## What changes were proposed in this pull request?
TaskResultGetter tries to deserialize the TaskEndReason before handling the failed task. If an error is thrown during deserialization, the failed task won't be handled, which leaves the job hanging.
The PR proposes to handle the failed task in a finally block.
## How was this patch tested?
In my case I hit a NoClassDefFoundError and the job hangs. Manually verified the patch can fix it.
Author: Rui Li <rui.li@intel.com>
Author: Rui Li <lirui@apache.org>
Author: Rui Li <shlr@cn.ibm.com>
Closes#12775 from lirui-intel/SPARK-14958.
This commit changes Utils.writeByteBuffer so that it does not change
the position of the ByteBuffer that it writes out, and adds a unit test for
this functionality.
cc mridulm
Author: Kay Ousterhout <kayousterhout@gmail.com>
Closes#16462 from kayousterhout/SPARK-19062.
## What changes were proposed in this pull request?
There are many locations in the Spark repo where the same word occurs consecutively. Sometimes they are appropriately placed, but many times they are not. This PR removes the inappropriately duplicated words.
## How was this patch tested?
N/A since only docs or comments were updated.
Author: Niranjan Padmanabhan <niranjan.padmanabhan@gmail.com>
Closes#16455 from neurons/np.structure_streaming_doc.
## What changes were proposed in this pull request?
Add `finally` clause for `sc.stop()` in the `test("register and deregister Spark listener from SparkContext")`.
## How was this patch tested?
Pass the build and unit tests.
Author: Weiqing Yang <yangweiqing001@gmail.com>
Closes#16426 from weiqingy/testIssue.
## What changes were proposed in this pull request?
This is to workaround an implicit result of #4947 which suppressed the
original Kryo exception if the overflow happened during serialization.
## How was this patch tested?
`KryoSerializerSuite` was augmented to reflect this change.
Author: Sergei Lebedev <superbobry@gmail.com>
Closes#16416 from superbobry/patch-1.
## What changes were proposed in this pull request?
The root cause of this issue is that RegisterWorkerResponse and LaunchExecutor are sent via two different channels (TCP connections) and their order is not guaranteed.
This PR changes the master and worker codes to use `workerRef` to send RegisterWorkerResponse, so that RegisterWorkerResponse and LaunchExecutor are sent via the same connection. Hence `LaunchExecutor` will always be after `RegisterWorkerResponse` and never be ignored.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#16345 from zsxwing/SPARK-17755.
## What changes were proposed in this pull request?
In current Spark we could add customized SparkListener through `SparkContext#addListener` API, but there's no equivalent API to remove the registered one. In our scenario SparkListener will be added repeatedly accordingly to the changed environment. If lacks the ability to remove listeners, there might be many registered listeners finally, this is unnecessary and potentially affects the performance. So here propose to add an API to remove registered listener.
## How was this patch tested?
Add an unit test to verify it.
Author: jerryshao <sshao@hortonworks.com>
Closes#16382 from jerryshao/SPARK-18975.
## What changes were proposed in this pull request?
There are several tests failing due to resource-closing-related and path-related problems on Windows as below.
- `SQLQuerySuite`:
```
- specifying database name for a temporary table is not allowed *** FAILED *** (125 milliseconds)
org.apache.spark.sql.AnalysisException: Path does not exist: file:/C:projectsspark arget mpspark-1f4471ab-aac0-4239-ae35-833d54b37e52;
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:382)
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:370)
```
- `JsonSuite`:
```
- Loading a JSON dataset from a text file with SQL *** FAILED *** (94 milliseconds)
org.apache.spark.sql.AnalysisException: Path does not exist: file:/C:projectsspark arget mpspark-c918a8b7-fc09-433c-b9d0-36c0f78ae918;
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:382)
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:370)
```
- `StateStoreSuite`:
```
- SPARK-18342: commit fails when rename fails *** FAILED *** (16 milliseconds)
java.lang.IllegalArgumentException: java.net.URISyntaxException: Relative path in absolute URI: StateStoreSuite29777261fs://C:%5Cprojects%5Cspark%5Ctarget%5Ctmp%5Cspark-ef349862-7281-4963-aaf3-add0d670a4ad%5C?????-2218c2f8-2cf6-4f80-9cdf-96354e8246a77685899733421033312/0
at org.apache.hadoop.fs.Path.initialize(Path.java:206)
at org.apache.hadoop.fs.Path.<init>(Path.java:116)
at org.apache.hadoop.fs.Path.<init>(Path.java:89)
...
Cause: java.net.URISyntaxException: Relative path in absolute URI: StateStoreSuite29777261fs://C:%5Cprojects%5Cspark%5Ctarget%5Ctmp%5Cspark-ef349862-7281-4963-aaf3-add0d670a4ad%5C?????-2218c2f8-2cf6-4f80-9cdf-96354e8246a77685899733421033312/0
at java.net.URI.checkPath(URI.java:1823)
at java.net.URI.<init>(URI.java:745)
at org.apache.hadoop.fs.Path.initialize(Path.java:203)
```
- `HDFSMetadataLogSuite`:
```
- FileManager: FileContextManager *** FAILED *** (94 milliseconds)
java.io.IOException: Failed to delete: C:\projects\spark\target\tmp\spark-415bb0bd-396b-444d-be82-04599e025f21
at org.apache.spark.util.Utils$.deleteRecursively(Utils.scala:1010)
at org.apache.spark.sql.test.SQLTestUtils$class.withTempDir(SQLTestUtils.scala:127)
at org.apache.spark.sql.execution.streaming.HDFSMetadataLogSuite.withTempDir(HDFSMetadataLogSuite.scala:38)
- FileManager: FileSystemManager *** FAILED *** (78 milliseconds)
java.io.IOException: Failed to delete: C:\projects\spark\target\tmp\spark-ef8222cd-85aa-47c0-a396-bc7979e15088
at org.apache.spark.util.Utils$.deleteRecursively(Utils.scala:1010)
at org.apache.spark.sql.test.SQLTestUtils$class.withTempDir(SQLTestUtils.scala:127)
at org.apache.spark.sql.execution.streaming.HDFSMetadataLogSuite.withTempDir(HDFSMetadataLogSuite.scala:38)
```
And, there are some tests being failed due to the length limitation on cmd in Windows as below:
- `LauncherBackendSuite`:
```
- local: launcher handle *** FAILED *** (30 seconds, 120 milliseconds)
The code passed to eventually never returned normally. Attempted 283 times over 30.0960053 seconds. Last failure message: The reference was null. (LauncherBackendSuite.scala:56)
org.scalatest.exceptions.TestFailedDueToTimeoutException:
at org.scalatest.concurrent.Eventually$class.tryTryAgain$1(Eventually.scala:420)
at org.scalatest.concurrent.Eventually$class.eventually(Eventually.scala:438)
- standalone/client: launcher handle *** FAILED *** (30 seconds, 47 milliseconds)
The code passed to eventually never returned normally. Attempted 282 times over 30.037987100000002 seconds. Last failure message: The reference was null. (LauncherBackendSuite.scala:56)
org.scalatest.exceptions.TestFailedDueToTimeoutException:
at org.scalatest.concurrent.Eventually$class.tryTryAgain$1(Eventually.scala:420)
at org.scalatest.concurrent.Eventually$class.eventually(Eventually.scala:438)
```
The executed command is, https://gist.github.com/HyukjinKwon/d3fdd2e694e5c022992838a618a516bd, which is 16K length; however, the length limitation is 8K on Windows. So, it is being failed to launch.
