KeDeng has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/18054 )

Change subject: KUDU-3340 Disable compact on the specified table
......................................................................


Patch Set 4:

(6 comments)

http://gerrit.cloudera.org:8080/#/c/18054/3//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/18054/3//COMMIT_MSG@7
PS3, Line 7: KUDU-3340 Disable compact on the specified table
> Should this affect all compactions? Or just rowset merge compactions?
Yes, in this scenario, the three types of compact are meaningless and can be 
closed together.


http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/common/common.proto
File src/kudu/common/common.proto:

http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/common/common.proto@476
PS3, Line 476:   // It's a compact operation switch.
             :   // If it was be setted to true, the table will not do compact.
             :   optional bool disable_rowset_comp
> Typically these extra configs are "off by default". To follow that style, h
Done


http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/common/common.proto@476
PS3, Line 476:   // It's a compact operation switch.
             :   // If it was be setted to true, the table will not do compact.
             :   optional bool disable_rowset_comp
> +1
Done


http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/integration-tests/alter_table-test.cc
File src/kudu/integration-tests/alter_table-test.cc:

http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/integration-tests/alter_table-test.cc@2536
PS3, Line 2536:   ASSERT_OK(table_alterer->AlterExtraConfig(
> Does it make sense to add a test scenario which verifies that the actual fu
Done


http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/tablet/tablet.cc
File src/kudu/tablet/tablet.cc:

http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/tablet/tablet.cc@2394
PS3, Line 2394:   MonoDelta elapse = MonoTime::Now() - 
last_update_workload_stats_time_;
              :   if (metrics_) {
              :     int64_t scans_started = metrics_->scans_started->value();
              :
> I agree with Andrew. The `workload_score` is used to define how 'hot' table
Done


http://gerrit.cloudera.org:8080/#/c/18054/3/src/kudu/tablet/tablet.cc@2394
PS3, Line 2394:   MonoDelta elapse = MonoTime::Now() - 
last_update_workload_stats_time_;
              :   if (metrics_) {
              :     int64_t scans_started = metrics_->scans_started->value();
              :
> I don't think this is the score we want to be updating. You likely want to
Done



--
To view, visit http://gerrit.cloudera.org:8080/18054
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ia8452bd9151f345fcad72bb9e0f07cd78432757e
Gerrit-Change-Number: 18054
Gerrit-PatchSet: 4
Gerrit-Owner: KeDeng <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: KeDeng <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Yifan Zhang <[email protected]>
Gerrit-Reviewer: Yingchun Lai <[email protected]>
Gerrit-Comment-Date: Tue, 14 Dec 2021 03:48:32 +0000
Gerrit-HasComments: Yes

Reply via email to