[jira] [Commented] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15143344#comment-15143344 ] Rajesh Balamohan commented on TEZ-3107: --- [~bikassaha] - Please review when you find sometime. >

[jira] [Commented] (TEZ-3104) Tez fails on Bzip2 intermediate output format on hadoop 2.7.1 and earlier

[ https://issues.apache.org/jira/browse/TEZ-3104?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15142767#comment-15142767 ] Rajesh Balamohan commented on TEZ-3104: --- It works around the sync issues in bzip2.isNativeBzip2Loaded.

[jira] [Commented] (TEZ-3113) massive increase of run time using PipelinedSorter rather than DefaultSorter

[ https://issues.apache.org/jira/browse/TEZ-3113?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15142954#comment-15142954 ] Rajesh Balamohan commented on TEZ-3113: --- Is it possible to share the logs and the sort.mb settings for

[jira] [Updated] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3107: -- Attachment: TEZ-3107.2.patch > tez-tools: Log warn msgs in case ATS has wrong values (e.g

[jira] [Updated] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3107: -- Attachment: (was: TEZ-3107.2.patch) > tez-tools: Log warn msgs in case ATS has wrong values

[jira] [Updated] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3107: -- Attachment: TEZ-3107.3.patch Attaching the correct patch. > tez-tools: Log warn msgs in case ATS

[jira] [Commented] (TEZ-3103) Shuffle can hang when memory to memory merging enabled

[ https://issues.apache.org/jira/browse/TEZ-3103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15143583#comment-15143583 ] Rajesh Balamohan commented on TEZ-3103: --- +1. lgtm. Thanks [~jlowe] > Shuffle can hang when memory to

[jira] [Updated] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3107: -- Attachment: TEZ-3107.4.patch Thanks [~bikassaha]. Fixed in latest patch. Will commit shortly. >

[jira] [Commented] (TEZ-3032) DAG start time getting logged using system time instead of recorded time in startTime field

[ https://issues.apache.org/jira/browse/TEZ-3032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15142999#comment-15142999 ] Rajesh Balamohan commented on TEZ-3032: --- Should this is backported to 0.7 as well? > DAG start time

[jira] [Updated] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3107: -- Attachment: TEZ-3107.1.patch Event info contain the correct values for start and end time. In

[jira] [Created] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

Rajesh Balamohan created TEZ-3107: - Summary: tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime) Key: TEZ-3107 URL: https://issues.apache.org/jira/browse/TEZ-3107

[jira] [Commented] (TEZ-3107) tez-tools: Log warn msgs in case ATS has wrong values (e.g startTime > finishTime)

[ https://issues.apache.org/jira/browse/TEZ-3107?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15141872#comment-15141872 ] Rajesh Balamohan commented on TEZ-3107: --- It didn't have TEZ-3032 in the version that was tried out. >

[jira] [Created] (TEZ-3098) tez-tools: Provide filter options in tfile loader

Rajesh Balamohan created TEZ-3098: - Summary: tez-tools: Provide filter options in tfile loader Key: TEZ-3098 URL: https://issues.apache.org/jira/browse/TEZ-3098 Project: Apache Tez Issue

[jira] [Updated] (TEZ-3098) tez-tools: Provide filter options in tfile loader

[ https://issues.apache.org/jira/browse/TEZ-3098?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3098: -- Attachment: TEZ-3098.1.patch > tez-tools: Provide filter options in tfile loader >

[jira] [Commented] (TEZ-3098) tez-tools: Provide filter options in tfile loader

[ https://issues.apache.org/jira/browse/TEZ-3098?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15134207#comment-15134207 ] Rajesh Balamohan commented on TEZ-3098: --- [~gopalv] - Can you plz review when you find time? >

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Attachment: TEZ-3093.2.patch > CriticalPathAnalyzer should be accessible via zeppelin >

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Target Version/s: 0.7.1 > CriticalPathAnalyzer should be accessible via zeppelin >

[jira] [Updated] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3089: -- Fix Version/s: (was: 0.7.2) 0.7.1 > TaskConcurrencyAnalyzer can return

[jira] [Commented] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15133557#comment-15133557 ] Rajesh Balamohan commented on TEZ-3089: --- branch-0.7: commit ca0ecab2c82a4e52c5adc58d2551fe45e084c5d0