This PR proposes to fix the test failures on Windows and skip the tests failed due to the length limitation
## How was this patch tested?
Manually tested via AppVeyor
**Before**
`SQLQuerySuite `: https://ci.appveyor.com/project/spark-test/spark/build/306-pr-references
`JsonSuite`: https://ci.appveyor.com/project/spark-test/spark/build/307-pr-references
`StateStoreSuite` : https://ci.appveyor.com/project/spark-test/spark/build/305-pr-references
`HDFSMetadataLogSuite`: https://ci.appveyor.com/project/spark-test/spark/build/304-pr-references
`LauncherBackendSuite`: https://ci.appveyor.com/project/spark-test/spark/build/303-pr-references
**After**
`SQLQuerySuite`: https://ci.appveyor.com/project/spark-test/spark/build/293-SQLQuerySuite
`JsonSuite`: https://ci.appveyor.com/project/spark-test/spark/build/294-JsonSuite
`StateStoreSuite`: https://ci.appveyor.com/project/spark-test/spark/build/297-StateStoreSuite
`HDFSMetadataLogSuite`: https://ci.appveyor.com/project/spark-test/spark/build/319-pr-references
`LauncherBackendSuite`: failed test skipped.
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16335 from HyukjinKwon/more-fixes-on-windows.
## What changes were proposed in this pull request?
Spark's current task cancellation / task killing mechanism is "best effort" because some tasks may not be interruptible or may not respond to their "killed" flags being set. If a significant fraction of a cluster's task slots are occupied by tasks that have been marked as killed but remain running then this can lead to a situation where new jobs and tasks are starved of resources that are being used by these zombie tasks.
This patch aims to address this problem by adding a "task reaper" mechanism to executors. At a high-level, task killing now launches a new thread which attempts to kill the task and then watches the task and periodically checks whether it has been killed. The TaskReaper will periodically re-attempt to call `TaskRunner.kill()` and will log warnings if the task keeps running. I modified TaskRunner to rename its thread at the start of the task, allowing TaskReaper to take a thread dump and filter it in order to log stacktraces from the exact task thread that we are waiting to finish. If the task has not stopped after a configurable timeout then the TaskReaper will throw an exception to trigger executor JVM death, thereby forcibly freeing any resources consumed by the zombie tasks.
This feature is flagged off by default and is controlled by four new configurations under the `spark.task.reaper.*` namespace. See the updated `configuration.md` doc for details.
## How was this patch tested?
Tested via a new test case in `JobCancellationSuite`, plus manual testing.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#16189 from JoshRosen/cancellation.
## What changes were proposed in this pull request?
Right now we serialize the empty task metrics once per task – Since this is shared across all tasks we could use the same serialized task metrics across all tasks of a stage.
## How was this patch tested?
- [x] Run tests on EC2 to measure performance improvement
Author: Shivaram Venkataraman <shivaram@cs.berkeley.edu>
Closes#16261 from shivaram/task-metrics-one-copy.
## What changes were proposed in this pull request?
`NoSuchElementException` will throw since https://github.com/apache/spark/pull/15056 if a broadcast cannot cache in memory. The reason is that that change cannot cover `!unrolled.hasNext` in `next()` function.
This change is to cover the `!unrolled.hasNext` and check `hasNext` before calling `next` in `blockManager.getLocalValues` to make it more robust.
We can cache and read broadcast even it cannot fit in memory from this pull request.
Exception log:
```
16/12/10 10:10:04 INFO UnifiedMemoryManager: Will not store broadcast_131 as the required space (1048576 bytes) exceeds our memory limit (122764 bytes)
16/12/10 10:10:04 WARN MemoryStore: Failed to reserve initial memory threshold of 1024.0 KB for computing block broadcast_131 in memory.
16/12/10 10:10:04 WARN MemoryStore: Not enough space to cache broadcast_131 in memory! (computed 384.0 B so far)
16/12/10 10:10:04 INFO MemoryStore: Memory use = 95.6 KB (blocks) + 0.0 B (scratch space shared across 0 tasks(s)) = 95.6 KB. Storage limit = 119.9 KB.
16/12/10 10:10:04 ERROR Utils: Exception encountered
java.util.NoSuchElementException
at org.apache.spark.util.collection.PrimitiveVector$$anon$1.next(PrimitiveVector.scala:58)
at org.apache.spark.storage.memory.PartiallyUnrolledIterator.next(MemoryStore.scala:700)
at org.apache.spark.util.CompletionIterator.next(CompletionIterator.scala:30)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1$$anonfun$2.apply(TorrentBroadcast.scala:210)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1$$anonfun$2.apply(TorrentBroadcast.scala:210)
at scala.Option.map(Option.scala:146)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1.apply(TorrentBroadcast.scala:210)
at org.apache.spark.util.Utils$.tryOrIOException(Utils.scala:1269)
at org.apache.spark.broadcast.TorrentBroadcast.readBroadcastBlock(TorrentBroadcast.scala:206)
at org.apache.spark.broadcast.TorrentBroadcast._value$lzycompute(TorrentBroadcast.scala:66)
at org.apache.spark.broadcast.TorrentBroadcast._value(TorrentBroadcast.scala:66)
at org.apache.spark.broadcast.TorrentBroadcast.getValue(TorrentBroadcast.scala:96)
at org.apache.spark.broadcast.Broadcast.value(Broadcast.scala:70)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:86)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:53)
at org.apache.spark.scheduler.Task.run(Task.scala:108)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:282)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
16/12/10 10:10:04 ERROR Executor: Exception in task 1.0 in stage 86.0 (TID 134423)
java.io.IOException: java.util.NoSuchElementException
at org.apache.spark.util.Utils$.tryOrIOException(Utils.scala:1276)
at org.apache.spark.broadcast.TorrentBroadcast.readBroadcastBlock(TorrentBroadcast.scala:206)
at org.apache.spark.broadcast.TorrentBroadcast._value$lzycompute(TorrentBroadcast.scala:66)
at org.apache.spark.broadcast.TorrentBroadcast._value(TorrentBroadcast.scala:66)
at org.apache.spark.broadcast.TorrentBroadcast.getValue(TorrentBroadcast.scala:96)
at org.apache.spark.broadcast.Broadcast.value(Broadcast.scala:70)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:86)
at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:53)
at org.apache.spark.scheduler.Task.run(Task.scala:108)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:282)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.util.NoSuchElementException
at org.apache.spark.util.collection.PrimitiveVector$$anon$1.next(PrimitiveVector.scala:58)
at org.apache.spark.storage.memory.PartiallyUnrolledIterator.next(MemoryStore.scala:700)
at org.apache.spark.util.CompletionIterator.next(CompletionIterator.scala:30)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1$$anonfun$2.apply(TorrentBroadcast.scala:210)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1$$anonfun$2.apply(TorrentBroadcast.scala:210)
at scala.Option.map(Option.scala:146)
at org.apache.spark.broadcast.TorrentBroadcast$$anonfun$readBroadcastBlock$1.apply(TorrentBroadcast.scala:210)
at org.apache.spark.util.Utils$.tryOrIOException(Utils.scala:1269)
... 12 more
```
## How was this patch tested?
Add unit test
Author: Yuming Wang <wgyumg@gmail.com>
Closes#16252 from wangyum/SPARK-18827.
