[SPARK-22344][SPARKR] Set java.io.tmpdir for SparkR tests

This PR sets the java.io.tmpdir for CRAN checks and also disables the hsperfdata for the JVM when running CRAN checks. Together this prevents files from being left behind in `/tmp` ## How was this patch tested? Tested manually on a clean EC2 machine Author: Shivaram Venkataraman <shivaram@cs.berkeley.edu> Closes #19589 from shivaram/sparkr-tmpdir-clean.
apache · Oct 30, 2017 · 1fe2761 · 1fe2761
1 parent 659acf1
commit 1fe2761
Show file tree

Hide file tree

Showing 3 changed files with 20 additions and 3 deletions.
diff --git a/R/pkg/inst/tests/testthat/test_basic.R b/R/pkg/inst/tests/testthat/test_basic.R
@@ -18,7 +18,8 @@
 context("basic tests for CRAN")
 
 test_that("create DataFrame from list or data.frame", {
-  sparkR.session(master = sparkRTestMaster, enableHiveSupport = FALSE)
+  sparkR.session(master = sparkRTestMaster, enableHiveSupport = FALSE,
+                 sparkConfig = sparkRTestConfig)
 
   i <- 4
   df <- createDataFrame(data.frame(dummy = 1:i))
@@ -49,7 +50,8 @@ test_that("create DataFrame from list or data.frame", {
 })
 
 test_that("spark.glm and predict", {
-  sparkR.session(master = sparkRTestMaster, enableHiveSupport = FALSE)
+  sparkR.session(master = sparkRTestMaster, enableHiveSupport = FALSE,
+                 sparkConfig = sparkRTestConfig)
 
   training <- suppressWarnings(createDataFrame(iris))
   # gaussian family

diff --git a/R/pkg/tests/run-all.R b/R/pkg/tests/run-all.R
@@ -36,8 +36,17 @@ invisible(lapply(sparkRWhitelistSQLDirs,
 sparkRFilesBefore <- list.files(path = sparkRDir, all.files = TRUE)
 
 sparkRTestMaster <- "local[1]"
+sparkRTestConfig <- list()
 if (identical(Sys.getenv("NOT_CRAN"), "true")) {
   sparkRTestMaster <- ""
+} else {
+  # Disable hsperfdata on CRAN
+  old_java_opt <- Sys.getenv("_JAVA_OPTIONS")
+  Sys.setenv("_JAVA_OPTIONS" = paste("-XX:-UsePerfData", old_java_opt))
+  tmpDir <- tempdir()
+  tmpArg <- paste0("-Djava.io.tmpdir=", tmpDir)
+  sparkRTestConfig <- list(spark.driver.extraJavaOptions = tmpArg,
+                            spark.executor.extraJavaOptions = tmpArg)
 }
 
 test_package("SparkR")

diff --git a/R/pkg/vignettes/sparkr-vignettes.Rmd b/R/pkg/vignettes/sparkr-vignettes.Rmd
@@ -36,6 +36,12 @@ opts_hooks$set(eval = function(options) {
   }
   options
 })
+r_tmp_dir <- tempdir()
+tmp_arg <- paste("-Djava.io.tmpdir=", r_tmp_dir, sep = "")
+sparkSessionConfig <- list(spark.driver.extraJavaOptions = tmp_arg,
+                           spark.executor.extraJavaOptions = tmp_arg)
+old_java_opt <- Sys.getenv("_JAVA_OPTIONS")
+Sys.setenv("_JAVA_OPTIONS" = paste("-XX:-UsePerfData", old_java_opt, sep = " "))
 ```
 
 ## Overview
@@ -57,7 +63,7 @@ We use default settings in which it runs in local mode. It auto downloads Spark
 
 ```{r, include=FALSE}
 install.spark()
-sparkR.session(master = "local[1]")
+sparkR.session(master = "local[1]", sparkConfig = sparkSessionConfig, enableHiveSupport = FALSE)
 ```
 ```{r, eval=FALSE}
 sparkR.session()