spark git commit: [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly()
Repository: spark Updated Branches: refs/heads/branch-2.0 23050c8a1 -> f50c4372c [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly() This patch replaces a single `awaitUninterruptibly()` call with a plain `await()` call in Spark's `network-common` library in order to fix a bug which may cause tasks to be uncancellable. In Spark's Netty RPC layer, `TransportClientFactory.createClient()` calls `awaitUninterruptibly()` on a Netty future while waiting for a connection to be established. This creates problem when a Spark task is interrupted while blocking in this call (which can happen in the event of a slow connection which will eventually time out). This has bad impacts on task cancellation when `interruptOnCancel = true`. As an example of the impact of this problem, I experienced significant numbers of uncancellable "zombie tasks" on a production cluster where several tasks were blocked trying to connect to a dead shuffle server and then continued running as zombies after I cancelled the associated Spark stage. The zombie tasks ran for several minutes with the following stack: ``` java.lang.Object.wait(Native Method) java.lang.Object.wait(Object.java:460) io.netty.util.concurrent.DefaultPromise.await0(DefaultPromise.java:607) io.netty.util.concurrent.DefaultPromise.awaitUninterruptibly(DefaultPromise.java:301) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:224) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:179) => holding Monitor(java.lang.Object1849476028}) org.apache.spark.network.shuffle.ExternalShuffleClient$1.createAndStart(ExternalShuffleClient.java:105) org.apache.spark.network.shuffle.RetryingBlockFetcher.fetchAllOutstanding(RetryingBlockFetcher.java:140) org.apache.spark.network.shuffle.RetryingBlockFetcher.start(RetryingBlockFetcher.java:120) org.apache.spark.network.shuffle.ExternalShuffleClient.fetchBlocks(ExternalShuffleClient.java:114) org.apache.spark.storage.ShuffleBlockFetcherIterator.sendRequest(ShuffleBlockFetcherIterator.scala:169) org.apache.spark.storage.ShuffleBlockFetcherIterator.fetchUpToMaxBytes(ShuffleBlockFetcherIterator.scala: 350) org.apache.spark.storage.ShuffleBlockFetcherIterator.initialize(ShuffleBlockFetcherIterator.scala:286) org.apache.spark.storage.ShuffleBlockFetcherIterator.(ShuffleBlockFetcherIterator.scala:120) org.apache.spark.shuffle.BlockStoreShuffleReader.read(BlockStoreShuffleReader.scala:45) org.apache.spark.sql.execution.ShuffledRowRDD.compute(ShuffledRowRDD.scala:169) org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:323) org.apache.spark.rdd.RDD.iterator(RDD.scala:287) [...] ``` As far as I can tell, `awaitUninterruptibly()` might have been used in order to avoid having to declare that methods throw `InterruptedException` (this code is written in Java, hence the need to use checked exceptions). This patch simply replaces this with a regular, interruptible `await()` call,. This required several interface changes to declare a new checked exception (these are internal interfaces, though, and this change doesn't significantly impact binary compatibility). An alternative approach would be to wrap `InterruptedException` into `IOException` in order to avoid having to change interfaces. The problem with this approach is that the `network-shuffle` project's `RetryingBlockFetcher` code treats `IOExceptions` as transitive failures when deciding whether to retry fetches, so throwing a wrapped `IOException` might cause an interrupted shuffle fetch to be retried, further prolonging the lifetime of a cancelled zombie task. Note that there are three other `awaitUninterruptibly()` in the codebase, but those calls have a hard 10 second timeout and are waiting on a `close()` operation which is expected to complete near instantaneously, so the impact of uninterruptibility there is much smaller. Manually. Author: Josh Rosen Closes #16866 from JoshRosen/SPARK-19529. (cherry picked from commit 1c4d10b10c78d138b55e381ec6828e04fef70d6f) Signed-off-by: Cheng Lian Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/f50c4372 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/f50c4372 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/f50c4372 Branch: refs/heads/branch-2.0 Commit: f50c4372c3ebd91c0f6c094a7c4d1dd08f3cdb30 Parents: 23050c8 Author: Josh Rosen Authored: Mon Feb 13 11:04:27 2017 -0800 Committer: Cheng Lian Committed: Mon Feb 13 12:57:29 2017 -0800 -- .../network/client/TransportClientFactory.java | 10 ++ .../spark/network/TransportClientFactorySuite.java | 6 -- .../network/shuffle/ExternalShuffleClient.java | 4 ++-- .../spark/network/shuffle/RetryingBlockFetcher.java | 3 ++- .../shuffle/mesos/MesosExternalShu