## What changes were proposed in this pull request?
There are several tests failing due to resource-closing-related and path-related problems on Windows as below.
- `RPackageUtilsSuite`:
```
- build an R package from a jar end to end *** FAILED *** (1 second, 625 milliseconds)
java.io.IOException: Unable to delete file: C:\projects\spark\target\tmp\1481729427517-0\a\dep2\d\dep2-d.jar
at org.apache.commons.io.FileUtils.forceDelete(FileUtils.java:2279)
at org.apache.commons.io.FileUtils.cleanDirectory(FileUtils.java:1653)
at org.apache.commons.io.FileUtils.deleteDirectory(FileUtils.java:1535)
- faulty R package shows documentation *** FAILED *** (359 milliseconds)
java.io.IOException: Unable to delete file: C:\projects\spark\target\tmp\1481729428970-0\dep1-c.jar
at org.apache.commons.io.FileUtils.forceDelete(FileUtils.java:2279)
at org.apache.commons.io.FileUtils.cleanDirectory(FileUtils.java:1653)
at org.apache.commons.io.FileUtils.deleteDirectory(FileUtils.java:1535)
- SparkR zipping works properly *** FAILED *** (47 milliseconds)
java.util.regex.PatternSyntaxException: Unknown character property name {r} near index 4
C:\projects\spark\target\tmp\1481729429282-0
^
at java.util.regex.Pattern.error(Pattern.java:1955)
at java.util.regex.Pattern.charPropertyNodeFor(Pattern.java:2781)
```
- `InputOutputMetricsSuite`:
```
- input metrics for old hadoop with coalesce *** FAILED *** (240 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics with cache and coalesce *** FAILED *** (109 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics for new Hadoop API with coalesce *** FAILED *** (0 milliseconds)
java.lang.IllegalArgumentException: Wrong FS: file://C:\projects\spark\target\tmp\spark-9366ec94-dac7-4a5c-a74b-3e7594a692ab\test\InputOutputMetricsSuite.txt, expected: file:///
at org.apache.hadoop.fs.FileSystem.checkPath(FileSystem.java:642)
at org.apache.hadoop.fs.FileSystem.makeQualified(FileSystem.java:462)
at org.apache.hadoop.fs.FilterFileSystem.makeQualified(FilterFileSystem.java:114)
- input metrics when reading text file *** FAILED *** (110 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics on records read - simple *** FAILED *** (125 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics on records read - more stages *** FAILED *** (110 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics on records - New Hadoop API *** FAILED *** (16 milliseconds)
java.lang.IllegalArgumentException: Wrong FS: file://C:\projects\spark\target\tmp\spark-3f10a1a4-7820-4772-b821-25fd7523bf6f\test\InputOutputMetricsSuite.txt, expected: file:///
at org.apache.hadoop.fs.FileSystem.checkPath(FileSystem.java:642)
at org.apache.hadoop.fs.FileSystem.makeQualified(FileSystem.java:462)
at org.apache.hadoop.fs.FilterFileSystem.makeQualified(FilterFileSystem.java:114)
- input metrics on records read with cache *** FAILED *** (93 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input read/write and shuffle read/write metrics all line up *** FAILED *** (93 milliseconds)
java.io.IOException: Not a file: file:/C:/projects/spark/core/ignored
at org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
at org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:202)
at org.apache.spark.rdd.RDD$$anonfun$partitions$2.apply(RDD.scala:252)
- input metrics with interleaved reads *** FAILED *** (0 milliseconds)
java.lang.IllegalArgumentException: Wrong FS: file://C:\projects\spark\target\tmp\spark-2638d893-e89b-47ce-acd0-bbaeee78dd9b\InputOutputMetricsSuite_cart.txt, expected: file:///
at org.apache.hadoop.fs.FileSystem.checkPath(FileSystem.java:642)
at org.apache.hadoop.fs.FileSystem.makeQualified(FileSystem.java:462)
at org.apache.hadoop.fs.FilterFileSystem.makeQualified(FilterFileSystem.java:114)
- input metrics with old CombineFileInputFormat *** FAILED *** (157 milliseconds)
17947 was not greater than or equal to 300000 (InputOutputMetricsSuite.scala:324)
org.scalatest.exceptions.TestFailedException:
at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
at org.scalatest.FunSuite.newAssertionFailedException(FunSuite.scala:1555)
at org.scalatest.Assertions$AssertionsHelper.macroAssert(Assertions.scala:466)
- input metrics with new CombineFileInputFormat *** FAILED *** (16 milliseconds)
java.lang.IllegalArgumentException: Wrong FS: file://C:\projects\spark\target\tmp\spark-11920c08-19d8-4c7c-9fba-28ed72b79f80\test\InputOutputMetricsSuite.txt, expected: file:///
at org.apache.hadoop.fs.FileSystem.checkPath(FileSystem.java:642)
at org.apache.hadoop.fs.FileSystem.makeQualified(FileSystem.java:462)
at org.apache.hadoop.fs.FilterFileSystem.makeQualified(FilterFileSystem.java:114)
```
- `ReplayListenerSuite`:
```
- End-to-end replay *** FAILED *** (121 milliseconds)
java.io.IOException: No FileSystem for scheme: C
at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2421)
at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2428)
- End-to-end replay with compression *** FAILED *** (516 milliseconds)
java.io.IOException: No FileSystem for scheme: C
at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2421)
at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2428)
at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:88)
```
- `EventLoggingListenerSuite`:
```
- End-to-end event logging *** FAILED *** (7 seconds, 435 milliseconds)
java.io.IOException: No FileSystem for scheme: C
at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2421)
at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2428)
at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:88)
- End-to-end event logging with compression *** FAILED *** (1 second)
java.io.IOException: No FileSystem for scheme: C
at org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2421)
at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2428)
at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:88)
- Event log name *** FAILED *** (16 milliseconds)
"file:/[]base-dir/app1" did not equal "file:/[C:/]base-dir/app1" (EventLoggingListenerSuite.scala:123)
org.scalatest.exceptions.TestFailedException:
at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
at org.scalatest.FunSuite.newAssertionFailedException(FunSuite.scala:1555)
at org.scalatest.Assertions$AssertionsHelper.macroAssert(Assertions.scala:466)
```
This PR proposes to fix the test failures on Windows
## How was this patch tested?
Manually tested via AppVeyor
**Before**
`RPackageUtilsSuite`: https://ci.appveyor.com/project/spark-test/spark/build/273-RPackageUtilsSuite-before
`InputOutputMetricsSuite`: https://ci.appveyor.com/project/spark-test/spark/build/272-InputOutputMetricsSuite-before
`ReplayListenerSuite`: https://ci.appveyor.com/project/spark-test/spark/build/274-ReplayListenerSuite-before
`EventLoggingListenerSuite`: https://ci.appveyor.com/project/spark-test/spark/build/275-EventLoggingListenerSuite-before
**After**
`RPackageUtilsSuite`: https://ci.appveyor.com/project/spark-test/spark/build/270-RPackageUtilsSuite
`InputOutputMetricsSuite`: https://ci.appveyor.com/project/spark-test/spark/build/271-InputOutputMetricsSuite
`ReplayListenerSuite`: https://ci.appveyor.com/project/spark-test/spark/build/277-ReplayListenerSuite-after
`EventLoggingListenerSuite`: https://ci.appveyor.com/project/spark-test/spark/build/278-EventLoggingListenerSuite-after
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16305 from HyukjinKwon/RPackageUtilsSuite-InputOutputMetricsSuite.
