Mike Percy has posted comments on this change. Change subject: tablet copy: Allow voting from crashed initial tablet copies ......................................................................
Patch Set 3: (4 comments) http://gerrit.cloudera.org:8080/#/c/7961/3/src/kudu/master/sys_catalog.cc File src/kudu/master/sys_catalog.cc: Line 243: boost::none, > mind adding a /* name= */ comment here so it's clearer the argument? the re Done http://gerrit.cloudera.org:8080/#/c/7961/3/src/kudu/tablet/tablet-harness.h File src/kudu/tablet/tablet-harness.h: Line 103: boost::none, > same Done http://gerrit.cloudera.org:8080/#/c/7961/3/src/kudu/tserver/tablet_copy_client.cc File src/kudu/tserver/tablet_copy_client.cc: PS3, Line 293: if (last_logged_opid) { : *superblock_->mutable_tombstone_last_logged_opid() = *last_logged_opid; : } : : > I don't think it functionally matters, but I think this probably should go Done PS3, Line 401: RETURN_NOT_OK(meta_->ReplaceSuperBlock(*superblock_)); > Hmm. Let me make sure we have coverage for this case by adding a test. To answer your question, yes, according to the tests and the header doc for TabletMetadata::DeleteTabletData(), if last_logged_opid is passed as boost::none, then the previous value will be retained. But I've added it to the header doc of TSTabletManager::DeleteTabletData() as well. -- To view, visit http://gerrit.cloudera.org:8080/7961 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: Idbc9cecac701c24ebe05759adbbb1fbe2b140506 Gerrit-PatchSet: 3 Gerrit-Project: kudu Gerrit-Branch: master Gerrit-Owner: Mike Percy <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Dan Burkert <[email protected]> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy <[email protected]> Gerrit-Reviewer: Todd Lipcon <[email protected]> Gerrit-HasComments: Yes