spark git commit: [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly()
Repository: spark Updated Branches: refs/heads/branch-2.1 2968d8c06 -> 5db234730 [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly() This patch replaces a single `awaitUninterruptibly()` call with a plain `await()` call in Spark's `network-common` library in order to fix a bug which may cause tasks to be uncancellable. In Spark's Netty RPC layer, `TransportClientFactory.createClient()` calls `awaitUninterruptibly()` on a Netty future while waiting for a connection to be established. This creates problem when a Spark task is interrupted while blocking in this call (which can happen in the event of a slow connection which will eventually time out). This has bad impacts on task cancellation when `interruptOnCancel = true`. As an example of the impact of this problem, I experienced significant numbers of uncancellable "zombie tasks" on a production cluster where several tasks were blocked trying to connect to a dead shuffle server and then continued running as zombies after I cancelled the associated Spark stage. The zombie tasks ran for several minutes with the following stack: ``` java.lang.Object.wait(Native Method) java.lang.Object.wait(Object.java:460) io.netty.util.concurrent.DefaultPromise.await0(DefaultPromise.java:607) io.netty.util.concurrent.DefaultPromise.awaitUninterruptibly(DefaultPromise.java:301) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:224) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:179) => holding Monitor(java.lang.Object1849476028}) org.apache.spark.network.shuffle.ExternalShuffleClient$1.createAndStart(ExternalShuffleClient.java:105) org.apache.spark.network.shuffle.RetryingBlockFetcher.fetchAllOutstanding(RetryingBlockFetcher.java:140) org.apache.spark.network.shuffle.RetryingBlockFetcher.start(RetryingBlockFetcher.java:120) org.apache.spark.network.shuffle.ExternalShuffleClient.fetchBlocks(ExternalShuffleClient.java:114) org.apache.spark.storage.ShuffleBlockFetcherIterator.sendRequest(ShuffleBlockFetcherIterator.scala:169) org.apache.spark.storage.ShuffleBlockFetcherIterator.fetchUpToMaxBytes(ShuffleBlockFetcherIterator.scala: 350) org.apache.spark.storage.ShuffleBlockFetcherIterator.initialize(ShuffleBlockFetcherIterator.scala:286) org.apache.spark.storage.ShuffleBlockFetcherIterator.(ShuffleBlockFetcherIterator.scala:120) org.apache.spark.shuffle.BlockStoreShuffleReader.read(BlockStoreShuffleReader.scala:45) org.apache.spark.sql.execution.ShuffledRowRDD.compute(ShuffledRowRDD.scala:169) org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:323) org.apache.spark.rdd.RDD.iterator(RDD.scala:287) [...] ``` As far as I can tell, `awaitUninterruptibly()` might have been used in order to avoid having to declare that methods throw `InterruptedException` (this code is written in Java, hence the need to use checked exceptions). This patch simply replaces this with a regular, interruptible `await()` call,. This required several interface changes to declare a new checked exception (these are internal interfaces, though, and this change doesn't significantly impact binary compatibility). An alternative approach would be to wrap `InterruptedException` into `IOException` in order to avoid having to change interfaces. The problem with this approach is that the `network-shuffle` project's `RetryingBlockFetcher` code treats `IOExceptions` as transitive failures when deciding whether to retry fetches, so throwing a wrapped `IOException` might cause an interrupted shuffle fetch to be retried, further prolonging the lifetime of a cancelled zombie task. Note that there are three other `awaitUninterruptibly()` in the codebase, but those calls have a hard 10 second timeout and are waiting on a `close()` operation which is expected to complete near instantaneously, so the impact of uninterruptibility there is much smaller. Manually. Author: Josh Rosen Closes #16866 from JoshRosen/SPARK-19529. (cherry picked from commit 1c4d10b10c78d138b55e381ec6828e04fef70d6f) Signed-off-by: Cheng Lian Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/5db23473 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/5db23473 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/5db23473 Branch: refs/heads/branch-2.1 Commit: 5db23473008a58fb9a7f77ad8b01bcdc2c5f2d9c Parents: 2968d8c Author: Josh Rosen Authored: Mon Feb 13 11:04:27 2017 -0800 Committer: Cheng Lian Committed: Mon Feb 13 12:49:37 2017 -0800 -- .../network/client/TransportClientFactory.java | 10 ++ .../spark/network/TransportClientFactorySuite.java | 6 -- .../network/shuffle/ExternalShuffleClient.java | 4 ++-- .../spark/network/shuffle/RetryingBlockFetcher.java | 3 ++- .../shuffle/mesos/MesosExternalShu
spark git commit: [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly()