## What changes were proposed in this pull request?
This builds upon the blacklisting introduced in SPARK-17675 to add blacklisting of executors and nodes for an entire Spark application. Resources are blacklisted based on tasks that fail, in tasksets that eventually complete successfully; they are automatically returned to the pool of active resources based on a timeout. Full details are available in a design doc attached to the jira.
## How was this patch tested?
Added unit tests, ran them via Jenkins, also ran a handful of them in a loop to check for flakiness.
The added tests include:
- verifying BlacklistTracker works correctly
- verifying TaskSchedulerImpl interacts with BlacklistTracker correctly (via a mock BlacklistTracker)
- an integration test for the entire scheduler with blacklisting in a few different scenarios
Author: Imran Rashid <irashid@cloudera.com>
Author: mwws <wei.mao@intel.com>
Closes#14079 from squito/blacklist-SPARK-8425.
## What changes were proposed in this pull request?
This PR proposes to fix the tests failed on Windows as below:
```
[info] - pipe with empty partition *** FAILED *** (672 milliseconds)
[info] Set(0, 4, 5) did not equal Set(0, 5, 6) (PipedRDDSuite.scala:145)
[info] org.scalatest.exceptions.TestFailedException:
...
```
In this case, `wc -c` counts the characters on both Windows and Linux but the newlines characters on Windows are `\r\n` which are two. So, the counts ends up one more for each.
```
[info] - test pipe exports map_input_file *** FAILED *** (62 milliseconds)
[info] java.lang.IllegalStateException: Subprocess exited with status 1. Command ran: printenv map_input_file
[info] at org.apache.spark.rdd.PipedRDD$$anon$1.hasNext(PipedRDD.scala:178)
...
```
```
[info] - test pipe exports mapreduce_map_input_file *** FAILED *** (172 milliseconds)
[info] java.lang.IllegalStateException: Subprocess exited with status 1. Command ran: printenv mapreduce_map_input_file
[info] at org.apache.spark.rdd.PipedRDD$$anon$1.hasNext(PipedRDD.scala:178)
...
```
`printenv` command prints the environment variables; however, when environment variables are set to `ProcessBuilder` as lower-cased keys, `printenv` in Windows ignores and does not print this although it is actually set and accessible. (this was tested in [here](https://ci.appveyor.com/project/spark-test/spark/build/208-PipedRDDSuite) for upper-cases with this [diff](https://github.com/apache/spark/compare/master...spark-test:74d39da) and [here](https://ci.appveyor.com/project/spark-test/spark/build/203-PipedRDDSuite) for lower-cases with this [diff](https://github.com/apache/spark/compare/master...spark-test:fde5e37f28032c15a8d8693ba033a8a779a26317). It seems a bug in `printenv`.
(BTW, note that environment variables on Windows are case-insensitive).
This is (I believe) a thirdparty tool on Windows that resembles `printenv` on Linux (installed in AppVeyor environment or Windows Server 2012 R2). This command does not exist, at least, for Windows 7 and 10 (manually tested).
On Windows, we can use `cmd.exe /C set [varname]` officially for this purpose. We could fix the tests with this in order to test if the environment variable is set.
## How was this patch tested?
Manually tested via AppVeyor.
**Before**
https://ci.appveyor.com/project/spark-test/spark/build/194-PipedRDDSuite
**After**
https://ci.appveyor.com/project/spark-test/spark/build/226-PipedRDDSuite
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16254 from HyukjinKwon/pipe-errors.
## What changes were proposed in this pull request?
Currently, some tests are being failed and hanging on Windows due to this problem. For the reason in SPARK-18718, some tests using `local-cluster` mode were disabled on Windows due to the length limitation by paths given to classpaths.
The limitation seems roughly 32K (see the [blog in MS](https://blogs.msdn.microsoft.com/oldnewthing/20031210-00/?p=41553/) and [another reference](https://support.thoughtworks.com/hc/en-us/articles/213248526-Getting-around-maximum-command-line-length-is-32767-characters-on-Windows)) but in `local-cluster` mode, executors were being launched as processes with the command such as [here](https://gist.github.com/HyukjinKwon/5bc81061c250d4af5a180869b59d42ea) in (only) tests.
This length is roughly 40K due to the classpaths given to `java` command. However, it seems duplicates are almost half of them. So, if we deduplicate the paths, it seems reduced to roughly 20K with the command, [here](https://gist.github.com/HyukjinKwon/dad0c8db897e5e094684a2dc6a417790).
Maybe, we should consider as some more paths are added in the future but it seems better than disabling all the tests for now with minimised changes.
Therefore, this PR proposes to deduplicate the paths in classpaths in case of launching executors as processes in `local-cluster` mode.
## How was this patch tested?
Existing tests in `ShuffleSuite` and `BroadcastJoinSuite` manually via AppVeyor
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16266 from HyukjinKwon/disable-local-cluster-tests.
There is a small race in SchedulerIntegrationSuite.
The test assumes that the taskscheduler thread
processing that last task will finish before the DAGScheduler processes
the task event and notifies the job waiter, but that is not 100%
guaranteed.
ran the test locally a bunch of times, never failed, though admittedly
it never failed locally for me before either. However I am nearly 100%
certain this is what caused the failure of one jenkins build
https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/68694/consoleFull
(which is long gone now, sorry -- I fixed it as part of
https://github.com/apache/spark/pull/14079 initially)
Author: Imran Rashid <irashid@cloudera.com>
Closes#16270 from squito/sched_integ_flakiness.
## What changes were proposed in this pull request?
Some places in SQL may call `RpcEndpointRef.askWithRetry` (e.g., ParquetFileFormat.buildReader -> SparkContext.broadcast -> ... -> BlockManagerMaster.updateBlockInfo -> RpcEndpointRef.askWithRetry), which will finally call `Await.result`. It may cause `java.lang.IllegalArgumentException: spark.sql.execution.id is already set` when running in Scala ForkJoinPool.
This PR includes the following changes to fix this issue:
- Remove `ThreadUtils.awaitResult`
- Rename `ThreadUtils. awaitResultInForkJoinSafely` to `ThreadUtils.awaitResult`
- Replace `Await.result` in RpcTimeout with `ThreadUtils.awaitResult`.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#16230 from zsxwing/fix-SPARK-13747.
This change moves the logic that translates Spark configuration to
commons-crypto configuration to the network-common module. It also
extends TransportConf and ConfigProvider to provide the necessary
interfaces for the translation to work.
As part of the change, I removed SystemPropertyConfigProvider, which
was mostly used as an "empty config" in unit tests, and adjusted the
very few tests that required a specific config.
I also changed the config keys for AES encryption to live under the
"spark.network." namespace, which is more correct than their previous
names under "spark.authenticate.".
Tested via existing unit test.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#16200 from vanzin/SPARK-18773.
## What changes were proposed in this pull request?
During history server startup, the spark configuration is examined. If security.authentication is
set, log at debug and set the value to false, so that {{SecurityManager}} can be created.