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Attachment: TEZ-3093.addendum.patch renamed. Will commit to branch-0.7 as well. >

[jira] [Updated] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3089: -- Target Version/s: 0.7.2 > TaskConcurrencyAnalyzer can return negative task count with very large

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Fix Version/s: 0.7.1 > CriticalPathAnalyzer should be accessible via zeppelin >

[jira] [Commented] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15133565#comment-15133565 ] Rajesh Balamohan commented on TEZ-3093: --- committed addendum to master: >>> commit

[jira] [Updated] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3089: -- Target Version/s: 0.7.1 (was: 0.7.2) > TaskConcurrencyAnalyzer can return negative task count

[jira] [Assigned] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan reassigned TEZ-3093: - Assignee: Rajesh Balamohan > CriticalPathAnalyzer should be accessible via zeppelin >

[jira] [Updated] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3089: -- Fix Version/s: 0.7.2 > TaskConcurrencyAnalyzer can return negative task count with very large

[jira] [Commented] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15132771#comment-15132771 ] Rajesh Balamohan commented on TEZ-3093: --- Normally, output-dir is populated by parsing the command line

[jira] [Commented] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15128135#comment-15128135 ] Rajesh Balamohan commented on TEZ-3089: --- Start and finish are the only event types expected in this

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Issue Type: Improvement (was: Bug) > CriticalPathAnalyzer should be accessible via zeppelin >

[jira] [Created] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

Rajesh Balamohan created TEZ-3093: - Summary: CriticalPathAnalyzer should be accessible via zeppelin Key: TEZ-3093 URL: https://issues.apache.org/jira/browse/TEZ-3093 Project: Apache Tez

[jira] [Updated] (TEZ-3093) CriticalPathAnalyzer should be accessible via zeppelin

[ https://issues.apache.org/jira/browse/TEZ-3093?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3093: -- Attachment: TEZ-3093.1.patch [~bikassaha] - Please review when you find time. >

[jira] [Created] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

Rajesh Balamohan created TEZ-3089: - Summary: TaskConcurrencyAnalyzer can return negative task count with very large jobs Key: TEZ-3089 URL: https://issues.apache.org/jira/browse/TEZ-3089 Project:

[jira] [Updated] (TEZ-3089) TaskConcurrencyAnalyzer can return negative task count with very large jobs

[ https://issues.apache.org/jira/browse/TEZ-3089?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3089: -- Attachment: TEZ-3089.1.patch [~bikassaha] - Please review when you find time. >

[jira] [Commented] (TEZ-2974) Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles

[ https://issues.apache.org/jira/browse/TEZ-2974?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15127317#comment-15127317 ] Rajesh Balamohan commented on TEZ-2974: --- Thanks [~gopalv]. Patch worked fine with test-case in master

[jira] [Created] (TEZ-3079) Fix tez-tfile parser documentation

Rajesh Balamohan created TEZ-3079: - Summary: Fix tez-tfile parser documentation Key: TEZ-3079 URL: https://issues.apache.org/jira/browse/TEZ-3079 Project: Apache Tez Issue Type: Improvement

[jira] [Updated] (TEZ-3079) Fix tez-tfile parser documentation

[ https://issues.apache.org/jira/browse/TEZ-3079?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-3079: -- Attachment: TEZ-3079.1.patch [~gopalv] - Please review. > Fix tez-tfile parser documentation >

[jira] [Commented] (TEZ-3076) Reduce merge memory overhead to support large number of in-memory mapoutputs

[ https://issues.apache.org/jira/browse/TEZ-3076?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118630#comment-15118630 ] Rajesh Balamohan commented on TEZ-3076: --- Thanks [~jeaglesham]. 2 minor comments 1.. This removes the

[jira] [Commented] (TEZ-3076) Reduce merge memory overhead to support large number of in-memory mapoutputs

[ https://issues.apache.org/jira/browse/TEZ-3076?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118789#comment-15118789 ] Rajesh Balamohan commented on TEZ-3076: --- Yeah, thanks [~gopalv]. > Reduce merge memory overhead to

