Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/12488 )
Change subject: KUDU-2701 Fix compaction loop due to using wrong rowset size ...................................................................... Patch Set 1: (3 comments) Just passing through; hope you get http://gerrit.cloudera.org:8080/#/c/12488/1//COMMIT_MSG Commit Message: http://gerrit.cloudera.org:8080/#/c/12488/1//COMMIT_MSG@14 PS1, Line 14: the size of the whole rowset should be used Can you be more explicit about which missing pieces are critical? Is it the bloom? The adhoc index (in the case of a composite primary key)? The UNDOs? http://gerrit.cloudera.org:8080/#/c/12488/1//COMMIT_MSG@15 PS1, Line 15: values Nit: this is used as a verb, right? Took a while to parse the sentence; maybe use a different verb that doesn't sound so much like a noun? http://gerrit.cloudera.org:8080/#/c/12488/1//COMMIT_MSG@19 PS1, Line 19: For example, I discovered a case where a tablet : had 8 rowsets of "size" 16MB. In reality, this size was the base data : size plus REDOs. Small rowset compaction policy determined that it would : be good to compact these 8 rowsets, believing that it would produce 4 : rowsets of size 32MB. However, the rowsets were actually 32MB in size, : and compacting them produced 8 rowsets of size 32MB identical to the : previous 8, and therefore 8 rowsets that appeared to be of size 16MB to : compaction policy. Thus these 8 were chosen to be compacted, and so : on... Would be nice to create a test case that models this and shows that compaction won't loop anymore. -- To view, visit http://gerrit.cloudera.org:8080/12488 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I21b7ff6333137aaf1e98ef4849691dd08e24e007 Gerrit-Change-Number: 12488 Gerrit-PatchSet: 1 Gerrit-Owner: Will Berkeley <[email protected]> Gerrit-Reviewer: Adar Dembo <[email protected]> Gerrit-Reviewer: Andrew Wong <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Comment-Date: Thu, 14 Feb 2019 23:33:14 +0000 Gerrit-HasComments: Yes