## How was this patch tested?
A new test in `HistoryServerSuite` sets the `spark.authenticate` property to true, tries to create a security manager via a new package-private method `HistoryServer.createSecurityManager(SparkConf)`. This is the method used in `HistoryServer.main`. All other instantiations of a security manager in `HistoryServerSuite` have been switched to the new method, for consistency with the production code.
Author: Steve Loughran <stevel@apache.org>
Closes#13579 from steveloughran/history/SPARK-15844-security.
## What changes were proposed in this pull request?
This PR proposes to fix some tests being failed on Windows as below for several problems.
### Incorrect path handling
- FileSuite
```
[info] - binary file input as byte array *** FAILED *** (500 milliseconds)
[info] "file:/C:/projects/spark/target/tmp/spark-e7c3a3b8-0a4b-4a7f-9ebe-7c4883e48624/record-bytestream-00000.bin" did not contain "C:\projects\spark\target\tmp\spark-e7c3a3b8-0a4b-4a7f-9ebe-7c4883e48624\record-bytestream-00000.bin" (FileSuite.scala:258)
[info] org.scalatest.exceptions.TestFailedException:
[info] at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
...
```
```
[info] - Get input files via old Hadoop API *** FAILED *** (1 second, 94 milliseconds)
[info] Set("/C:/projects/spark/target/tmp/spark-cf5b1f8b-c5ed-43e0-8d17-546ebbfa8200/output/part-00000", "/C:/projects/spark/target/tmp/spark-cf5b1f8b-c5ed-43e0-8d17-546ebbfa8200/output/part-00001") did not equal Set("C:\projects\spark\target\tmp\spark-cf5b1f8b-c5ed-43e0-8d17-546ebbfa8200\output/part-00000", "C:\projects\spark\target\tmp\spark-cf5b1f8b-c5ed-43e0-8d17-546ebbfa8200\output/part-00001") (FileSuite.scala:535)
[info] org.scalatest.exceptions.TestFailedException:
[info] at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
...
```
```
[info] - Get input files via new Hadoop API *** FAILED *** (313 milliseconds)
[info] Set("/C:/projects/spark/target/tmp/spark-12bc1540-1111-4df6-9c4d-79e0e614407c/output/part-00000", "/C:/projects/spark/target/tmp/spark-12bc1540-1111-4df6-9c4d-79e0e614407c/output/part-00001") did not equal Set("C:\projects\spark\target\tmp\spark-12bc1540-1111-4df6-9c4d-79e0e614407c\output/part-00000", "C:\projects\spark\target\tmp\spark-12bc1540-1111-4df6-9c4d-79e0e614407c\output/part-00001") (FileSuite.scala:549)
[info] org.scalatest.exceptions.TestFailedException:
...
```
- TaskResultGetterSuite
```
[info] - handling results larger than max RPC message size *** FAILED *** (1 second, 579 milliseconds)
[info] 1 did not equal 0 Expect result to be removed from the block manager. (TaskResultGetterSuite.scala:129)
[info] org.scalatest.exceptions.TestFailedException:
[info] ...
[info] Cause: java.net.URISyntaxException: Illegal character in path at index 12: string:///C:\projects\spark\target\tmp\spark-93c485af-68da-440f-a907-aac7acd5fc25\repro\MyException.java
[info] at java.net.URI$Parser.fail(URI.java:2848)
[info] at java.net.URI$Parser.checkChars(URI.java:3021)
...
```
```
[info] - failed task deserialized with the correct classloader (SPARK-11195) *** FAILED *** (0 milliseconds)
[info] java.lang.IllegalArgumentException: Illegal character in path at index 12: string:///C:\projects\spark\target\tmp\spark-93c485af-68da-440f-a907-aac7acd5fc25\repro\MyException.java
[info] at java.net.URI.create(URI.java:852)
...
```
- SparkSubmitSuite
```
[info] java.lang.IllegalArgumentException: Illegal character in path at index 12: string:///C:\projects\spark\target\tmp\1481210831381-0\870903339\MyLib.java
[info] at java.net.URI.create(URI.java:852)
[info] at org.apache.spark.TestUtils$.org$apache$spark$TestUtils$$createURI(TestUtils.scala:112)
...
```
### Incorrect separate for JarEntry
After the path fix from above, then `TaskResultGetterSuite` throws another exception as below:
```
[info] - failed task deserialized with the correct classloader (SPARK-11195) *** FAILED *** (907 milliseconds)
[info] java.lang.ClassNotFoundException: repro.MyException
[info] at java.net.URLClassLoader.findClass(URLClassLoader.java:381)
...
```
This is because `Paths.get` concatenates the given paths to an OS-specific path (Windows `\` and Linux `/`). However, for `JarEntry` we should comply ZIP specification meaning it should be always `/` according to ZIP specification.
See `4.4.17 file name: (Variable)` in https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
### Long path problem on Windows
Some tests in `ShuffleSuite` via `ShuffleNettySuite` were skipped due to the same reason with SPARK-18718
## How was this patch tested?
Manually via AppVeyor.
**Before**
- `FileSuite`, `TaskResultGetterSuite`,`SparkSubmitSuite`
https://ci.appveyor.com/project/spark-test/spark/build/164-tmp-windows-base (please grep each to check each)
- `ShuffleSuite`
https://ci.appveyor.com/project/spark-test/spark/build/157-tmp-windows-base
**After**
- `FileSuite`
https://ci.appveyor.com/project/spark-test/spark/build/166-FileSuite
- `TaskResultGetterSuite`
https://ci.appveyor.com/project/spark-test/spark/build/173-TaskResultGetterSuite
- `SparkSubmitSuite`
https://ci.appveyor.com/project/spark-test/spark/build/167-SparkSubmitSuite
- `ShuffleSuite`
https://ci.appveyor.com/project/spark-test/spark/build/176-ShuffleSuite
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16234 from HyukjinKwon/test-errors-windows.
## What changes were proposed in this pull request?
There is an outstanding issue that existed for a long time: Sometimes the shuffle blocks are corrupt and can't be decompressed. We recently hit this in three different workloads, sometimes we can reproduce it by every try, sometimes can't. I also found that when the corruption happened, the beginning and end of the blocks are correct, the corruption happen in the middle. There was one case that the string of block id is corrupt by one character. It seems that it's very likely the corruption is introduced by some weird machine/hardware, also the checksum (16 bits) in TCP is not strong enough to identify all the corruption.
Unfortunately, Spark does not have checksum for shuffle blocks or broadcast, the job will fail if any corruption happen in the shuffle block from disk, or broadcast blocks during network. This PR try to detect the corruption after fetching shuffle blocks by decompressing them, because most of the compression already have checksum in them. It will retry the block, or failed with FetchFailure, so the previous stage could be retried on different (still random) machines.
Checksum for broadcast will be added by another PR.
## How was this patch tested?
Added unit tests
Author: Davies Liu <davies@databricks.com>
Closes#15923 from davies/detect_corrupt.
## What changes were proposed in this pull request?
* This PR changes `JVMObjectTracker` from `object` to `class` and let its instance associated with each RBackend. So we can manage the lifecycle of JVM objects when there are multiple `RBackend` sessions. `RBackend.close` will clear the object tracker explicitly.
* I assume that `SQLUtils` and `RRunner` do not need to track JVM instances, which could be wrong.
