Dan Burkert has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11570 )

Change subject: Extract connection retrying and HA support from HmsCatalog
......................................................................


Patch Set 2:

(11 comments)

http://gerrit.cloudera.org:8080/#/c/11570/2//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/11570/2//COMMIT_MSG@19
PS2, Line 19: No new tests are provided, since the class is tested by the 
existing
> IIUC this only tests HaClient<HmsClient>. Maybe until we're ready for more
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/hms/hms_catalog.h
File src/kudu/hms/hms_catalog.h:

http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/hms/hms_catalog.h@37
PS2, Line 37: class HostPort;
> nit: alphabetical order
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/hms/hms_catalog.h@177
PS2, Line 177: static bool IsFatalError(const Status& status);
> Remove?
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/hms/hms_catalog.cc
File src/kudu/hms/hms_catalog.cc:

http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/hms/hms_catalog.cc@276
PS2, Line 276:       Slice hms_database;
             :       Slice hms_table;
             :       RETURN_NOT_OK(ParseTableName(name, &hms_database, 
&hms_table));
> Could be lifted out of the Execute() call.
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h
File src/kudu/thrift/client.h:

http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@119
PS2, Line 119:  // Client options
> nit: Move to L121.
addresses_ is an option for configuring the underlying thrift client, even if 
it isn't part of the ClientOptions struct.


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@136
PS2, Line 136:  HaClient is defined inline so that it can be instantiated with 
HmsClient and
> One way around this is to do something like:
That would require the user to cast the Client* to the correct concrete type, 
effectively losing type safety.


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@137
PS2, Line 137: which live in modules which are not
             : // linked to be the thrift module.
> Not quite follow this. Do you mean 'which live in modules to which the thri
oops, I had an extraneous word in there.  Hopefully it's more clear now, 
otherwise I can rewrite.


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@186
PS2, Line 186:   // TODO(todd): wrapping this in a TRACE_EVENT scope and a 
LOG_IF_SLOW and such
             :   // would be helpful. Perhaps a TRACE message and/or a 
TRACE_COUNTER_INCREMENT
             :   // too to keep track of how much time is spent in calls to HMS 
for a given
             :   // CreateTable call. That will also require propagating the 
current Trace
             :   // object into the 'Rpc' object. Note that the HmsClient class 
already has
             :   // LOG_IF_SLOW calls internally.
> Should be updated to remove HMS references.
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@262
PS2, Line 262:       VLOG(1) << "Call to " << Service::kServiceName << " 
failed: " << task_status.ToString();
> nit: mind rewriting this with `Substitute()`?
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@274
PS2, Line 274: << "Call to " << Service::kServiceName << " failed after "
             :                  << options_.retry_count << " retries: " << 
first_failure.ToString();
> And this?
Done


http://gerrit.cloudera.org:8080/#/c/11570/2/src/kudu/thrift/client.h@305
PS2, Line 305:       VLOG(1) << "Connected to " << Service::kServiceName << " " 
<< address.ToString();
> And this
Done



--
To view, visit http://gerrit.cloudera.org:8080/11570
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Id8135f4c3995bba0ba28384c35696a1771ff5296
Gerrit-Change-Number: 11570
Gerrit-PatchSet: 2
Gerrit-Owner: Dan Burkert <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
Gerrit-Reviewer: Hao Hao <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Tidy Bot
Gerrit-Comment-Date: Fri, 05 Oct 2018 18:30:33 +0000
Gerrit-HasComments: Yes

Reply via email to