[jira] [Comment Edited] (SPARK-18085) SPIP: Better History Server scalability for many / large applications
[ https://issues.apache.org/jira/browse/SPARK-18085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16148093#comment-16148093 ] Marcelo Vanzin edited comment on SPARK-18085 at 9/1/17 6:37 PM: [~jincheng] do you have code that can reproduce that? The code in the exception hasn't really changed, so this is probably an artifact of how the new code is recording data from the application. From your description (when we clicking stages with no tasks successful) I haven't been able to reproduce this; a stage that has only failed tasks still renders fine with the code in my branch. was (Author: vanzin): [~jincheng] do you have code that can reproduce that? The code in the exception hasn't really changed, so this is probably an artifact of how the new code is recording data from the application. For your description (when we clicking stages with no tasks successful) I haven't been able to reproduce this; a stage that has only failed tasks still renders fine with the code in my branch. > SPIP: Better History Server scalability for many / large applications > - > > Key: SPARK-18085 > URL: https://issues.apache.org/jira/browse/SPARK-18085 > Project: Spark > Issue Type: Umbrella > Components: Spark Core, Web UI >Affects Versions: 2.0.0 >Reporter: Marcelo Vanzin > Labels: SPIP > Attachments: spark_hs_next_gen.pdf > > > It's a known fact that the History Server currently has some annoying issues > when serving lots of applications, and when serving large applications. > I'm filing this umbrella to track work related to addressing those issues. > I'll be attaching a document shortly describing the issues and suggesting a > path to how to solve them. -- This message was sent by Atlassian JIRA (v6.4.14#64029) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Comment Edited] (SPARK-18085) SPIP: Better History Server scalability for many / large applications
[ https://issues.apache.org/jira/browse/SPARK-18085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16125150#comment-16125150 ] duyanghao edited comment on SPARK-18085 at 8/14/17 2:32 AM: [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time.What you have done is for quick loading logs for subsequent use. I am afraid that only solves one part of above problems(for subsequent use),and one more important problem is that how to quick load logs at the first time. was (Author: duyanghao): [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time.What you have done is for quick loading logs for subsequent use. I am afraid that only solves one part of above problems(for subsequent use),one more important problem is that how to quick load logs at the first time. > SPIP: Better History Server scalability for many / large applications > - > > Key: SPARK-18085 > URL: https://issues.apache.org/jira/browse/SPARK-18085 > Project: Spark > Issue Type: Umbrella > Components: Spark Core, Web UI >Affects Versions: 2.0.0 >Reporter: Marcelo Vanzin > Labels: SPIP > Attachments: spark_hs_next_gen.pdf > > > It's a known fact that the History Server currently has some annoying issues > when serving lots of applications, and when serving large applications. > I'm filing this umbrella to track work related to addressing those issues. > I'll be attaching a document shortly describing the issues and suggesting a > path to how to solve them. -- This message was sent by Atlassian JIRA (v6.4.14#64029) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Comment Edited] (SPARK-18085) SPIP: Better History Server scalability for many / large applications
[ https://issues.apache.org/jira/browse/SPARK-18085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16125150#comment-16125150 ] duyanghao edited comment on SPARK-18085 at 8/14/17 2:31 AM: [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time.What you have done is for quick loading logs for subsequent use. I am afraid that only solves one part of above problems(for subsequent use),one more important problem is that how to quick load logs at the first time. was (Author: duyanghao): [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time.What you have done is for quick loading logs for subsequent use. > SPIP: Better History Server scalability for many / large applications > - > > Key: SPARK-18085 > URL: https://issues.apache.org/jira/browse/SPARK-18085 > Project: Spark > Issue Type: Umbrella > Components: Spark Core, Web UI >Affects Versions: 2.0.0 >Reporter: Marcelo Vanzin > Labels: SPIP > Attachments: spark_hs_next_gen.pdf > > > It's a known fact that the History Server currently has some annoying issues > when serving lots of applications, and when serving large applications. > I'm filing this umbrella to track work related to addressing those issues. > I'll be attaching a document shortly describing the issues and suggesting a > path to how to solve them. -- This message was sent by Atlassian JIRA (v6.4.14#64029) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Comment Edited] (SPARK-18085) SPIP: Better History Server scalability for many / large applications
[ https://issues.apache.org/jira/browse/SPARK-18085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16125150#comment-16125150 ] duyanghao edited comment on SPARK-18085 at 8/14/17 2:28 AM: [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time.What you have done is for quick loading logs for subsequent use. was (Author: duyanghao): [~vanzin] so what you mean is that your project has done nothing about loading large event logs(100G+) and loading history summary page when the event log directory is large at the first time. > SPIP: Better History Server scalability for many / large applications > - > > Key: SPARK-18085 > URL: https://issues.apache.org/jira/browse/SPARK-18085 > Project: Spark > Issue Type: Umbrella > Components: Spark Core, Web UI >Affects Versions: 2.0.0 >Reporter: Marcelo Vanzin > Labels: SPIP > Attachments: spark_hs_next_gen.pdf > > > It's a known fact that the History Server currently has some annoying issues > when serving lots of applications, and when serving large applications. > I'm filing this umbrella to track work related to addressing those issues. > I'll be attaching a document shortly describing the issues and suggesting a > path to how to solve them. -- This message was sent by Atlassian JIRA (v6.4.14#64029) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Comment Edited] (SPARK-18085) SPIP: Better History Server scalability for many / large applications
[ https://issues.apache.org/jira/browse/SPARK-18085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16124421#comment-16124421 ] duyanghao edited comment on SPARK-18085 at 8/12/17 3:11 AM: [~vanzin] Could you have a summary of your progress(solved problems and unsolved problems) since we all have the same problems for loading large event logs(100G+) and loading history summary page when the event log directory is large. We all want to know how good your solution is(of course from your test results) and your plans to do it. was (Author: duyanghao): [~vanzin] Could you have a summary of your progress(solved problems and unsolved problems) since we all have the same problems for loading large event logs(100G+) and loading history summary page when the event log directory is large. We all want to know how good your solution is(of course from your test results)? > SPIP: Better History Server scalability for many / large applications > - > > Key: SPARK-18085 > URL: https://issues.apache.org/jira/browse/SPARK-18085 > Project: Spark > Issue Type: Umbrella > Components: Spark Core, Web UI >Affects Versions: 2.0.0 >Reporter: Marcelo Vanzin > Labels: SPIP > Attachments: spark_hs_next_gen.pdf > > > It's a known fact that the History Server currently has some annoying issues > when serving lots of applications, and when serving large applications. > I'm filing this umbrella to track work related to addressing those issues. > I'll be attaching a document shortly describing the issues and suggesting a > path to how to solve them. -- This message was sent by Atlassian JIRA (v6.4.14#64029) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org