* Small refactor of `SerDe.sqlSerDe` to increase readability.
## How was this patch tested?
* Added unit tests for `JVMObjectTracker`.
* Wait for Jenkins to run full tests.
Author: Xiangrui Meng <meng@databricks.com>
Closes#16154 from mengxr/SPARK-17822.
## What changes were proposed in this pull request?
When `SparkContext.stop` is called in `Utils.tryOrStopSparkContext` (the following three places), it will cause deadlock because the `stop` method needs to wait for the thread running `stop` to exit.
- ContextCleaner.keepCleaning
- LiveListenerBus.listenerThread.run
- TaskSchedulerImpl.start
This PR adds `SparkContext.stopInNewThread` and uses it to eliminate the potential deadlock. I also removed my changes in #15775 since they are not necessary now.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#16178 from zsxwing/fix-stop-deadlock.
## What changes were proposed in this pull request?
- Removed the`attempt.completed ` filter so cleaner would include the orphan inprogress files.
- Use loading time for inprogress files as lastUpdated. Keep using the modTime for completed files. First one will prevent deletion of inprogress job files. Second one will ensure that lastUpdated time won't change for completed jobs in an event of HistoryServer reboot.
## How was this patch tested?
Added new unittests and via existing tests.
Author: Ergin Seyfe <eseyfe@fb.com>
Closes#16165 from seyfe/clear_old_inprogress_files.
## What changes were proposed in this pull request?
There are some tests failed on Windows due to the wrong format of path and the limitation of path length as below:
This PR proposes both to fix the failed tests by fixing the path for the tests below:
- `InsertSuite`
```
Exception encountered when attempting to run a suite with class name: org.apache.spark.sql.sources.InsertSuite *** ABORTED *** (12 seconds, 547 milliseconds)
org.apache.spark.sql.AnalysisException: Path does not exist: file:/C:projectsspark arget mpspark-177945ef-9128-42b4-8c07-de31f78bbbd6;
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:382)
at org.apache.spark.sql.execution.datasources.DataSource$$anonfun$14.apply(DataSource.scala:370)
at scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:241)
```
- `PathOptionSuite`
```
- path option also exist for write path *** FAILED *** (1 second, 93 milliseconds)
"C:[projectsspark arget mp]spark-5ab34a58-df8d-..." did not equal "C:[\projects\spark\target\tmp\]spark-5ab34a58-df8d-..." (PathOptionSuite.scala:93)
org.scalatest.exceptions.TestFailedException:
at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
at org.scalatest.FunSuite.newAssertionFailedException(FunSuite.scala:1555)
...
```
- `UDFSuite`
```
- SPARK-8005 input_file_name *** FAILED *** (2 seconds, 234 milliseconds)
"file:///C:/projects/spark/target/tmp/spark-e4e5720a-2006-48f9-8b11-797bf59794bf/part-00001-26fb05e4-603d-471d-ae9d-b9549e0c7765.snappy.parquet" did not contain "C:\projects\spark\target\tmp\spark-e4e5720a-2006-48f9-8b11-797bf59794bf" (UDFSuite.scala:67)
org.scalatest.exceptions.TestFailedException:
at org.scalatest.Assertions$class.newAssertionFailedException(Assertions.scala:500)
at org.scalatest.FunSuite.newAssertionFailedException(FunSuite.scala:1555)
...
```
and to skip the tests belows which are being failed on Windows due to path length limitation.
- `SparkLauncherSuite`
```
Test org.apache.spark.launcher.SparkLauncherSuite.testChildProcLauncher failed: java.lang.AssertionError: expected:<0> but was:<1>, took 0.062 sec
at org.apache.spark.launcher.SparkLauncherSuite.testChildProcLauncher(SparkLauncherSuite.java:177)
...
```
The stderr from the process is `The filename or extension is too long` which is equivalent to the one below.
- `BroadcastJoinSuite`
```
04:09:40.882 ERROR org.apache.spark.deploy.worker.ExecutorRunner: Error running executor
java.io.IOException: Cannot run program "C:\Progra~1\Java\jdk1.8.0\bin\java" (in directory "C:\projects\spark\work\app-20161205040542-0000\51658"): CreateProcess error=206, The filename or extension is too long
at java.lang.ProcessBuilder.start(ProcessBuilder.java:1048)
at org.apache.spark.deploy.worker.ExecutorRunner.org$apache$spark$deploy$worker$ExecutorRunner$$fetchAndRunExecutor(ExecutorRunner.scala:167)
at org.apache.spark.deploy.worker.ExecutorRunner$$anon$1.run(ExecutorRunner.scala:73)
Caused by: java.io.IOException: CreateProcess error=206, The filename or extension is too long
at java.lang.ProcessImpl.create(Native Method)
at java.lang.ProcessImpl.<init>(ProcessImpl.java:386)
at java.lang.ProcessImpl.start(ProcessImpl.java:137)
at java.lang.ProcessBuilder.start(ProcessBuilder.java:1029)
... 2 more
04:09:40.929 ERROR org.apache.spark.deploy.worker.ExecutorRunner: Error running executor
(appearently infinite same error messages)
...
```
## How was this patch tested?
Manually tested via AppVeyor.
**Before**
`InsertSuite`: https://ci.appveyor.com/project/spark-test/spark/build/148-InsertSuite-pr
`PathOptionSuite`: https://ci.appveyor.com/project/spark-test/spark/build/139-PathOptionSuite-pr
`UDFSuite`: https://ci.appveyor.com/project/spark-test/spark/build/143-UDFSuite-pr
`SparkLauncherSuite`: https://ci.appveyor.com/project/spark-test/spark/build/141-SparkLauncherSuite-pr
`BroadcastJoinSuite`: https://ci.appveyor.com/project/spark-test/spark/build/145-BroadcastJoinSuite-pr
**After**
`PathOptionSuite`: https://ci.appveyor.com/project/spark-test/spark/build/140-PathOptionSuite-pr
`SparkLauncherSuite`: https://ci.appveyor.com/project/spark-test/spark/build/142-SparkLauncherSuite-pr
`UDFSuite`: https://ci.appveyor.com/project/spark-test/spark/build/144-UDFSuite-pr
`InsertSuite`: https://ci.appveyor.com/project/spark-test/spark/build/147-InsertSuite-pr
`BroadcastJoinSuite`: https://ci.appveyor.com/project/spark-test/spark/build/149-BroadcastJoinSuite-pr
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#16147 from HyukjinKwon/fix-tests.
## What changes were proposed in this pull request?
Fix reservoir sampling bias for small k. An off-by-one error meant that the probability of replacement was slightly too high -- k/(l-1) after l element instead of k/l, which matters for small k.
## How was this patch tested?
Existing test plus new test case.
Author: Sean Owen <sowen@cloudera.com>
Closes#16129 from srowen/SPARK-18678.
## What changes were proposed in this pull request?
`spark.sql.unsafe.enabled` is deprecated since 1.6. There still are codes in UI to check it. We should remove it and clean the codes.
## How was this patch tested?
Changes to related existing unit test.
Please review http://spark.apache.org/contributing.html before opening a pull request.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Closes#16095 from viirya/remove-deprecated-config-code.
The problem exists because it's not possible to just concatenate encrypted
partition data from different spill files; currently each partition would
have its own initial vector to set up encryption, and the final merged file
should contain a single initial vector for each merged partiton, otherwise
iterating over each record becomes really hard.