Repository: spark Updated Branches: refs/heads/master ab88b2410 -> 1c4d10b10 [SPARK-19529] TransportClientFactory.createClient() shouldn't call awaitUninterruptibly() ## What changes were proposed in this pull request? This patch replaces a single `awaitUninterruptibly()` call with a plain `await()` call in Spark's `network-common` library in order to fix a bug which may cause tasks to be uncancellable. In Spark's Netty RPC layer, `TransportClientFactory.createClient()` calls `awaitUninterruptibly()` on a Netty future while waiting for a connection to be established. This creates problem when a Spark task is interrupted while blocking in this call (which can happen in the event of a slow connection which will eventually time out). This has bad impacts on task cancellation when `interruptOnCancel = true`. As an example of the impact of this problem, I experienced significant numbers of uncancellable "zombie tasks" on a production cluster where several tasks were blocked trying to connect to a dead shuffle server and then continued running as zombies after I cancelled the associated Spark stage. The zombie tasks ran for several minutes with the following stack: ``` java.lang.Object.wait(Native Method) java.lang.Object.wait(Object.java:460) io.netty.util.concurrent.DefaultPromise.await0(DefaultPromise.java:607) io.netty.util.concurrent.DefaultPromise.awaitUninterruptibly(DefaultPromise.java:301) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:224) org.apache.spark.network.client.TransportClientFactory.createClient(TransportClientFactory.java:179) => holding Monitor(java.lang.Object1849476028}) org.apache.spark.network.shuffle.ExternalShuffleClient$1.createAndStart(ExternalShuffleClient.java:105) org.apache.spark.network.shuffle.RetryingBlockFetcher.fetchAllOutstanding(RetryingBlockFetcher.java:140) org.apache.spark.network.shuffle.RetryingBlockFetcher.start(RetryingBlockFetcher.java:120) org.apache.spark.network.shuffle.ExternalShuffleClient.fetchBlocks(ExternalShuffleClient.java:114) org.apache.spark.storage.ShuffleBlockFetcherIterator.sendRequest(ShuffleBlockFetcherIterator.scala:169) org.apache.spark.storage.ShuffleBlockFetcherIterator.fetchUpToMaxBytes(ShuffleBlockFetcherIterator.scala: 350) org.apache.spark.storage.ShuffleBlockFetcherIterator.initialize(ShuffleBlockFetcherIterator.scala:286) org.apache.spark.storage.ShuffleBlockFetcherIterator.(ShuffleBlockFetcherIterator.scala:120) org.apache.spark.shuffle.BlockStoreShuffleReader.read(BlockStoreShuffleReader.scala:45) org.apache.spark.sql.execution.ShuffledRowRDD.compute(ShuffledRowRDD.scala:169) org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:323) org.apache.spark.rdd.RDD.iterator(RDD.scala:287) [...] ``` As far as I can tell, `awaitUninterruptibly()` might have been used in order to avoid having to declare that methods throw `InterruptedException` (this code is written in Java, hence the need to use checked exceptions). This patch simply replaces this with a regular, interruptible `await()` call,. This required several interface changes to declare a new checked exception (these are internal interfaces, though, and this change doesn't significantly impact binary compatibility). An alternative approach would be to wrap `InterruptedException` into `IOException` in order to avoid having to change interfaces. The problem with this approach is that the `network-shuffle` project's `RetryingBlockFetcher` code treats `IOExceptions` as transitive failures when deciding whether to retry fetches, so throwing a wrapped `IOException` might cause an interrupted shuffle fetch to be retried, further prolonging the lifetime of a cancelled zombie task. Note that there are three other `awaitUninterruptibly()` in the codebase, but those calls have a hard 10 second timeout and are waiting on a `close()` operation which is expected to complete near instantaneously, so the impact of uninterruptibility there is much smaller. ## How was this patch tested? Manually. Author: Josh Rosen Closes #16866 from JoshRosen/SPARK-19529. Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/1c4d10b1 Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/1c4d10b1 Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/1c4d10b1 Branch: refs/heads/master Commit: 1c4d10b10c78d138b55e381ec6828e04fef70d6f Parents: ab88b24 Author: Josh Rosen Authored: Mon Feb 13 11:04:27 2017 -0800 Committer: Cheng Lian Committed: Mon Feb 13 11:04:27 2017 -0800 -- .../network/client/TransportClientFactory.java | 10 ++ .../spark/network/TransportClientFactorySuite.java | 6 -- .../network/shuffle/ExternalShuffleClient.java | 4 ++-- .../spark/network/shuffle/RetryingBlockFetcher.java | 3 ++- .../shuffle/mesos/MesosExternalShuffleClient.java |