Adar Dembo has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11753 )

Change subject: authz: verify tokens on scans
......................................................................


Patch Set 2:

(3 comments)

Just a partial review so far; curious whether you agree with my control flow 
suggestion (which would make interdiff reviewing tough).

http://gerrit.cloudera.org:8080/#/c/11753/2//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/11753/2//COMMIT_MSG@17
PS2, Line 17:     if uses pk:
Does "if uses pk" mean "if projection includes pk"?


http://gerrit.cloudera.org:8080/#/c/11753/2/src/kudu/common/schema.h
File src/kudu/common/schema.h:

http://gerrit.cloudera.org:8080/#/c/11753/2/src/kudu/common/schema.h@588
PS2, Line 588:   std::vector<ColumnId> get_key_column_ids() const {
This could be simpler a la CreateKeyProjection:

    col_ids.assign(col_ids_.begin(), col_ids_.begin() + num_key_columns_);


http://gerrit.cloudera.org:8080/#/c/11753/2/src/kudu/tserver/tablet_service.cc
File src/kudu/tserver/tablet_service.cc:

http://gerrit.cloudera.org:8080/#/c/11753/2/src/kudu/tserver/tablet_service.cc@1495
PS2, Line 1495:   if (FLAGS_tserver_enforce_access_control && 
req->has_new_scan_request()) {
Splitting the authz checks into two chunks makes these RPC implementations less 
readable. Is there any way we could lead with  
LookupRunningTabletReplicaOrRespond, then do the authz work? I imagine the 
split is so that LRTROR doesn't leak information, but maybe it can call a 
variant of MayHaveScanPrivileges() on error to figure out whether to leak or 
not?



--
To view, visit http://gerrit.cloudera.org:8080/11753
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I7a5d81cf215a5d936f8853feba05778038764905
Gerrit-Change-Number: 11753
Gerrit-PatchSet: 2
Gerrit-Owner: Andrew Wong <[email protected]>
Gerrit-Reviewer: Adar Dembo <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
Gerrit-Reviewer: Hao Hao <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Tidy Bot (241)
Gerrit-Comment-Date: Mon, 11 Mar 2019 19:39:38 +0000
Gerrit-HasComments: Yes

Reply via email to