[jira] [Commented] (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13096253#comment-13096253 ] Jitendra Nath Pandey commented on HDFS-826: --- Uploaded patch for 20-security

[jira] [Commented] (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13096267#comment-13096267 ] Suresh Srinivas commented on HDFS-826: -- +1 for the patch. I committed the patch to

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12876756#action_12876756 ] Nicolas Spiegelberg commented on HDFS-826: -- Note: HDFS-826-0.20-v2.patch can be

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12853869#action_12853869 ] Hudson commented on HDFS-826: - Integrated in Hdfs-Patch-h5.grid.sp2.yahoo.net #302 (See

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12853959#action_12853959 ] Hudson commented on HDFS-826: - Integrated in Hadoop-Hdfs-trunk #275 (See

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12853660#action_12853660 ] Hudson commented on HDFS-826: - Integrated in Hdfs-Patch-h2.grid.sp2.yahoo.net #146 (See

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12845232#action_12845232 ] Hudson commented on HDFS-826: - Integrated in Hadoop-Hdfs-trunk-Commit #214 (See

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12844119#action_12844119 ] stack commented on HDFS-826: Just to say that I'm +1 on committing Replicable4.txt (I've already

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12843225#action_12843225 ] Hairong Kuang commented on HDFS-826: hope this is ok with you and thanks for reviewing

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841754#action_12841754 ] dhruba borthakur commented on HDFS-826: --- Hi Hairong, Thanks for the review. I would

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12842204#action_12842204 ] stack commented on HDFS-826: Trying to make a version of this patch for 0.20 and having a bit of

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841424#action_12841424 ] Hairong Kuang commented on HDFS-826: Dhruba, sorry for this late thought. :-) It might be

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841494#action_12841494 ] dhruba borthakur commented on HDFS-826: --- Hauirong: it makes the application code easier

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841717#action_12841717 ] Hairong Kuang commented on HDFS-826: I do not think that making the user aware of whether

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12840555#action_12840555 ] Hadoop QA commented on HDFS-826: -1 overall. Here are the results of testing the latest

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841021#action_12841021 ] dhruba borthakur commented on HDFS-826: --- anybody wants to re-review this one? Thanks.

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12841051#action_12841051 ] stack commented on HDFS-826: Patch looks good to me (Didn't try it, just reviewed patch). Allow

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12840285#action_12840285 ] stack commented on HDFS-826: @ Konstantin It looks like DFSOS implements it: {code} +class

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12840287#action_12840287 ] stack commented on HDFS-826: If making a new patch, take a look at the javadoc for the method

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12840297#action_12840297 ] Hairong Kuang commented on HDFS-826: I do not think that Replicable interface is

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12839837#action_12839837 ] Todd Lipcon commented on HDFS-826: -- Patch looks good to me. I question whether returning 0

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12839853#action_12839853 ] dhruba borthakur commented on HDFS-826: --- Thanks Todd for the review. I agree with your

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12839858#action_12839858 ] Hadoop QA commented on HDFS-826: -1 overall. Here are the results of testing the latest

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12839875#action_12839875 ] Konstantin Shvachko commented on HDFS-826: -- What is the point of introducing new

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12799775#action_12799775 ] dhruba borthakur commented on HDFS-826: --- To clarify by client you mean FileContext

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12799464#action_12799464 ] ryan rawson commented on HDFS-826: -- Why wouldn't DFSClient just handle the write pipeline

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12799503#action_12799503 ] dhruba borthakur commented on HDFS-826: --- If the client gets an error it must close the

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12799622#action_12799622 ] Eli Collins commented on HDFS-826: -- To clarify by client you mean FileContext right? ie from

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12796810#action_12796810 ] Konstantin Shvachko commented on HDFS-826: -- I remember we intended to implement but

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12796278#action_12796278 ] dhruba borthakur commented on HDFS-826: --- My thinking is that most hadoop apps should be

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12794910#action_12794910 ] dhruba borthakur commented on HDFS-826: --- If we do not introduce Replicable, then

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12792502#action_12792502 ] dhruba borthakur commented on HDFS-826: --- One question that arises is what to return

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12792568#action_12792568 ] Hairong Kuang commented on HDFS-826: I suspect we should return t intended replication

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12791682#action_12791682 ] dhruba borthakur commented on HDFS-826: --- Hi Hairong, would you support creating a new

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12789932#action_12789932 ] Hairong Kuang commented on HDFS-826: Unfortunately FSDataoutputStream needs a new API to

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12789426#action_12789426 ] Hairong Kuang commented on HDFS-826: Would polling be good enough for your application?

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12789638#action_12789638 ] dhruba borthakur commented on HDFS-826: --- Would polling be good enough for your

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12788953#action_12788953 ] Hairong Kuang commented on HDFS-826: The problem is that if datanode# in the write

[jira] Commented: (HDFS-826) Allow a mechanism for an application to detect that datanode(s) have died in the write pipeline

[ https://issues.apache.org/jira/browse/HDFS-826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12788954#action_12788954 ] dhruba borthakur commented on HDFS-826: --- The application is such that it can start