[jira] [Comment Edited] (TEZ-3076) Reduce merge memory overhead to support large number of in-memory mapoutputs

[ https://issues.apache.org/jira/browse/TEZ-3076?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15118630#comment-15118630 ] Rajesh Balamohan edited comment on TEZ-3076 at 1/27/16 7:25 AM: Thanks

[jira] [Resolved] (TEZ-2898) tez tools : swimlanes.py is broken in master

[ https://issues.apache.org/jira/browse/TEZ-2898?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan resolved TEZ-2898. --- Resolution: Fixed Hadoop Flags: Reviewed Fix Version/s: 0.8.3 Target

[jira] [Commented] (TEZ-2898) tez tools : swimlanes.py is broken in master

[ https://issues.apache.org/jira/browse/TEZ-2898?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15107710#comment-15107710 ] Rajesh Balamohan commented on TEZ-2898: --- Will commit .1 version shortly as the reconstruction of

[jira] [Updated] (TEZ-2898) tez tools : swimlanes.py is broken in master

[ https://issues.apache.org/jira/browse/TEZ-2898?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2898: -- Attachment: TEZ-2898.2.patch Thanks [~gopalv]. Attaching .2 version, which tries to construct

[jira] [Commented] (TEZ-2914) Ability to limit vertex concurrency

[ https://issues.apache.org/jira/browse/TEZ-2914?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15070640#comment-15070640 ] Rajesh Balamohan commented on TEZ-2914: --- Yes, that would avoid the map get call in normal path. Rest

[jira] [Commented] (TEZ-2914) Ability to limit vertex concurrency

[ https://issues.apache.org/jira/browse/TEZ-2914?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15069370#comment-15069370 ] Rajesh Balamohan commented on TEZ-2914: --- In most cases, concurrency would be set to -1 (as in

[jira] [Commented] (TEZ-2538) ADDITIONAL_SPILL_COUNT wrongly populated for DefaultSorter with multiple partitions

[ https://issues.apache.org/jira/browse/TEZ-2538?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15061048#comment-15061048 ] Rajesh Balamohan commented on TEZ-2538: --- committed to branch-0.7 >> commit

[jira] [Updated] (TEZ-2538) ADDITIONAL_SPILL_COUNT wrongly populated for DefaultSorter with multiple partitions

[ https://issues.apache.org/jira/browse/TEZ-2538?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2538: -- Fix Version/s: (was: 0.8.0-alpha) 0.7.1 > ADDITIONAL_SPILL_COUNT wrongly

[jira] [Commented] (TEZ-2538) ADDITIONAL_SPILL_COUNT wrongly populated for DefaultSorter with multiple partitions

[ https://issues.apache.org/jira/browse/TEZ-2538?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15060081#comment-15060081 ] Rajesh Balamohan commented on TEZ-2538: --- [~bikassaha], [~zjffdu] - Would be good to add this in 0.7.x?

[jira] [Commented] (TEZ-2918) Make progress notifications in IOs

[ https://issues.apache.org/jira/browse/TEZ-2918?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15057839#comment-15057839 ] Rajesh Balamohan commented on TEZ-2918: --- lgtm. +1. > Make progress notifications in IOs >

[jira] [Resolved] (TEZ-3001) Build broken against 2.2 in branch 0.7

[ https://issues.apache.org/jira/browse/TEZ-3001?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan resolved TEZ-3001. --- Resolution: Fixed Committed addendum patch in branch-0.7 and checked with hadoop 2.2.0 >>

[jira] [Assigned] (TEZ-3001) Build broken against 2.2 in branch 0.7

[ https://issues.apache.org/jira/browse/TEZ-3001?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan reassigned TEZ-3001: - Assignee: Rajesh Balamohan > Build broken against 2.2 in branch 0.7 >

[jira] [Commented] (TEZ-3001) Build broken against 2.2 in branch 0.7

[ https://issues.apache.org/jira/browse/TEZ-3001?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15059072#comment-15059072 ] Rajesh Balamohan commented on TEZ-3001: --- Addendum fix

[jira] [Updated] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

[ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2496: -- Fix Version/s: 0.7.1 > Consider scheduling tasks in ShuffleVertexManager based on the partition

