You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Harsh J (JIRA)" <ji...@apache.org> on 2012/05/30 14:04:24 UTC
[jira] [Created] (HDFS-3475) Make the replication monitor
multipliers configurable
Harsh J created HDFS-3475:
-----------------------------
Summary: Make the replication monitor multipliers configurable
Key: HDFS-3475
URL: https://issues.apache.org/jira/browse/HDFS-3475
Project: Hadoop HDFS
Issue Type: Improvement
Affects Versions: 3.0.0
Reporter: Harsh J
Assignee: Harsh J
Priority: Trivial
BlockManager currently hardcodes the following two constants:
{code}
private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
{code}
These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13398912#comment-13398912 ]
Harsh J commented on HDFS-3475:
-------------------------------
Btw I finally played around with the block repl multiplier on a local cluster and I could notice increase in usage when a DN went down and the blocks began moving about, with a higher multiplier value. The logs grepped and counted too showed higher "asking DN to replicate" count per DN heartbeat at NN.
This should be cautiously increased though, cause even if your network is awesome, your disks cumulatively need to keep up with it anyway. And too much of payload per DN command transfer-back is bad as well. This is why the defaults, I've stuck them as-is and added warnings in documented props.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Suresh Srinivas (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Suresh Srinivas updated HDFS-3475:
----------------------------------
Release Note:
This change adds two new configuration parameters.
# {{dfs.namenode.invalidate.work.pct.per.iteration}} for controlling deletion rate of blocks.
# {{dfs.namenode.replication.work.multiplier.per.iteration}} for controlling replication rate. This in turn allows controlling the time it takes for decommissioning.
Please see hdfs-default.xml for detailed description.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.1.0-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Eli Collins (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403288#comment-13403288 ]
Eli Collins commented on HDFS-3475:
-----------------------------------
+1 Looks good. Style nit going forward: no need to use "this.foo" when referencing members in the constructor, understand you're doing so to be consistent with the surrounding code (which doesn't need to do so either)
Per offline conversation Harsh tested this on a real cluster with higher multiplier and verfied that re-replication proceeded in kind. Harsh, seems like the current defaults (which you've preserved here) should be revisited? They seem a little low to me, if you concur please file a separate jira for that (to raise them).
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Eli Collins (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13404241#comment-13404241 ]
Eli Collins commented on HDFS-3475:
-----------------------------------
Thanks Harsh, makes sense.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403176#comment-13403176 ]
Hadoop QA commented on HDFS-3475:
---------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12533831/HDFS-3475.patch
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
+1 javadoc. The javadoc tool did not generate any warning messages.
+1 eclipse:eclipse. The patch built with eclipse:eclipse.
+1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
+1 core tests. The patch passed unit tests in hadoop-hdfs-project/hadoop-hdfs.
+1 contrib tests. The patch passed contrib unit tests.
Test results: https://builds.apache.org/job/PreCommit-HDFS-Build/2717//testReport/
Console output: https://builds.apache.org/job/PreCommit-HDFS-Build/2717//console
This message is automatically generated.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Harsh J updated HDFS-3475:
--------------------------
Attachment: HDFS-3475.patch
I noticed I missed out on using one variable (the invalidating pct value). I've fixed that now.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Harsh J updated HDFS-3475:
--------------------------
Target Version/s: 2.0.1-alpha, 3.0.0 (was: 3.0.0)
Affects Version/s: (was: 3.0.0)
2.0.0-alpha
Status: Patch Available (was: Open)
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication and invalidation
multipliers configurable
Posted by "Eli Collins (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eli Collins updated HDFS-3475:
------------------------------
Summary: Make the replication and invalidation multipliers configurable (was: Make the replication monitor multipliers configurable)
> Make the replication and invalidation multipliers configurable
> --------------------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.2-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13399617#comment-13399617 ]
Hadoop QA commented on HDFS-3475:
---------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12533056/HDFS-3475.patch
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 javadoc. The javadoc tool appears to have generated 13 warning messages.
+1 eclipse:eclipse. The patch built with eclipse:eclipse.
+1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
-1 core tests. The patch failed these unit tests in hadoop-hdfs-project/hadoop-hdfs:
org.apache.hadoop.hdfs.server.datanode.TestBlockRecovery
org.apache.hadoop.hdfs.server.namenode.metrics.TestNameNodeMetrics
+1 contrib tests. The patch passed contrib unit tests.
