Repository: spark Updated Branches: refs/heads/branch-2.2 154bbc959 -> 768d0b7ce
[SPARK-23489][SQL][TEST][BRANCH-2.2] HiveExternalCatalogVersionsSuite should verify the downloaded file ## What changes were proposed in this pull request? This is a backport of #21210 because `branch-2.2` also faces the same failures. Although [SPARK-22654](https://issues.apache.org/jira/browse/SPARK-22654) made `HiveExternalCatalogVersionsSuite` download from Apache mirrors three times, it has been flaky because it didn't verify the downloaded file. Some Apache mirrors terminate the downloading abnormally, the *corrupted* file shows the following errors. ``` gzip: stdin: not in gzip format tar: Child returned status 1 tar: Error is not recoverable: exiting now 22:46:32.700 WARN org.apache.spark.sql.hive.HiveExternalCatalogVersionsSuite: ===== POSSIBLE THREAD LEAK IN SUITE o.a.s.sql.hive.HiveExternalCatalogVersionsSuite, thread names: Keep-Alive-Timer ===== *** RUN ABORTED *** java.io.IOException: Cannot run program "./bin/spark-submit" (in directory "/tmp/test-spark/spark-2.2.0"): error=2, No such file or directory ``` This has been reported weirdly in two ways. For example, the above case is reported as Case 2 `no failures`. - Case 1. [Test Result (1 failure / +1)](https://amplab.cs.berkeley.edu/jenkins/view/Spark%20QA%20Test%20(Dashboard)/job/spark-master-test-sbt-hadoop-2.7/4389/) - Case 2. [Test Result (no failures)](https://amplab.cs.berkeley.edu/jenkins/view/Spark%20QA%20Test%20(Dashboard)/job/spark-master-test-maven-hadoop-2.6/4811/) This PR aims to make `HiveExternalCatalogVersionsSuite` more robust by verifying the downloaded `tgz` file by extracting and checking the existence of `bin/spark-submit`. If it turns out that the file is empty or corrupted, `HiveExternalCatalogVersionsSuite` will do retry logic like the download failure. ## How was this patch tested? Pass the Jenkins. Author: Dongjoon Hyun <dongj...@apache.org> Closes #21232 from dongjoon-hyun/SPARK-23489-2. Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/768d0b7c Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/768d0b7c Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/768d0b7c Branch: refs/heads/branch-2.2 Commit: 768d0b7cecb8716d517279df2228a404681e4f95 Parents: 154bbc9 Author: Dongjoon Hyun <dongj...@apache.org> Authored: Thu May 3 17:10:15 2018 -0700 Committer: gatorsmile <gatorsm...@gmail.com> Committed: Thu May 3 17:10:15 2018 -0700 ---------------------------------------------------------------------- .../hive/HiveExternalCatalogVersionsSuite.scala | 35 ++++++++++---------- 1 file changed, 18 insertions(+), 17 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/768d0b7c/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveExternalCatalogVersionsSuite.scala ---------------------------------------------------------------------- diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveExternalCatalogVersionsSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveExternalCatalogVersionsSuite.scala index a3d5b94..2b37047 100644 --- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveExternalCatalogVersionsSuite.scala +++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveExternalCatalogVersionsSuite.scala @@ -57,30 +57,31 @@ class HiveExternalCatalogVersionsSuite extends SparkSubmitTestUtils { for (i <- 0 until 3) { val preferredMirror = Seq("wget", "https://www.apache.org/dyn/closer.lua?preferred=true", "-q", "-O", "-").!!.trim - val url = s"$preferredMirror/spark/spark-$version/spark-$version-bin-hadoop2.7.tgz" + val filename = s"spark-$version-bin-hadoop2.7.tgz" + val url = s"$preferredMirror/spark/spark-$version/$filename" logInfo(s"Downloading Spark $version from $url") if (Seq("wget", url, "-q", "-P", path).! == 0) { - return + val downloaded = new File(sparkTestingDir, filename).getCanonicalPath + val targetDir = new File(sparkTestingDir, s"spark-$version").getCanonicalPath + + Seq("mkdir", targetDir).! + val exitCode = Seq("tar", "-xzf", downloaded, "-C", targetDir, "--strip-components=1").! + Seq("rm", downloaded).! + + // For a corrupted file, `tar` returns non-zero values. However, we also need to check + // the extracted file because `tar` returns 0 for empty file. + val sparkSubmit = new File(sparkTestingDir, s"spark-$version/bin/spark-submit") + if (exitCode == 0 && sparkSubmit.exists()) { + return + } else { + Seq("rm", "-rf", targetDir).! + } } logWarning(s"Failed to download Spark $version from $url") } fail(s"Unable to download Spark $version") } - - private def downloadSpark(version: String): Unit = { - tryDownloadSpark(version, sparkTestingDir.getCanonicalPath) - - val downloaded = new File(sparkTestingDir, s"spark-$version-bin-hadoop2.7.tgz").getCanonicalPath - val targetDir = new File(sparkTestingDir, s"spark-$version").getCanonicalPath - - Seq("mkdir", targetDir).! - - Seq("tar", "-xzf", downloaded, "-C", targetDir, "--strip-components=1").! - - Seq("rm", downloaded).! - } - private def genDataDir(name: String): String = { new File(tmpDataDir, name).getCanonicalPath } @@ -125,7 +126,7 @@ class HiveExternalCatalogVersionsSuite extends SparkSubmitTestUtils { PROCESS_TABLES.testingVersions.zipWithIndex.foreach { case (version, index) => val sparkHome = new File(sparkTestingDir, s"spark-$version") if (!sparkHome.exists()) { - downloadSpark(version) + tryDownloadSpark(version, sparkTestingDir.getCanonicalPath) } val args = Seq( --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org