[
https://issues.apache.org/jira/browse/HDFS-14576?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16865718#comment-16865718
]
He Xiaoqiao edited comment on HDFS-14576 at 6/17/19 3:56 PM:
-------------------------------------------------------------
Thanks [~jojochuang],[~dineshchitlangia] for your quick response.
{quote}I suppose this can still be an issue at 10k node scale.{quote}
Correct, I think it is more serious for larger cluster according to my
experience. I will submit draft patch later and welcome discussion.
About load fsimage, some solutions has been proposed, for instance HDFS-7784
loading in parallel. I think it is time to reevaluate this solution. FYI.
was (Author: hexiaoqiao):
Thanks [~jojochuang],[~dineshchitlangia] for your quick response.
{quote}I suppose this can still be an issue at 10k node scale.{quote}
Correct, I think it is more serious for larger cluster according to my
experience. I will submit draft patch later and welcome discussion.
About load fsimage, there are some solution proposed, for instance HDFS-7784
loading in parallel. I think it is time to reevaluate this solution. FYI.
> Avoid block report retry and slow down namenode startup
> -------------------------------------------------------
>
> Key: HDFS-14576
> URL: https://issues.apache.org/jira/browse/HDFS-14576
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: namenode
> Reporter: He Xiaoqiao
> Assignee: He Xiaoqiao
> Priority: Major
>
> During namenode startup, the load will be very high since it has to process
> every datanodes blockreport one by one. If there are hundreds datanodes block
> reports pending process, the issue will be more serious even
> #processFirstBlockReport is processed a lot more efficiently than ordinary
> block reports. Then some of datanode will retry blockreport and lengthens
> restart times. I think we should filter the block report request (via
> datanode blockreport retries) which has be processed and return directly then
> shorten down restart time. I want to state this proposal may be obvious only
> for large cluster.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
