[
https://issues.apache.org/jira/browse/HDFS-16682?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17607344#comment-17607344
]
Erik Krogen commented on HDFS-16682:
------------------------------------
I remember at the time we built this feature, [~shv] had reservations about
exposing these as configurable parameters. Konstantin, do you have any
commentary here?
One high-level comment is that if we do add configs, they should be namespaced
to make their purpose more clear as a part of the observer namenode feature
specifically.
> [SBN Read] make estimated transactions configurable
> ---------------------------------------------------
>
> Key: HDFS-16682
> URL: https://issues.apache.org/jira/browse/HDFS-16682
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Reporter: zhengchenyu
> Assignee: zhengchenyu
> Priority: Minor
> Labels: pull-request-available
> Time Spent: 20m
> Remaining Estimate: 0h
>
> In GlobalStateIdContext, ESTIMATED_TRANSACTIONS_PER_SECOND and
> ESTIMATED_SERVER_TIME_MULTIPLIER should be configured.
> These parameter depends on different cluster's load. In the other way, these
> config will help use to simulate observer namenode was far behind.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
