I've bugged JD too), but this p
I actually made this a lot worse than it needed to be, I had replaced 
'environments' (map of maps) with just a single environment, but forgot to 
finish removing all the uses.  Now it's just a single additional map shared by 
the whole cluster.


Line 220:     }
> Nit: can you inline this two lines below?
Done


Line 299:    * Starts a process using the provided command and configures it to 
be daemon,
> Nit: inline
Done


Line 320:     
thread.setName(Iterables.getLast(Splitter.on(File.separatorChar).split(command.get(0)))
 + ":" + port);
> Update.
Done


PS1, Line 336:    * Starts a previously killed master process on the specified 
port.
             :    * @param port which port the master was listening on for RPCs
             :    * @throws Exception
> This is also in MiniKdc.java. Perhaps it can be decomposed into a helper me
I ended up removing this.


http://gerrit.cloudera.org:8080/#/c/5150/1/java/kudu-client/src/test/java/org/apache/kudu/client/TestMiniKdc.java
File java/kudu-client/src/test/java/org/apache/kudu/client/TestMiniKdc.java:

Line 43:       assertFalse(klist.contains("[email protected]"));
> Why did this change?
We no longer allow multiple logged in client principals due to moving from DIR 
to FILE typed ccache.  A similar change was made on the C++ side in 
123b9918c7005f65010c5d431f4e3cac459e7a31


-- 
To view, visit http://gerrit.cloudera.org:8080/5150
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I5131edb1f2bda443f7980a4aad86362666b3b6f5
Gerrit-PatchSet: 2
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Dan Burkert <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
Gerrit-Reviewer: Jean-Daniel Cryans <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-HasComments: Yes

Reply via email to