[
https://issues.apache.org/jira/browse/HDFS-11360?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lantao Jin updated HDFS-11360:
------------------------------
Description:
With distcp configuration, there are two or more "nameservices" setting in
hdfs-site.xml, for example:
{code}
<property>
<name>dfs.nameservices</name>
<value>one-nn-ha,two-nn-ha</value>
</property>
{code}
If the HDFS Balancer also launches in that node, it will create two IPC threads
to connect both of NNs. And block removing both happens in this Balancer.
Although I didn't find any block removing between different clusters, the
behavior still became weird and unexpected.
So the best way is adding a configuration that appoint to a decided nameservice
to launch. I can offer a patch. Any better ideal?
was:
With distcp configuration, there are two or more "nameservices" setting in
hdfs-site.xml, for example:
{code}
<property>
<name>dfs.nameservices</name>
<value>one-nn-ha,two-nn-ha</value>
</property>
{code}
If the HDFS Balancer also launches in that node, it will create two IPC threads
to connect both of NNs. And block removing also happens in the this Balancer.
Although I didn't find any block removing between different nodes, the behavior
still became weird and unexpected.
So the best way is adding a configuration that appoint to a decided nameservice
to launch. I can offer a patch. Any better ideal?
> HDFS balancer need to appoint to a decided nameservice
> ------------------------------------------------------
>
> Key: HDFS-11360
> URL: https://issues.apache.org/jira/browse/HDFS-11360
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: balancer & mover
> Reporter: Lantao Jin
> Priority: Minor
>
> With distcp configuration, there are two or more "nameservices" setting in
> hdfs-site.xml, for example:
> {code}
> <property>
> <name>dfs.nameservices</name>
> <value>one-nn-ha,two-nn-ha</value>
> </property>
> {code}
> If the HDFS Balancer also launches in that node, it will create two IPC
> threads to connect both of NNs. And block removing both happens in this
> Balancer. Although I didn't find any block removing between different
> clusters, the behavior still became weird and unexpected.
> So the best way is adding a configuration that appoint to a decided
> nameservice to launch. I can offer a patch. Any better ideal?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
