[
https://issues.apache.org/jira/browse/HDFS-10368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15587810#comment-15587810
]
Rakesh R edited comment on HDFS-10368 at 10/19/16 6:25 AM:
---
Thanks [~andrew.wang] for the reminder. Earlier we had identified one property
and deprecated it as part of HDFS-9869.
{code}
https://hadoop.apache.org/docs/r3.0.0-alpha1/hadoop-project-dist/hadoop-common/DeprecatedProperties.html
dfs.replication.pending.timeout.sec =>
dfs.namenode.reconstruction.pending.timeout-sec
dfs.namenode.replication.pending.timeout-sec =>
dfs.namenode.reconstruction.pending.timeout-sec
{code}
Like mentioned, this jira to revisit the replication related configs, following
are few replication config items. I remember somewhere we discussed to consider
{{EC internal blocks are a special form of replicas}}, imho documentation would
be sufficient rather than renaming these configs to reconstruction or
redundancy. Do we have docs conveying the same? Makes sense to you?
{code}
dfs.namenode.replication.considerLoad
dfs.namenode.replication.considerLoad.factor
dfs.namenode.replication.min
dfs.namenode.replication.interval
dfs.namenode.replication.work.multiplier.per.iteration
dfs.namenode.replication.max-streams
dfs.namenode.replication.max-streams-hard-limit
{code}
was (Author: rakeshr):
Thanks [~andrew.wang] for the reminder. Earlier we had identified one property
and deprecated it as part of HDFS-9869.
{code}
https://hadoop.apache.org/docs/r3.0.0-alpha1/hadoop-project-dist/hadoop-common/DeprecatedProperties.html
dfs.replication.pending.timeout.sec =>
dfs.namenode.reconstruction.pending.timeout-sec
dfs.namenode.replication.pending.timeout-sec =>
dfs.namenode.reconstruction.pending.timeout-sec
{code}
Like mentioned, this jira to revisit the replication related configs, following
are few replication config items. Do we need to modify any of these to
reconstruction or redundancy? I remember somewhere we dicussed to consider EC
internal blocks are a special form of replicas, do we documented the same
anywhere?
{code}
dfs.namenode.replication.considerLoad
dfs.namenode.replication.considerLoad.factor
dfs.namenode.replication.min
dfs.namenode.replication.interval
dfs.namenode.replication.work.multiplier.per.iteration
dfs.namenode.replication.max-streams
dfs.namenode.replication.max-streams-hard-limit
{code}
> Erasure Coding: Deprecate replication-related config keys
> -
>
> Key: HDFS-10368
> URL: https://issues.apache.org/jira/browse/HDFS-10368
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: erasure-coding
>Reporter: Rakesh R
>Assignee: Rakesh R
> Labels: hdfs-ec-3.0-must-do
>
> This jira is to visit the replication based config keys and deprecate them(if
> necessary) in order to make it more meaningful.
> Please refer [discussion
> thread|https://issues.apache.org/jira/browse/HDFS-9869?focusedCommentId=15249363=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15249363]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
