Alexey Serbin has posted comments on this change.

Change subject: [rpc] handling ERROR_UNAVAILABLE RPC error
......................................................................


Patch Set 9:

(5 comments)

http://gerrit.cloudera.org:8080/#/c/6640/9/src/kudu/client/client.h
File src/kudu/client/client.h:

Line 528:   friend class kudu::SecurityUnknownTskTest;
> the kudu:: namespace shouldn't be necessary here.
Unfortunately, it's needed: the compiler does not see the class otherwise.


http://gerrit.cloudera.org:8080/#/c/6640/9/src/kudu/integration-tests/security-unknown-tsk-itest.cc
File src/kudu/integration-tests/security-unknown-tsk-itest.cc:

Line 163:   void ReplaceAuthnToken(shared_ptr<KuduClient>& client,
> warning: non-const reference parameter 'client', make it const or use a poi
Done


Line 216:   gscoped_ptr<KuduTableCreator> 
table_creator(client->NewTableCreator());
> Is it necessary to have more than a single tablet for this test to exercise
I wanted to exercise the path where requests are sent to multiple tablet 
servers: that's for broader coverage if looking at the client implementation as 
a blackbox.

Yep, it seems hash partitioning is a better option here.


Line 287:   // path: the meta-cache already contains information on the 
corresponding
> Is there a different error that occurs if a meta-cache lookup happens?
It's the same TimedOut, but error path is little bit different.  But that's not 
the crucial point here.  The crucial point here is to exercise client-->tserver 
RPC path.

The paths related to the meta-data looking are exercised above  while trying to 
create the same table again (GetTableSchema fails) and opening a scanner below 
(GetTableLocations fails)


Line 335:         if (!importer_do_run) {
> If this is just for short circuiting the failure case, why not just not joi
Good idea.  I was thinking about shorter run-time in case of failure, though.  
But given failures are not expected here, it's not worth it, I agree.


-- 
To view, visit http://gerrit.cloudera.org:8080/6640
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I87d780a4ad88c15ceaacfddf6c1b69ed053bb959
Gerrit-PatchSet: 9
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: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-HasComments: Yes

Reply via email to