Test results: https://builds.apache.org/job/PreCommit-HDFS-Build/2685//testReport/
Console output: https://builds.apache.org/job/PreCommit-HDFS-Build/2685//console
This message is automatically generated.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403324#comment-13403324 ]
Hudson commented on HDFS-3475:
------------------------------
Integrated in Hadoop-Hdfs-trunk-Commit #2471 (See [https://builds.apache.org/job/Hadoop-Hdfs-trunk-Commit/2471/])
HDFS-3475. Make the replication monitor multipliers configurable. Contributed by Harsh J Chouraria (Revision 1355089)
Result = SUCCESS
eli : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1355089
Files :
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403920#comment-13403920 ]
Hudson commented on HDFS-3475:
------------------------------
Integrated in Hadoop-Mapreduce-trunk #1124 (See [https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1124/])
HDFS-3475. Make the replication monitor multipliers configurable. Contributed by Harsh J Chouraria (Revision 1355089)
Result = FAILURE
eli : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1355089
Files :
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403847#comment-13403847 ]
Hudson commented on HDFS-3475:
------------------------------
Integrated in Hadoop-Hdfs-trunk #1091 (See [https://builds.apache.org/job/Hadoop-Hdfs-trunk/1091/])
HDFS-3475. Make the replication monitor multipliers configurable. Contributed by Harsh J Chouraria (Revision 1355089)
Result = FAILURE
eli : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1355089
Files :
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Closed] (HDFS-3475) Make the replication and invalidation
rates configurable
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy closed HDFS-3475.
-------------------------------
> Make the replication and invalidation rates configurable
> --------------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.2-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403449#comment-13403449 ]
Hudson commented on HDFS-3475:
------------------------------
Integrated in Hadoop-Mapreduce-trunk-Commit #2422 (See [https://builds.apache.org/job/Hadoop-Mapreduce-trunk-Commit/2422/])
HDFS-3475. Make the replication monitor multipliers configurable. Contributed by Harsh J Chouraria (Revision 1355089)
Result = FAILURE
eli : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1355089
Files :
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13398872#comment-13398872 ]
Harsh J commented on HDFS-3475:
-------------------------------
The javadocs warnings are not related to this patch. The tests warn I've explained in my previous comment.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Harsh J updated HDFS-3475:
--------------------------
Attachment: HDFS-3475.patch
Hi Aaron - Thanks! Can you take a look at this patch? I believe the existing tests already cover the invalidation and replication work and the defaults are what there were as constants previously. Let me know if you need me to do any more changes!
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 3.0.0
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Aaron T. Myers (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13400599#comment-13400599 ]
Aaron T. Myers commented on HDFS-3475:
--------------------------------------
One small comment: I think you should add some info to the hdfs-default.xml description for "{{dfs.namenode.invalidate.work.pct.per.iteration}}" saying that the value should be between 0-100, or whatever's appropriate. For that matter, since this is a brand new config, you might want to change it to be in the range 0 - 1.0, which I think is a more common way in the Hadoop code base to represent percentages.
Other than that the patch looks good. +1 pending a fix for the above and an explanation of the two test failures.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Adam Muise (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13426186#comment-13426186 ]
Adam Muise commented on HDFS-3475:
----------------------------------
Note on this value: while we tested at 100, this may be too high for a cluster under even moderate workload. The cluster in question had very powerful nodes and dual-bonded 10Gb interfaces. We also had to increase the DataNode memory in the range of 4-6Gb. You may choose to start the value at 10 and go up based on your available memory for the Datanode, I/O, and network capacity.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.1.0-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403332#comment-13403332 ]
Hudson commented on HDFS-3475:
------------------------------
Integrated in Hadoop-Common-trunk-Commit #2403 (See [https://builds.apache.org/job/Hadoop-Common-trunk-Commit/2403/])
HDFS-3475. Make the replication monitor multipliers configurable. Contributed by Harsh J Chouraria (Revision 1355089)
Result = SUCCESS
eli : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1355089
Files :
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/blockmanagement/BlockManager.java
* /hadoop/common/trunk/hadoop-hdfs-project/hadoop-hdfs/src/main/resources/hdfs-default.xml
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13427126#comment-13427126 ]
Harsh J commented on HDFS-3475:
-------------------------------
Thanks for following up Adam. Yes, and also since this is a multiplier config, growth of blocks will induce higher memory, so this must only be done if greatly needed and must be carefully monitored.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.1.0-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13398829#comment-13398829 ]
Hadoop QA commented on HDFS-3475:
---------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12532920/HDFS-3475.patch
against trunk revision .
+1 @author. The patch does not contain any @author tags.
-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.
+1 javac. The applied patch does not increase the total number of javac compiler warnings.
-1 javadoc. The javadoc tool appears to have generated 13 warning messages.
+1 eclipse:eclipse. The patch built with eclipse:eclipse.
