[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-23 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14145541#comment-14145541 ] Hari Shreedharan commented on SPARK-3129: - Sure. Thanks Matei! > Prevent data los

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-23 Thread Matei Zaharia (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14145537#comment-14145537 ] Matei Zaharia commented on SPARK-3129: -- Alright, in that case, this sounds pretty goo

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-23 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14145411#comment-14145411 ] Hari Shreedharan commented on SPARK-3129: - It is per node, single threaded. > Pre

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-23 Thread Matei Zaharia (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14145324#comment-14145324 ] Matei Zaharia commented on SPARK-3129: -- Is that 100 MB/s per node or in total? That s

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-22 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14143775#comment-14143775 ] Hari Shreedharan commented on SPARK-3129: - I did multiple rounds of testing and it

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-19 Thread Matei Zaharia (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14141382#comment-14141382 ] Matei Zaharia commented on SPARK-3129: -- So Hari, what is the maximum sustainable rate

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-18 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14140049#comment-14140049 ] Hari Shreedharan commented on SPARK-3129: - Do these numbers look ok enough to you

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-18 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14140048#comment-14140048 ] Hari Shreedharan commented on SPARK-3129: - Reducing the buffer size decreases the

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-18 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14140033#comment-14140033 ] Hari Shreedharan commented on SPARK-3129: - So I did some benchmarking on EC2, writ

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-18 Thread Saisai Shao (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14138706#comment-14138706 ] Saisai Shao commented on SPARK-3129: Strongly agree with Matei's comment, I think we c

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-17 Thread Matei Zaharia (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14138281#comment-14138281 ] Matei Zaharia commented on SPARK-3129: -- Great, it will be nice to see how fast this i

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-17 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14138032#comment-14138032 ] Hari Shreedharan commented on SPARK-3129: - Thanks Matei for the background. I had

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-17 Thread Matei Zaharia (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14138014#comment-14138014 ] Matei Zaharia commented on SPARK-3129: -- Hari, have you actually benchmarked a WAL bas

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-16 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14135841#comment-14135841 ] Hari Shreedharan commented on SPARK-3129: - As long as at least one executor contai

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-16 Thread Patrick Wendell (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14135789#comment-14135789 ] Patrick Wendell commented on SPARK-3129: I think for this it's worth considering a

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-16 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14135116#comment-14135116 ] Hari Shreedharan commented on SPARK-3129: - It looks like Akka makes it difficult t

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-10 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14129279#comment-14129279 ] Hari Shreedharan commented on SPARK-3129: - [~sowen] Thanks! That fixed the issue!

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-10 Thread Sean Owen (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14129256#comment-14129256 ] Sean Owen commented on SPARK-3129: -- [~hshreedharan] Just manually add the src dir in the

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-10 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14129252#comment-14129252 ] Hari Shreedharan commented on SPARK-3129: - FYI here is the branch where I am doing

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-08 Thread Thomas Graves (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125981#comment-14125981 ] Thomas Graves commented on SPARK-3129: -- yes that should be enough. > Prevent data lo

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-08 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125828#comment-14125828 ] Hari Shreedharan commented on SPARK-3129: - Correct me if I am wrong here, it looks

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-08 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125810#comment-14125810 ] Hari Shreedharan commented on SPARK-3129: - (I am not too familiar with how UGI get

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-08 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125790#comment-14125790 ] Hari Shreedharan commented on SPARK-3129: - [~tgraves] - It looks like the Security

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-08 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125784#comment-14125784 ] Hari Shreedharan commented on SPARK-3129: - Hi Saisai, You are correct that there

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Saisai Shao (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122274#comment-14122274 ] Saisai Shao commented on SPARK-3129: Hi [~hshreedharan]], thanks for your reply, is th

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122114#comment-14122114 ] Hari Shreedharan commented on SPARK-3129: - Looks like simply moving the code that

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122103#comment-14122103 ] Hari Shreedharan commented on SPARK-3129: - I am less worried about client mode, si

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Thomas Graves (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122064#comment-14122064 ] Thomas Graves commented on SPARK-3129: -- On yarn, it generates the secret automaticall

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122017#comment-14122017 ] Hari Shreedharan commented on SPARK-3129: - [~tgraves] - Am I correct in assuming t

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122006#comment-14122006 ] Hari Shreedharan commented on SPARK-3129: - Yes, so my initial goal is to be able t

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-09-04 Thread Saisai Shao (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14121160#comment-14121160 ] Saisai Shao commented on SPARK-3129: Hi [~hshreedharan], one more question: Is your d

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-21 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14105928#comment-14105928 ] Hari Shreedharan commented on SPARK-3129: - [~tgraves] - Thanks for the pointers. Y

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-20 Thread Saisai Shao (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14105086#comment-14105086 ] Saisai Shao commented on SPARK-3129: Hi Hari, I have some high level questions about t

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-20 Thread Thomas Graves (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14104077#comment-14104077 ] Thomas Graves commented on SPARK-3129: -- Yes that probably means using reflection. I

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-19 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14102579#comment-14102579 ] Hari Shreedharan commented on SPARK-3129: - The way the driver "finds" the executor

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-19 Thread Thomas Graves (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14102504#comment-14102504 ] Thomas Graves commented on SPARK-3129: -- A couple of random thoughts on this for yarn.

[jira] [Commented] (SPARK-3129) Prevent data loss in Spark Streaming

2014-08-19 Thread Hari Shreedharan (JIRA)
[ https://issues.apache.org/jira/browse/SPARK-3129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14102501#comment-14102501 ] Hari Shreedharan commented on SPARK-3129: - This doc is an early list of fixes. I m