To fix that, UnsafeShuffleWriter now decrypts the partitions when merging,
so that the merged file contains a single initial vector at the start of
the partition data.
Because it's not possible to do that using the fast transferTo path, when
encryption is enabled UnsafeShuffleWriter will revert back to using file
streams when merging. It may be possible to use a hybrid approach when
using encryption, using an intermediate direct buffer when reading from
files and encrypting the data, but that's better left for a separate patch.
As part of the change I made DiskBlockObjectWriter take a SerializerManager
instead of a "wrap stream" closure, since that makes it easier to test the
code without having to mock SerializerManager functionality.
Tested with newly added unit tests (UnsafeShuffleWriterSuite for the write
side and ExternalAppendOnlyMapSuite for integration), and by running some
apps that failed without the fix.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#15982 from vanzin/SPARK-18546.
## What changes were proposed in this pull request?
The method `TaskSchedulerImpl.runningTasksByExecutors()` accesses the mutable `executorIdToRunningTaskIds` map without proper synchronization. In addition, as markhamstra pointed out in #15986, the signature's use of parentheses is a little odd given that this is a pure getter method.
This patch fixes both issues.
## How was this patch tested?
Covered by existing tests.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#16073 from JoshRosen/runningTasksByExecutors-thread-safety.
## What changes were proposed in this pull request?
#15992 provided a solution to fix the bug, i.e. **receiver data can not be deserialized properly**. As zsxwing said, it is a critical bug, but we should not break APIs between maintenance releases. It may be a rational choice to close auto pick kryo serializer for Spark Streaming in the first step. I will continue #15992 to optimize the solution.
## How was this patch tested?
existing ut
Author: uncleGen <hustyugm@gmail.com>
Closes#16052 from uncleGen/SPARK-18617.
## What changes were proposed in this pull request?
_This is the master branch version of #15986; the original description follows:_
This patch fixes a critical resource leak in the TaskScheduler which could cause RDDs and ShuffleDependencies to be kept alive indefinitely if an executor with running tasks is permanently lost and the associated stage fails.
This problem was originally identified by analyzing the heap dump of a driver belonging to a cluster that had run out of shuffle space. This dump contained several `ShuffleDependency` instances that were retained by `TaskSetManager`s inside the scheduler but were not otherwise referenced. Each of these `TaskSetManager`s was considered a "zombie" but had no running tasks and therefore should have been cleaned up. However, these zombie task sets were still referenced by the `TaskSchedulerImpl.taskIdToTaskSetManager` map.
Entries are added to the `taskIdToTaskSetManager` map when tasks are launched and are removed inside of `TaskScheduler.statusUpdate()`, which is invoked by the scheduler backend while processing `StatusUpdate` messages from executors. The problem with this design is that a completely dead executor will never send a `StatusUpdate`. There is [some code](072f4c518c/core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala (L338)) in `statusUpdate` which handles tasks that exit with the `TaskState.LOST` state (which is supposed to correspond to a task failure triggered by total executor loss), but this state only seems to be used in Mesos fine-grained mode. There doesn't seem to be any code which performs per-task state cleanup for tasks that were running on an executor that completely disappears without sending any sort of final death message. The `executorLost` and [`removeExecutor`](072f4c518c/core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala (L527)) methods don't appear to perform any cleanup of the `taskId -> *` mappings, causing the leaks observed here.
This patch's fix is to maintain a `executorId -> running task id` mapping so that these `taskId -> *` maps can be properly cleaned up following an executor loss.
There are some potential corner-case interactions that I'm concerned about here, especially some details in [the comment](072f4c518c/core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala (L523)) in `removeExecutor`, so I'd appreciate a very careful review of these changes.
## How was this patch tested?
I added a new unit test to `TaskSchedulerImplSuite`.
/cc kayousterhout and markhamstra, who reviewed #15986.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#16045 from JoshRosen/fix-leak-following-total-executor-loss-master.
This change modifies the method used to propagate encryption keys used during
shuffle. Instead of relying on YARN's UserGroupInformation credential propagation,
this change explicitly distributes the key using the messages exchanged between
driver and executor during registration. When RPC encryption is enabled, this means
key propagation is also secure.
This allows shuffle encryption to work in non-YARN mode, which means that it's
easier to write unit tests for areas of the code that are affected by the feature.
The key is stored in the SecurityManager; because there are many instances of
that class used in the code, the key is only guaranteed to exist in the instance
managed by the SparkEnv. This path was chosen to avoid storing the key in the
SparkConf, which would risk having the key being written to disk as part of the
configuration (as, for example, is done when starting YARN applications).
Tested by new and existing unit tests (which were moved from the YARN module to
core), and by running apps with shuffle encryption enabled.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#15981 from vanzin/SPARK-18547.
## What changes were proposed in this pull request?
This adds tests to verify the interaction between TaskSetBlacklist and
TaskSchedulerImpl. TaskSetBlacklist was introduced by SPARK-17675 but
it neglected to add these tests.
This change does not fix any bugs -- it is just for increasing test
coverage.
## How was this patch tested?
Jenkins
Author: Imran Rashid <irashid@cloudera.com>
Closes#15644 from squito/taskset_blacklist_test_update.
## What changes were proposed in this pull request?
This patch adds a new property called `spark.secret.redactionPattern` that
allows users to specify a scala regex to decide which Spark configuration
properties and environment variables in driver and executor environments
contain sensitive information. When this regex matches the property or
environment variable name, its value is redacted from the environment UI and
various logs like YARN and event logs.
This change uses this property to redact information from event logs and YARN
logs. It also, updates the UI code to adhere to this property instead of
hardcoding the logic to decipher which properties are sensitive.
Here's an image of the UI post-redaction:
Here's the text in the YARN logs, post-redaction:
``HADOOP_CREDSTORE_PASSWORD -> *********(redacted)``
Here's the text in the event logs, post-redaction:
``...,"spark.executorEnv.HADOOP_CREDSTORE_PASSWORD":"*********(redacted)","spark.yarn.appMasterEnv.HADOOP_CREDSTORE_PASSWORD":"*********(redacted)",...``
## How was this patch tested?
1. Unit tests are added to ensure that redaction works.
2. A YARN job reading data off of S3 with confidential information
(hadoop credential provider password) being provided in the environment
variables of driver and executor. And, afterwards, logs were grepped to make
sure that no mention of secret password was present. It was also ensure that
the job was able to read the data off of S3 correctly, thereby ensuring that
the sensitive information was being trickled down to the right places to read
the data.
3. The event logs were checked to make sure no mention of secret password was
present.
4. UI environment tab was checked to make sure there was no secret information
being displayed.
Author: Mark Grover <mark@apache.org>
Closes#15971 from markgrover/master_redaction.
## What changes were proposed in this pull request?
This PR avoids that a result of an expression is negative due to signed integer overflow (e.g. 0x10?????? * 8 < 0). This PR casts each operand to `long` before executing a calculation. Since the result is interpreted as long, the result of the expression is positive.
## How was this patch tested?
Manually executed query82 of TPC-DS with 100TB
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#15907 from kiszk/SPARK-18458.
Adds --conf option to set spark configuration properties in mesos dispacther.
Properties provided with --conf take precedence over properties within the properties file.