[jira] [Commented] (TEZ-2987) TestVertexImpl.testTez2684 fails

[ https://issues.apache.org/jira/browse/TEZ-2987?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055234#comment-15055234 ] Rajesh Balamohan commented on TEZ-2987: --- Scratch my previous comment; its not possible to get a stage

[jira] [Updated] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

[ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2496: -- Attachment: TEZ-2496.9.branch-0.7.patch Rebasing this patch for branch-0.7 (after TEZ-2943,

[jira] [Updated] (TEZ-2684) ShuffleVertexManager.parsePartitionStats throws IllegalStateException: Stats should be initialized

[ https://issues.apache.org/jira/browse/TEZ-2684?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2684: -- Attachment: TEZ-2684.2.patch.branch-0.7 Rebasing the patch for branch-0.7. This would fail

[jira] [Commented] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

[ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055295#comment-15055295 ] Rajesh Balamohan commented on TEZ-2496: --- Committed to branch-0.7 >> commit

[jira] [Updated] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

[ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2496: -- Attachment: (was: TEZ-2496.9.branch-0.7.patch) > Consider scheduling tasks in

[jira] [Updated] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

[ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2496: -- Attachment: TEZ-2496.9.branch-0.7.patch > Consider scheduling tasks in ShuffleVertexManager based

[jira] [Comment Edited] (TEZ-2684) ShuffleVertexManager.parsePartitionStats throws IllegalStateException: Stats should be initialized

[ https://issues.apache.org/jira/browse/TEZ-2684?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055342#comment-15055342 ] Rajesh Balamohan edited comment on TEZ-2684 at 12/14/15 3:01 AM: - Rebasing

[jira] [Comment Edited] (TEZ-2684) ShuffleVertexManager.parsePartitionStats throws IllegalStateException: Stats should be initialized

[ https://issues.apache.org/jira/browse/TEZ-2684?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055342#comment-15055342 ] Rajesh Balamohan edited comment on TEZ-2684 at 12/14/15 3:02 AM: - Rebasing

[jira] [Updated] (TEZ-2684) ShuffleVertexManager.parsePartitionStats throws IllegalStateException: Stats should be initialized

[ https://issues.apache.org/jira/browse/TEZ-2684?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2684: -- Fix Version/s: 0.7.1 > ShuffleVertexManager.parsePartitionStats throws IllegalStateException:

[jira] [Updated] (TEZ-2993) TestHistoryParser fail in branch-0.7

[ https://issues.apache.org/jira/browse/TEZ-2993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2993: -- Attachment: TEZ-2993.branch-0.7.patch > TestHistoryParser fail in branch-0.7 >

[jira] [Commented] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055490#comment-15055490 ] Rajesh Balamohan commented on TEZ-2979: --- Committed to branch-0.7. >> commit

[jira] [Updated] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2979: -- Fix Version/s: 0.7.1 > FlakyTest: org.apache.tez.history.TestHistoryParser >

[jira] [Commented] (TEZ-2993) TestHistoryParser fail in branch-0.7

[ https://issues.apache.org/jira/browse/TEZ-2993?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15055470#comment-15055470 ] Rajesh Balamohan commented on TEZ-2993: --- TestHistoryParser is fixed in master with TEZ-2979 (haven't

[jira] [Updated] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2979: -- Attachment: TEZ-2979.branch-0.7.patch Rebasing for branch 0.7. Will commit shortly > FlakyTest:

[jira] [Commented] (TEZ-2987) TestVertexImpl.testTez2684 fails

[ https://issues.apache.org/jira/browse/TEZ-2987?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15050305#comment-15050305 ] Rajesh Balamohan commented on TEZ-2987: --- [~bikassaha] - In corner cases, it is possible to get

[jira] [Commented] (TEZ-2987) TestVertexImpl.testTez2684 fails

[ https://issues.apache.org/jira/browse/TEZ-2987?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15052142#comment-15052142 ] Rajesh Balamohan commented on TEZ-2987: --- onVertexStartedDone would be set to true when

[jira] [Updated] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2979: -- Attachment: TEZ-2979.1.patch Might be due to changes in TEZ-2581. [~zjffdu], [~bikassaha] - Can

