[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-19 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14140387#comment-14140387 ] Hudson commented on HDFS-6840: -- SUCCESS: Integrated in Hadoop-Yarn-trunk #685 (See

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-19 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14140539#comment-14140539 ] Hudson commented on HDFS-6840: -- FAILURE: Integrated in Hadoop-Mapreduce-trunk #1901 (See

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-19 Thread Hudson (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14140593#comment-14140593 ] Hudson commented on HDFS-6840: -- FAILURE: Integrated in Hadoop-Hdfs-trunk #1876 (See

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-18 Thread Aaron T. Myers (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14139643#comment-14139643 ] Aaron T. Myers commented on HDFS-6840: -- Latest patch looks good to me, +1. I agree

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-18 Thread Andrew Wang (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14139661#comment-14139661 ] Andrew Wang commented on HDFS-6840: --- Filed HADOOP-11107 as a follow-on for the random

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-18 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14139797#comment-14139797 ] Hadoop QA commented on HDFS-6840: - {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-18 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14139813#comment-14139813 ] Hadoop QA commented on HDFS-6840: - {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-18 Thread Aaron T. Myers (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14139826#comment-14139826 ] Aaron T. Myers commented on HDFS-6840: -- +1, branch-2 patch looks good to me. Thanks,

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-09-09 Thread Daryn Sharp (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14127156#comment-14127156 ] Daryn Sharp commented on HDFS-6840: --- In addition to Jason's comment, I'm mildly concerned

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-29 Thread Jason Lowe (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14115859#comment-14115859 ] Jason Lowe commented on HDFS-6840: -- The test failures appear to be unrelated, and they

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-28 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14113557#comment-14113557 ] Hadoop QA commented on HDFS-6840: - {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-28 Thread Yi Liu (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14113720#comment-14113720 ] Yi Liu commented on HDFS-6840: -- Thanks [~andrew.wang], the patch looks good to me. Just one

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-28 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14114661#comment-14114661 ] Hadoop QA commented on HDFS-6840: - {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-28 Thread Yi Liu (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14114667#comment-14114667 ] Yi Liu commented on HDFS-6840: -- Thanks [~andrew.wang], +1 (non-binding) Clients are always

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-21 Thread Ashwin Shankar (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14106428#comment-14106428 ] Ashwin Shankar commented on HDFS-6840: -- Since we are planning to make a 2.5.1 release,

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-13 Thread Daryn Sharp (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14095499#comment-14095499 ] Daryn Sharp commented on HDFS-6840: --- I'd like complete removal of the random seed. Why

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-13 Thread Daryn Sharp (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14095763#comment-14095763 ] Daryn Sharp commented on HDFS-6840: --- We believe but haven't proven that this

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-12 Thread Daryn Sharp (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14094111#comment-14094111 ] Daryn Sharp commented on HDFS-6840: --- bq. My impression of the old pseudo-sort was that

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-12 Thread Andrew Wang (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14094340#comment-14094340 ] Andrew Wang commented on HDFS-6840: --- Alright, I'm sufficiently convinced :) Is the fix as

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-12 Thread Jason Lowe (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14094706#comment-14094706 ] Jason Lowe commented on HDFS-6840: -- Commenting out the setSeed call in NetworkTopology on

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-12 Thread Yi Liu (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14095165#comment-14095165 ] Yi Liu commented on HDFS-6840: -- Simplest way is to set

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-11 Thread Jason Lowe (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14093584#comment-14093584 ] Jason Lowe commented on HDFS-6840: -- HDFS-6701 gives the option to randomize the returned

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-11 Thread Andrew Wang (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14093595#comment-14093595 ] Andrew Wang commented on HDFS-6840: --- Hey Jason, did you bisect this behavior to

[jira] [Commented] (HDFS-6840) Clients are always sent to the same datanode when read is off rack

2014-08-11 Thread Jason Lowe (JIRA)
[ https://issues.apache.org/jira/browse/HDFS-6840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14093613#comment-14093613 ] Jason Lowe commented on HDFS-6840: -- I think the previous behavior was not deterministic