You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ambari.apache.org by "Dmytro Sen (JIRA)" <ji...@apache.org> on 2015/04/20 17:18:59 UTC
[jira] [Commented] (AMBARI-10602) HDFS namenode_opt_newsize and
namenode_opt_maxnewsize not using number of datanodes in calculation
[ https://issues.apache.org/jira/browse/AMBARI-10602?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14503009#comment-14503009 ]
Dmytro Sen commented on AMBARI-10602:
-------------------------------------
local tests result
{noformat}
----------------------------------------------------------------------
Ran 263 tests in 8.238s
OK
----------------------------------------------------------------------
Total run:728
Total errors:0
Total failures:0
OK
{noformat}
> HDFS namenode_opt_newsize and namenode_opt_maxnewsize not using number of datanodes in calculation
> --------------------------------------------------------------------------------------------------
>
> Key: AMBARI-10602
> URL: https://issues.apache.org/jira/browse/AMBARI-10602
> Project: Ambari
> Issue Type: Bug
> Components: stacks
> Affects Versions: 2.1.0
> Reporter: Dmytro Sen
> Assignee: Dmytro Sen
> Fix For: 2.1.0
>
> Attachments: AMBARI-10602.patch
>
>
> Reviewing HDFS recommendation code for namenode_heapsize, namenode_opt_newsize and namenode_opt_maxnewsize, it looks like the calculations are not taking into account the number of datanodes and disks per datanode.
> HDFS recommendations currently gives the following output
> {code}
> "configurations": {
> "hadoop-env": {
> "properties": {
> "namenode_heapsize": "1024",
> "namenode_opt_maxnewsize": "256",
> "namenode_opt_newsize": "256"
> },
> "property_attributes": {
> "namenode_heapsize": {
> "maximum": "2887"
> },
> "dtnode_heapsize": {
> "maximum": "2887"
> }
> }
> },
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)