Dan Burkert has posted comments on this change.

Change subject: [java] KUDU-2013 re-acquire authn token if expired
......................................................................


Patch Set 14:

(11 comments)

http://gerrit.cloudera.org:8080/#/c/7250/14/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 1140:         ? new NonRecoverableException(statusTimedOut)
>From a quick look, it seems like a null cause will be passed all the way up 
>the chain without being dereferenced, so I don't think the ternary is 
>necessary.


Line 1382: 
extra space


http://gerrit.cloudera.org:8080/#/c/7250/14/java/kudu-client/src/main/java/org/apache/kudu/client/AuthnTokenReacquirer.java
File 
java/kudu-client/src/main/java/org/apache/kudu/client/AuthnTokenReacquirer.java:

Line 33: final class AuthnTokenReacquirer {
Add visibility/privacy annotations.


Line 38:   private final Object affectedRpcInfoListLock = new Object();
Make the name of this match the list (there's an extra 'info' currently).


Line 46:   private ArrayList<KuduRpc<?>> affectedRpcList = Lists.newArrayList();
I think this would be better called 'queuedRpcs' or 'rpcsAwaitingToken'.  
Affected is a little bit generic.  I prefer the first because it make the 
desired FIFO semantics explicit.


Line 82:     rpc.addTrace(new RpcTraceFrame.RpcTraceFrameBuilder(
I was thinking that this trace frame should be added when the RPC is popped, so 
that the timestamp includes the time waiting in the queue, but I'm not entirely 
sure.  I think we should get JD to weigh in.


Line 93:   private List<KuduRpc<?>> prepareRpcListToNotify() {
This naming can be improved.  The class is using a standard double-buffering 
technique, but the names and (lack of) comments don't make that clear.  I'd 
suggest 'swapQueuedRpcs'.


Line 96:       Preconditions.checkState(!affectedRpcList.isEmpty());
Move this after the synchronized block (and therefore on the local instead of 
the field), and considering making this an 'assert', since the method is 
private and this would indicate a logic bug in the class.


Line 109:       @Override
Put annotation after comments.


Line 128:       void handleAffectedRpcs() {
naming: 'retryQueuedRpcs' would be more indicative of what is happening.


PS14, Line 165:  handleAffectedRpcs
naming: 'failQueuedRpcs'


-- 
To view, visit http://gerrit.cloudera.org:8080/7250
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I0be620629c9a8345ecd5e5679c80ee76ca4eaa57
Gerrit-PatchSet: 14
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
Gerrit-Reviewer: Jean-Daniel Cryans <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Will Berkeley <[email protected]>
Gerrit-HasComments: Yes

Reply via email to