Fengling Wang has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11727 )

Change subject: [location_awareness] Add ts location to TSInfoPB
......................................................................


Patch Set 2:

(7 comments)

http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc
File src/kudu/integration-tests/table_locations-itest.cc:

http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@127
PS2, Line 127: properply
> Nit: properly
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@128
PS2, Line 128: class TableLocationsWithTSLocationTest : public KuduTest {
> You should be able to extend the existing test fixture; just add a virtual
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@138
PS2, Line 138:                                                    
"scripts/first_argument.sh");
> Nit: indentation
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@301
PS2, Line 301: req.set_max_returned_locations(3);
> Is there any significance of how many returned locations are there?
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@307
PS2, Line 307: EXPECT_EQ("", 
resp.tablet_locations(0).replicas(0).ts_info().location());
> What about checking additional 2 locations?
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@315
PS2, Line 315:
             :   { // Check that the master doesn't give back partial results 
while the table is being created.
             :     GetTableLocationsRequestPB req;
             :     GetTableLocationsResponsePB resp;
             :     RpcController controller;
             :     req.mutable_table()->set_table_name(table_name);
             :
             :     for (int i = 1; ; i++) {
             :       if (i > 10) {
             :         FAIL() << "Create table timed out";
             :       }
             :
             :       controller.Reset();
             :       ASSERT_OK(proxy_->GetTableLocations(req, &resp, 
&controller));
             :       SCOPED_TRACE(SecureDebugString(resp));
             :
             :       if (resp.has_error()) {
             :         ASSERT_EQ(MasterErrorPB::TABLET_NOT_RUNNING, 
resp.error().code());
             :         SleepFor(MonoDelta::FromMilliseconds(i * i * 100));
             :       } else {
             :         ASSERT_EQ(1, resp.tablet_locations().size());
             :         break;
             :       }
             :     }
             :   }
> OK, then please encapsulate it into a common helper method. And maybe combi
Done


http://gerrit.cloudera.org:8080/#/c/11727/2/src/kudu/integration-tests/table_locations-itest.cc@350
PS2, Line 350: ASSERT_EQ("/foo", 
resp.tablet_locations(0).replicas(0).ts_info().location());
> Does it make sense to add an additional assert on expected size of the tabl
Done



--
To view, visit http://gerrit.cloudera.org:8080/11727
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ic5865cb47698817097a7f62c968c5cfd80e07259
Gerrit-Change-Number: 11727
Gerrit-PatchSet: 2
Gerrit-Owner: Fengling Wang <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Fengling Wang <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Will Berkeley <[email protected]>
Gerrit-Comment-Date: Wed, 24 Oct 2018 04:18:26 +0000
Gerrit-HasComments: Yes

Reply via email to