[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-06-04 Thread felixcheung
Github user felixcheung commented on the issue: https://github.com/apache/spark/pull/21390 Thanks @JoshRosen for very detailed and thoughtful reply. Agreed TTL could be fragile, but I was very concern with what point > There is a related issue where shuffle files can be

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-06-02 Thread JoshRosen
Github user JoshRosen commented on the issue: https://github.com/apache/spark/pull/21390 Feel free to do the TTL in a followup. My feeling is that it won't be super useful in practice, though: 1. Cleanup of non-shuffle disk block manager files following executor exit only

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-06-02 Thread felixcheung
Github user felixcheung commented on the issue: https://github.com/apache/spark/pull/21390 sounds like we need to have some sort of lifetime management, TTL -like design shuffle file - should we have a new JIRA on that? ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-06-01 Thread gatorsmile
Github user gatorsmile commented on the issue: https://github.com/apache/spark/pull/21390 Thanks! Merged to master. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail:

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-29 Thread jiangxb1987
Github user jiangxb1987 commented on the issue: https://github.com/apache/spark/pull/21390 Are there any other concerns over this PR? --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-25 Thread JoshRosen
Github user JoshRosen commented on the issue: https://github.com/apache/spark/pull/21390 Yeah, this is only concerned with non-shuffle files which are located in the block manager temp directories (e.g. large sorter spill files). There is a related issue where shuffle files

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-25 Thread mccheah
Github user mccheah commented on the issue: https://github.com/apache/spark/pull/21390 Actually since this specifically applies to _non_-shuffle files I think Kubernetes will be fine here regardless. --- - To

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-25 Thread mccheah
Github user mccheah commented on the issue: https://github.com/apache/spark/pull/21390 For Kubernetes, without a story around the external shuffle service, all the scratch space used by executors will be cleaned up by Kubernetes itself. When we want shuffle data to persist across

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/91128/ Test PASSed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91128 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91128/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3565/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread jiangxb1987
Github user jiangxb1987 commented on the issue: https://github.com/apache/spark/pull/21390 @jerryshao Agree it should be useful to add a `debug-delay-sec` config for ease of developing, since this PR has already bring in a brunch of code changes, maybe we can add the config in a

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91128 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91128/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread jerryshao
Github user jerryshao commented on the issue: https://github.com/apache/spark/pull/21390 YARN will clean container local dirs when container (executor) is exited, so this may not be a problem in YARN. YARN has a useful configuration "yarn.nodemanager.delete.debug-delay-sec"

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/91083/ Test FAILed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test FAILed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-24 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91083 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91083/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3536/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91083 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91083/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread jiangxb1987
Github user jiangxb1987 commented on the issue: https://github.com/apache/spark/pull/21390 retest this please --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail:

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/91078/ Test FAILed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test FAILed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91078 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91078/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3533/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91078 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91078/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/91064/ Test PASSed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91064 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91064/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3521/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91064 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91064/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-23 Thread gatorsmile
Github user gatorsmile commented on the issue: https://github.com/apache/spark/pull/21390 retest this please --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail:

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test FAILed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/91006/ Test FAILed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91006 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91006/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3483/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #91006 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/91006/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test FAILed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test FAILed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/90942/ Test FAILed. ---

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #90942 has finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/90942/testReport)** for PR 21390 at commit

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread JoshRosen
Github user JoshRosen commented on the issue: https://github.com/apache/spark/pull/21390 Context for other reviewers: the issue addressed by this patch is actually a real issue in practice, especially for long-lived Spark clusters; I have seen this specific problem play a large

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Merged build finished. Test PASSed. --- - To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue: https://github.com/apache/spark/pull/21390 Test PASSed. Refer to this link for build results (access rights to CI server needed): https://amplab.cs.berkeley.edu/jenkins//job/testing-k8s-prb-make-spark-distribution/3448/

[GitHub] spark issue #21390: [SPARK-24340][Core] Clean up non-shuffle disk block mana...

2018-05-22 Thread SparkQA
Github user SparkQA commented on the issue: https://github.com/apache/spark/pull/21390 **[Test build #90942 has started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/90942/testReport)** for PR 21390 at commit