[jira] [Updated] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2979: -- Target Version/s: (was: 0.8.3) > FlakyTest: org.apache.tez.history.TestHistoryParser >

[jira] [Updated] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2979: -- Target Version/s: 0.8.3 > FlakyTest: org.apache.tez.history.TestHistoryParser >

[jira] [Assigned] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

[ https://issues.apache.org/jira/browse/TEZ-2979?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan reassigned TEZ-2979: - Assignee: Rajesh Balamohan > FlakyTest: org.apache.tez.history.TestHistoryParser >

[jira] [Created] (TEZ-2979) FlakyTest: org.apache.tez.history.TestHistoryParser

Rajesh Balamohan created TEZ-2979: - Summary: FlakyTest: org.apache.tez.history.TestHistoryParser Key: TEZ-2979 URL: https://issues.apache.org/jira/browse/TEZ-2979 Project: Apache Tez Issue

[jira] [Commented] (TEZ-2943) Auto reduce parallelism estimates badly on vertices with two disparate input sources

[ https://issues.apache.org/jira/browse/TEZ-2943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15048063#comment-15048063 ] Rajesh Balamohan commented on TEZ-2943: --- lgtm. +1. - maxThreshold logging may not be needed in the log

[jira] [Commented] (TEZ-2974) Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles

[ https://issues.apache.org/jira/browse/TEZ-2974?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15044541#comment-15044541 ] Rajesh Balamohan commented on TEZ-2974: --- test failure is unrelated to this change. > Tez tools:

[jira] [Updated] (TEZ-2974) Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles

[ https://issues.apache.org/jira/browse/TEZ-2974?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2974: -- Attachment: TEZ-2974.1.patch Attaching .1 patch. Instead of reading entire TFile into buffer

[jira] [Assigned] (TEZ-2974) Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles

[ https://issues.apache.org/jira/browse/TEZ-2974?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan reassigned TEZ-2974: - Assignee: Rajesh Balamohan > Tez tools: TFileRecordReader in tez-tools should support

[jira] [Created] (TEZ-2974) Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles

Rajesh Balamohan created TEZ-2974: - Summary: Tez tools: TFileRecordReader in tez-tools should support reading >2 GB tfiles Key: TEZ-2974 URL: https://issues.apache.org/jira/browse/TEZ-2974 Project:

[jira] [Commented] (TEZ-2964) Tez AMScheduler - assign to unallocated container

[ https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15029734#comment-15029734 ] Rajesh Balamohan commented on TEZ-2964: --- Should "tez.dag.recovery.max.unflushed.events" be bumped up

[jira] [Commented] (TEZ-2956) Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0

[ https://issues.apache.org/jira/browse/TEZ-2956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15025719#comment-15025719 ] Rajesh Balamohan commented on TEZ-2956: --- lgtm. +1 minor: Should

[jira] [Updated] (TEZ-2956) Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0

[ https://issues.apache.org/jira/browse/TEZ-2956?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2956: -- Attachment: TEZ-2956.2.patch Attaching the revised patch. Retaining the assumptions about the

[jira] [Created] (TEZ-2956) Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0

Rajesh Balamohan created TEZ-2956: - Summary: Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0 Key: TEZ-2956 URL: https://issues.apache.org/jira/browse/TEZ-2956 Project:

[jira] [Updated] (TEZ-2956) Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0

[ https://issues.apache.org/jira/browse/TEZ-2956?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2956: -- Attachment: With_Patch.png Without_Patch.png TEZ-2956_DAG.png >

[jira] [Commented] (TEZ-2956) Handle auto-reduce parallelism when the totalNumBipartiteSourceTasks is 0

[ https://issues.apache.org/jira/browse/TEZ-2956?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15015337#comment-15015337 ] Rajesh Balamohan commented on TEZ-2956: --- [~bikassaha] - Please review when you find time. > Handle

[jira] [Commented] (TEZ-2950) Poor performance of UnorderedPartitionedKVWriter

[ https://issues.apache.org/jira/browse/TEZ-2950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15012492#comment-15012492 ] Rajesh Balamohan commented on TEZ-2950: --- sizePerBuffer seems to be low as per the log. Can you

[jira] [Commented] (TEZ-2480) TEZ-2003: exception when closing output (ignored)