The reason for this PR is that for simple configuration or testing purposes we need to provide a property file (ideally a shared one for a cluster) even if we just provide a single property.
Manually tested.
Author: Stavros Kontopoulos <st.kontopoulos@gmail.com>
Author: Stavros Kontopoulos <stavros.kontopoulos@lightbend.com>
Closes#14650 from skonto/dipatcher_conf.
## What changes were proposed in this pull request?
This is a follow-up work of #15618.
Close file source;
For any newly created streaming context outside the withContext, explicitly close the context.
## How was this patch tested?
Existing unit tests.
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#15818 from wangmiao1981/rtest.
## What changes were proposed in this pull request?
We should call `setConf` if `OutputFormat` is `Configurable`, this should be done before we create `OutputCommitter` and `RecordWriter`.
This is follow up of #15769, see discussion [here](https://github.com/apache/spark/pull/15769/files#r87064229)
## How was this patch tested?
Add test of this case in `PairRDDFunctionsSuite`.
Author: jiangxingbo <jiangxb1987@gmail.com>
Closes#15823 from jiangxb1987/config-format.
## What changes were proposed in this pull request?
Application links generated on the history server UI no longer (regression from 1.6) contain the configured spark.ui.proxyBase in the links. To address this, made the uiRoot available globally to all javascripts for Web UI. Updated the mustache template (historypage-template.html) to include the uiroot for rendering links to the applications.
The existing test was not sufficient to verify the scenario where ajax call is used to populate the application listing template, so added a new selenium test case to cover this scenario.
## How was this patch tested?
Existing tests and a new unit test.
No visual changes to the UI.
Author: Vinayak <vijoshi5@in.ibm.com>
Closes#15742 from vijoshi/SPARK-16808_master.
## What changes were proposed in this pull request?
"StandaloneSchedulerBackend.dead" is called in a RPC thread, so it should not call "SparkContext.stop" in the same thread. "SparkContext.stop" will block until all RPC threads exit, if it's called inside a RPC thread, it will be dead-lock.
This PR add a thread local flag inside RPC threads. `SparkContext.stop` uses it to decide if launching a new thread to stop the SparkContext.
## How was this patch tested?
Jenkins
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#15775 from zsxwing/SPARK-18280.
## What changes were proposed in this pull request?
This PR port RDD API to use commit protocol, the changes made here:
1. Add new internal helper class that saves an RDD using a Hadoop OutputFormat named `SparkNewHadoopWriter`, it's similar with `SparkHadoopWriter` but uses commit protocol. This class supports the newer `mapreduce` API, instead of the old `mapred` API which is supported by `SparkHadoopWriter`;
2. Rewrite `PairRDDFunctions.saveAsNewAPIHadoopDataset` function, so it uses commit protocol now.
## How was this patch tested?
Exsiting test cases.
Author: jiangxingbo <jiangxb1987@gmail.com>
Closes#15769 from jiangxb1987/rdd-commit.
## What changes were proposed in this pull request?
When profiling heap dumps from the HistoryServer and live Spark web UIs, I found a large amount of memory being wasted on duplicated objects and strings. This patch's changes remove most of this duplication, resulting in over 40% memory savings for some benchmarks.
- **Task metrics** (6441f0624dfcda9c7193a64bfb416a145b5aabdf): previously, every `TaskUIData` object would have its own instances of `InputMetricsUIData`, `OutputMetricsUIData`, `ShuffleReadMetrics`, and `ShuffleWriteMetrics`, but for many tasks these metrics are irrelevant because they're all zero. This patch changes how we construct these metrics in order to re-use a single immutable "empty" value for the cases where these metrics are empty.
- **TaskInfo.accumulables** (ade86db901127bf13c0e0bdc3f09c933a093bb76): Previously, every `TaskInfo` object had its own empty `ListBuffer` for holding updates from named accumulators. Tasks which didn't use named accumulators still paid for the cost of allocating and storing this empty buffer. To avoid this overhead, I changed the `val` with a mutable buffer into a `var` which holds an immutable Scala list, allowing tasks which do not have named accumulator updates to share the same singleton `Nil` object.
- **String.intern() in JSONProtocol** (7e05630e9a78c455db8c8c499f0590c864624e05): in the HistoryServer, executor hostnames and ids are deserialized from JSON, leading to massive duplication of these string objects. By calling `String.intern()` on the deserialized values we can remove all of this duplication. Since Spark now requires Java 7+ we don't have to worry about string interning exhausting the permgen (see http://java-performance.info/string-intern-in-java-6-7-8/).
## How was this patch tested?
I ran
```
sc.parallelize(1 to 100000, 100000).count()
```
in `spark-shell` with event logging enabled, then loaded that event log in the HistoryServer, performed a full GC, and took a heap dump. According to YourKit, the changes in this patch reduced memory consumption by roughly 28 megabytes (or 770k Java objects):
Here's a table illustrating the drop in objects due to deduplication (the drop is <100k for some objects because some events were dropped from the listener bus; this is a separate, existing bug that I'll address separately after CPU-profiling):
Author: Josh Rosen <joshrosen@databricks.com>
Closes#15743 from JoshRosen/spark-ui-memory-usage.
## What changes were proposed in this pull request?
Close `FileStreams`, `ZipFiles` etc to release the resources after using. Not closing the resources will cause IO Exception to be raised while deleting temp files.
## How was this patch tested?
Existing tests
Author: U-FAREAST\tl <tl@microsoft.com>
Author: hyukjinkwon <gurwls223@gmail.com>
Author: Tao LI <tl@microsoft.com>
Closes#15618 from HyukjinKwon/SPARK-14914-1.
## What changes were proposed in this pull request?
[SPARK-18200](https://issues.apache.org/jira/browse/SPARK-18200) reports Apache Spark 2.x raises `java.lang.IllegalArgumentException: requirement failed: Invalid initial capacity` while running `triangleCount`. The root cause is that `VertexSet`, a type alias of `OpenHashSet`, does not allow zero as a initial size. This PR loosens the restriction to allow zero.
## How was this patch tested?
Pass the Jenkins test with a new test case in `OpenHashSetSuite`.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#15741 from dongjoon-hyun/SPARK-18200.
## What changes were proposed in this pull request?
Fix `Locale.US` for all usages of `DateFormat`, `NumberFormat`
## How was this patch tested?
Existing tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#15610 from srowen/SPARK-18076.
## What changes were proposed in this pull request?
Removing `appUIAddress` attribute since it is no longer in use.
## How was this patch tested?
Local build
Author: Jacek Laskowski <jacek@japila.pl>
Closes#15603 from jaceklaskowski/sparkui-fixes.
## What changes were proposed in this pull request?
While learning core scheduler code, I found two lines of wrong comments. This PR simply corrects the comments.
## How was this patch tested?
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#15631 from wangmiao1981/Rbug.
## What changes were proposed in this pull request?
(Please fill in changes proposed in this fix)
`Utils.getIteratorZipWithIndex` was added to deal with number of records > 2147483647 in one partition.
method `getIteratorZipWithIndex` accepts `startIndex` < 0, which leads to negative index.
This PR just adds a defensive check on `startIndex` to make sure it is >= 0.
## How was this patch tested?
Add a new unit test.
Author: Miao Wang <miaowang@Miaos-MacBook-Pro.local>
Closes#15639 from wangmiao1981/zip.
