[ https://issues.apache.org/jira/browse/HADOOP-12131?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14604225#comment-14604225 ]
Kengo Seki commented on HADOOP-12131: ------------------------------------- A possible solution is simply numbering each JDK uniquely. Output image is: {code} | Vote | Subsystem | Runtime | Comment ============================================================================ | +1 | mvninstall | 12m 25s | trunk passed | +1 | javac | 1m 46s | trunk passed with JDK v1.7.0_79 (1) | +1 | javac | 1m 44s | trunk passed with JDK v1.7.0_80 (2) | +1 | javac | 1m 46s | trunk passed with JDK v1.7.0_79 (3) | +1 | javadoc | 2m 17s | trunk passed with JDK v1.7.0_79 (1) | +1 | javadoc | 2m 32s | trunk passed with JDK v1.7.0_80 (2) | +1 | javadoc | 2m 17s | trunk passed with JDK v1.7.0_79 (3) | +1 | javac | 1m 41s | the patch passed with JDK v1.7.0_79 (1) | +1 | javac | 1m 43s | the patch passed with JDK v1.7.0_80 (2) | +1 | javac | 1m 41s | the patch passed with JDK v1.7.0_79 (3) | +1 | javadoc | 2m 10s | the patch passed with JDK v1.7.0_79 (1) | +1 | javadoc | 2m 14s | the patch passed with JDK v1.7.0_80 (2) | +1 | javadoc | 2m 10s | the patch passed with JDK v1.7.0_79 (3) | -1 | unit | 19m 58s | hadoop-common in the patch failed with | | | | JDK v1.7.0_79 (1). | -1 | unit | 27m 14s | hadoop-common in the patch failed with | | | | JDK v1.7.0_80 (2). | -1 | unit | 19m 58s | hadoop-common in the patch failed with | | | | JDK v1.7.0_79 (3). Reason | Tests JDK v1.7.0_79 (1) Failed unit tests | hadoop.crypto.TestCryptoCodec JDK v1.7.0_80 (2) Failed unit tests | hadoop.crypto.TestCryptoCodec JDK v1.7.0_79 (3) Failed unit tests | hadoop.crypto.TestCryptoCodec || Subsystem || Report/Notes || ============================================================================ | Default Java | (3) 1.7.0_79 | | Multi-JDK versions | (1) /usr/lib/jvm/java-1.7.0-openjdk-1.7.0.79.x86_64 | | | (2) /usr/lib/jvm/java-1.8.0-openjdk-1.8.0.45-28.b13.el6_6.x86 | | | (3) /usr/java/jdk1.7.0_79 | | unit | /private/tmp/test-patch-hadoop/43110/patch-unit-hadoop-common-project_hadoop-common-jdk1.7.0_79_1.txt | | unit | /private/tmp/test-patch-hadoop/43110/patch-unit-hadoop-common-project_hadoop-common-jdk1.7.0_80_2.txt | | unit | /private/tmp/test-patch-hadoop/43110/patch-unit-hadoop-common-project_hadoop-common-jdk1.7.0_79_3.txt | {code} > The different JDKs which versions are the same aren't distiguished on > multi-JDK mode > ------------------------------------------------------------------------------------ > > Key: HADOOP-12131 > URL: https://issues.apache.org/jira/browse/HADOOP-12131 > Project: Hadoop Common > Issue Type: Sub-task > Components: yetus > Affects Versions: HADOOP-12111 > Reporter: Kengo Seki > > If one installs the same version of Oracle JDK and OpenJDK and enables the > multi-JDK option, output files are overwritten by the last one because the > same value is assigned to $jdk and $fn. Example: > {code} > ============================================================================ > ============================================================================ > Pre-patch trunk javac compilation > ============================================================================ > ============================================================================ > mvn --batch-mode clean compile -DskipTests -Pnative -Drequire.libwebhdfs > -Drequire.snappy -Drequire.openssl -Drequire.fuse -Drequire.test.libhadoop > -Ptest-patch > > /tmp/test-patch-hadoop/25579/branch-javac-hadoop-common-project_hadoop-common-jdk1.7.0_79.txt > 2>&1 > Elapsed: 1m 0s > mvn --batch-mode clean compile -DskipTests -Pnative -Drequire.libwebhdfs > -Drequire.snappy -Drequire.openssl -Drequire.fuse -Drequire.test.libhadoop > -Ptest-patch > > /tmp/test-patch-hadoop/25579/branch-javac-hadoop-common-project_hadoop-common-jdk1.7.0_79.txt > 2>&1 > Elapsed: 0m 54s > {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)