Andrew Wong has posted comments on this change.

Change subject: mini_cluster: support multiple data dirs
......................................................................


Patch Set 11:

(14 comments)

http://gerrit.cloudera.org:8080/#/c/7211/11/src/kudu/integration-tests/internal_mini_cluster.cc
File src/kudu/integration-tests/internal_mini_cluster.cc:

PS11, Line 60: fs_root_
> Can we get rid of this member too? We could put this in the body of the con
Done, but it came at the cost of making opts_ non-const.


http://gerrit.cloudera.org:8080/#/c/7211/11/src/kudu/master/master-test.cc
File src/kudu/master/master-test.cc:

Line 1349: TEST_F(MasterTest, TestMultiDirMaster) {
> If this is just unit test coverage for the MiniMaster I think it should be 
Done


http://gerrit.cloudera.org:8080/#/c/7211/11/src/kudu/master/mini_master.cc
File src/kudu/master/mini_master.cc:

PS11, Line 59: data$0
> I think it would be more readable as "data-$0"
Done


Line 70: void MiniMaster::SetMasterAddresses(const vector<HostPort>& 
master_addrs) {
> nit: add CHECK(!master_) so this is only called when it's not running
Done


Line 71:   opts_.master_addresses = master_addrs;
> nit: pass by value and use std::move() here
Done


PS11, Line 85:   if (opts_.master_addresses.empty()) {
             :     return master_->WaitForCatalogManagerInit();
             :   }
> Add comment: // Wait for catalog manager to be ready if we only have a sing
Done


Line 92:   opts_.rpc_opts.rpc_bind_addresses = bound_rpc_.ToString();
> add: CHECK(!master_) so this is only called after Shutdown()
Done


PS11, Line 93: HostPort(bound_http_).host()
> no need to construct a HostPort here, just call bound_http_.host() and get 
Done


PS11, Line 94: HostPort(bound_http_).port()
> bound_http_.port()
Done


Line 95:   Shutdown();
> Restart() has unusual semantics in the minicluster* classes. They require S
Ah I see, makes sense. I'll save that for another patch then; I don't have 
anything against those semantics.

To just keep the original behavior (with no Shutdown()), I'll just take out 
Shutdown() from Restart(). In all the tests, if Shutdown() is always called 
beforehand, the Shutdown() here would just be a no-op anyway.


http://gerrit.cloudera.org:8080/#/c/7211/12/src/kudu/master/mini_master.cc
File src/kudu/master/mini_master.cc:

Line 100:   if (master_) {
> warning: an integer is interpreted as a character code when assigning it to
Done


http://gerrit.cloudera.org:8080/#/c/7211/12/src/kudu/tserver/mini_tablet_server-test.cc
File src/kudu/tserver/mini_tablet_server-test.cc:

Line 21
> warning: #includes are not sorted properly [llvm-include-order]
Done


http://gerrit.cloudera.org:8080/#/c/7211/11/src/kudu/tserver/mini_tablet_server.cc
File src/kudu/tserver/mini_tablet_server.cc:

PS11, Line 65: "data$0
> I know this is subjective but I think this would be a little more readable 
Done


http://gerrit.cloudera.org:8080/#/c/7211/11/src/kudu/tserver/tablet_server-test.cc
File src/kudu/tserver/tablet_server-test.cc:

Line 2486: TEST_F(TabletServerTest, TestMultiDirServer) {
> This should probably be in its own file if this is just a unit test for the
Done


-- 
To view, visit http://gerrit.cloudera.org:8080/7211
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I52c9352f1a3565d58149cf2c63d37246c6b39c23
Gerrit-PatchSet: 11
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Andrew Wong <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Mike Percy <[email protected]>
Gerrit-Reviewer: Tidy Bot
Gerrit-HasComments: Yes

Reply via email to