+1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.
+1 release audit. The applied patch does not increase the total number of release audit warnings.
+1 core tests. The patch passed unit tests in hadoop-hdfs-project/hadoop-hdfs.
+1 contrib tests. The patch passed contrib unit tests.
Test results: https://builds.apache.org/job/PreCommit-HDFS-Build/2680//testReport/
Console output: https://builds.apache.org/job/PreCommit-HDFS-Build/2680//console
This message is automatically generated.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Eli Collins (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eli Collins updated HDFS-3475:
------------------------------
Resolution: Fixed
Fix Version/s: 2.0.1-alpha
Target Version/s: (was: 2.0.1-alpha, 3.0.0)
Hadoop Flags: Reviewed
Status: Resolved (was: Patch Available)
I've committed this and merged to branch-2. Thanks Harsh!
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Aaron T. Myers (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13285943#comment-13285943 ]
Aaron T. Myers commented on HDFS-3475:
--------------------------------------
I'm +1 to making these configurable.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 3.0.0
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403894#comment-13403894 ]
Harsh J commented on HDFS-3475:
-------------------------------
The value for the multiplier we used was 100. This gave us a good boost, but was tried on a cluster that had a very good network gear.
Yes I do feel the default of 2 is too low today. However, it is low to prevent unnecessary replication in face of long-taking DN restarts and for that reason it does well not to consume too much network. Hence I had left in the defaults. It has to be raised only for fast networks, otherwise the reads/writes may get choked. So I do not think we need to raise it, given the variance of users who use Hadoop.
Let me know if you feel otherwise Eli.
P.s. The commit HDFS build failed cause of HDFS-RAID project issues, not HDFS. No alarm, hence.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.1-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Suresh Srinivas (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Suresh Srinivas updated HDFS-3475:
----------------------------------
This change adds two new configuration parameters.
# {{dfs.namenode.invalidate.work.pct.per.iteration}} for controlling deletion rate of blocks.
# {{dfs.namenode.invalidate.work.pct.per.iteration}} for controlling replication rate. This in turn allows controlling the time it takes for decommissioning.
Please see hdfs-default.xml for detailed description.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.1.0-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication monitor
multipliers configurable
Posted by "Harsh J (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Harsh J updated HDFS-3475:
--------------------------
Attachment: HDFS-3475.patch
Updated patch addresses Aaron's comments and also adds in checks for value being {{-gt zero}}.
Both tests mentioned pass for me locally with and without patch applied. Was something on trunk at the time possibly, its gone now.
[With patch applied]
{code}
-------------------------------------------------------
T E S T S
-------------------------------------------------------
-------------------------------------------------------
T E S T S
-------------------------------------------------------
Running org.apache.hadoop.hdfs.server.datanode.TestBlockRecovery
Tests run: 12, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 4.581 sec
Running org.apache.hadoop.hdfs.server.namenode.metrics.TestNameNodeMetrics
gTests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 29.291 sec
{code}
Thanks Aaron. After a Jenkins +1 on this new patch, I shall commit it in, unless you have further comments on the patch.
> Make the replication monitor multipliers configurable
> -----------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HDFS-3475) Make the replication and invalidation
rates configurable
Posted by "Eli Collins (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-3475?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eli Collins updated HDFS-3475:
------------------------------
Summary: Make the replication and invalidation rates configurable (was: Make the replication and invalidation multipliers configurable)
> Make the replication and invalidation rates configurable
> --------------------------------------------------------
>
> Key: HDFS-3475
> URL: https://issues.apache.org/jira/browse/HDFS-3475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.0-alpha
> Reporter: Harsh J
> Assignee: Harsh J
> Priority: Trivial
> Fix For: 2.0.2-alpha
>
> Attachments: HDFS-3475.patch, HDFS-3475.patch, HDFS-3475.patch
>
>
> BlockManager currently hardcodes the following two constants:
> {code}
> private static final int INVALIDATE_WORK_PCT_PER_ITERATION = 32;
> private static final int REPLICATION_WORK_MULTIPLIER_PER_ITERATION = 2;
> {code}
> These are used to throttle/limit the amount of deletion and replication-to-other-DN work done per heartbeat interval of a live DN.
> Not many have had reasons to want these changed so far but there have been a few requests I've faced over the past year from a variety of clusters I've helped maintain. I think with the improvements in disks and network thats already started to be rolled out in production environments out there, changing these may start making sense to some.
> Lets at least make it advanced-configurable with proper docs that warn adequately, with the defaults being what they are today. With hardcodes, it comes down to a recompile for admins, which is not something they may like.
> Please let me know your thoughts.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira