Dan Burkert has posted comments on this change. Change subject: WIP: Kudu Jepsen Tests - Initial Commit ......................................................................
Patch Set 4: (15 comments) http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/.gitignore File java/kudu-jepsen/.gitignore: Line 1: /target I think this shouldn't be necessary since the java/.gitignore contains target/. http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/CHANGELOG.md File java/kudu-jepsen/CHANGELOG.md: Line 1: # Change Log can probably be removed? http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/LICENSE File java/kudu-jepsen/LICENSE: Line 1: THE ACCOMPANYING PROGRAM IS PROVIDED UNDER THE TERMS OF THIS ECLIPSE PUBLIC Remove http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/doc/intro.md File java/kudu-jepsen/doc/intro.md: Line 1: # Introduction to jepsen.kudu remove http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/pom.xml File java/kudu-jepsen/pom.xml: Line 56: <exclusions> Could you document the reason for this exclusion? I wouldn't expect the client to work without the slf4j-api jar being on the classpath. http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/project.clj File java/kudu-jepsen/project.clj: PS1, Line 4: Eclipse Public License ASL http://gerrit.cloudera.org:8080/#/c/5492/4/java/kudu-jepsen/project.clj File java/kudu-jepsen/project.clj: Line 1: (defproject kudu "0.1.0-SNAPSHOT" Is this file necessary? Seems like we should have either the pom.xml or project.clj, not both. http://gerrit.cloudera.org:8080/#/c/5492/1/java/kudu-jepsen/src/main/clojure/jepsen/kudu.clj File java/kudu-jepsen/src/main/clojure/jepsen/kudu.clj: Line 17: ;; TODO allow to replace the binaries with locally built ones Yah, we should try and use the built in MiniCluster here, if possible. Would simplify this a lot. PS1, Line 54: whitespace http://gerrit.cloudera.org:8080/#/c/5492/4/java/kudu-jepsen/src/main/clojure/jepsen/kudu.clj File java/kudu-jepsen/src/main/clojure/jepsen/kudu.clj: Line 32: (when (> (count (:masters test)) 1) Why only set the master addresses when there are more than one master? This whole thing could be simplified to look more like kudu-cfg-tserver. PS4, Line 54: trailing whitespace PS4, Line 59: into [] I don't think the 'into []' is necessary, str/join should work fine over a seq. user=> (def foo ["foo" "bar"]) #'user/foo user=> (def bar ["baz" "bazzle"]) #'user/bar user=> (clojure.string/join "\n" (concat foo bar)) "foo\nbar\nbaz\nbazzle" PS4, Line 59: efn this could be a def PS4, Line 63: into [] Same. http://gerrit.cloudera.org:8080/#/c/5492/4/java/kudu-jepsen/src/main/clojure/jepsen/kudu/table.clj File java/kudu-jepsen/src/main/clojure/jepsen/kudu/table.clj: Line 40: (defn kv-table-options Any reason not to use hash partitioning? Would make this a lot easier, and it's more realistic for a K/V workload. -- To view, visit http://gerrit.cloudera.org:8080/5492 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I590c6e78840304b3131666c7037ff9a08dc77dea Gerrit-PatchSet: 4 Gerrit-Project: kudu Gerrit-Branch: master Gerrit-Owner: David Ribeiro Alves <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Dan Burkert <[email protected]> Gerrit-Reviewer: Jean-Daniel Cryans <[email protected]> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon <[email protected]> Gerrit-HasComments: Yes