[ https://issues.apache.org/jira/browse/TEZ-2480?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15012273#comment-15012273 ] Rajesh Balamohan commented on TEZ-2480: --- lgtm. +1 > TEZ-2003: exception when closing output

[jira] [Commented] (TEZ-1491) Tez reducer-side merge's counter update is slow

[ https://issues.apache.org/jira/browse/TEZ-1491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15009894#comment-15009894 ] Rajesh Balamohan commented on TEZ-1491: --- In merger case, this would be beneficial if we have many

[jira] [Resolved] (TEZ-2030) TEZ-UI some vertex level counters are not populated in dag-view

[ https://issues.apache.org/jira/browse/TEZ-2030?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan resolved TEZ-2030. --- Resolution: Cannot Reproduce Haven't seen this in a while with recent builds. Not sure if it

[jira] [Updated] (TEZ-2918) Make progress notifications in IOs

[ https://issues.apache.org/jira/browse/TEZ-2918?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2918: -- Attachment: jmh-microbenchmark.tar.gz Tried out a simple JMH microbenchmark which updates

[jira] [Commented] (TEZ-2918) Make progress notifications in IOs

[ https://issues.apache.org/jira/browse/TEZ-2918?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14984592#comment-14984592 ] Rajesh Balamohan commented on TEZ-2918: --- Patch introduces volatile updates in inner loops of the

[jira] [Comment Edited] (TEZ-2918) Make progress notifications in IOs

[ https://issues.apache.org/jira/browse/TEZ-2918?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14984592#comment-14984592 ] Rajesh Balamohan edited comment on TEZ-2918 at 11/2/15 3:22 AM: Patch

[jira] [Comment Edited] (TEZ-2814) ATSImportTool has a return statement in a finally block

[ https://issues.apache.org/jira/browse/TEZ-2814?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14980686#comment-14980686 ] Rajesh Balamohan edited comment on TEZ-2814 at 10/29/15 4:06 PM: - [~hitesh]

[jira] [Assigned] (TEZ-2814) ATSImportTool has a return statement in a finally block

[ https://issues.apache.org/jira/browse/TEZ-2814?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan reassigned TEZ-2814: - Assignee: Rajesh Balamohan > ATSImportTool has a return statement in a finally block >

[jira] [Comment Edited] (TEZ-2244) PipelinedSorter: Progressive allocation for sort-buffers

[ https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978725#comment-14978725 ] Rajesh Balamohan edited comment on TEZ-2244 at 10/28/15 4:44 PM: -

[jira] [Updated] (TEZ-2244) PipelinedSorter: Progressive allocation for sort-buffers

[ https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2244: -- Attachment: TEZ-2244.6.patch Made minor changes to comments. > PipelinedSorter: Progressive

[jira] [Comment Edited] (TEZ-2244) PipelinedSorter: Progressive allocation for sort-buffers

[ https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978808#comment-14978808 ] Rajesh Balamohan edited comment on TEZ-2244 at 10/28/15 5:23 PM: - [~hitesh]

[jira] [Commented] (TEZ-2244) PipelinedSorter: Progressive allocation for sort-buffers

[ https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978808#comment-14978808 ] Rajesh Balamohan commented on TEZ-2244: --- [~hitesh] - It depends on the sort buffers. So if > 2GB is

[jira] [Updated] (TEZ-2244) PipelinedSorter: Progressive allocation for sort-buffers

[ https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2244: -- Attachment: TEZ-2244.8.patch Fixing the javadoc issue. Will commit shortly. > PipelinedSorter:

[jira] [Updated] (TEZ-2021) Tez tool to analyze shuffle performance in large clusters by mining task logs

[ https://issues.apache.org/jira/browse/TEZ-2021?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Rajesh Balamohan updated TEZ-2021: -- Attachment: TEZ-2021.3.patch Rebasing the patch as the log formats have changed in the recent

[jira] [Commented] (TEZ-2849) Implement Specific Workaround for JDK-8026049 & JDK-8073093

[ https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14976975#comment-14976975 ] Rajesh Balamohan commented on TEZ-2849: --- lgtm. +1. Is the explicit cast on SpanIterator intensional

<    2   3   4   5   6   7   8   9   10   11   >