[
https://issues.apache.org/jira/browse/HDFS-1547?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12978904#action_12978904
]
Suresh Srinivas commented on HDFS-1547:
---------------------------------------
Thanks guys for the comments.
Summary of proposed changes:
# When datanodes are added to exclude file:
#* Currently registered datanodes will be decommissioned. (same as today)
#* Previously registered datanodes after namenode start, when registers back
will be decommissioned. (same as today)
#* New datanodes will be allowed to register and will be decommissioned.
(changed)
#* After a node is decommissioned it is allowed to communicate with the
namenode. (changed).
#* Decommissioned node and decommissioning node free storage capacity does not
count towards the free storage capacity of the cluster. (changed)
# NameNode WebUI changes:
#* In cluster summary, live namenodes will show additional decommissioned node
count "Live Nodes : <Live node count> (Decommissioned <Decommissioned count>)"
#* In cluster summary, dead namenodes will show additional decommissioned node
count "Dead Nodes : <Dead node count> (Decommissioned <Decommissioned count>)"
#* In Live Nodes web page, in addition to live nodes listed today, a separate
table (not column) will list decommissioned nodes.
#* In Dead Nodes web page, in addition to live nodes listed today, a separate
table (not column) will list decommissioned nodes.
# I will rename the configuration parameter from "dfs.hosts.exclude" to
"dfs.hosts.decom". Will use key deprecation mechanism to support the older
config param "dfs.host.exclude", for backward compatibility.
# The documentation associated with include/exclude file is confusing and
incorrect in some places. Will update the doc as well.
There are other enhancements that came out of our discussions. I will open
jiras to track them.
> Improve decommission mechanism
> ------------------------------
>
> Key: HDFS-1547
> URL: https://issues.apache.org/jira/browse/HDFS-1547
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: name-node
> Affects Versions: 0.23.0
> Reporter: Suresh Srinivas
> Assignee: Suresh Srinivas
> Fix For: 0.23.0
>
>
> Current decommission mechanism driven using exclude file has several issues.
> This bug proposes some changes in the mechanism for better manageability. See
> the proposal in the next comment for more details.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
