[
https://issues.apache.org/jira/browse/HDFS-10808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15450594#comment-15450594
]
Lei (Eddy) Xu commented on HDFS-10808:
--------------------------------------
Thanks for the explanation, [~anu]
bq. The intend of having the setExitFlag was to reduce the complexity of
multiple exits from the loop.
Wouldn't {{break}} is the simplest way to exit the while loop?
bq. Fortunately for us, the way we do cancel is by cancelling the executor and
not by relying on this flag.
It makes me wonder why we have the flag in the first place. And for the
comments of the following code:
{code}
// check if someone told us exit, treat this as an interruption
// point
// for the thread, since both getNextBlock and moveBlocAcrossVolume
// can take some time.
if (!shouldRun()) {
break;
}
{code}
If {{shouldRun}} is not used for canceling, it is very confused to me because
all the {{exitFlag}} and {{shouldRun()}} are only consumed within
{{copyBlocks()}}, and use atomic boolean for {{shouldRun}} flag.
> DiskBalancer does not execute multi-steps plan-redux
> ----------------------------------------------------
>
> Key: HDFS-10808
> URL: https://issues.apache.org/jira/browse/HDFS-10808
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: balancer & mover
> Reporter: Anu Engineer
> Assignee: Anu Engineer
> Attachments: HDFS-10808.001.patch, HDFS-10808.002.patch
>
>
> This is to redo of the fix in HDFS-10598
--
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
