hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Arpit Agarwal (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (HDFS-11282) Document the missing metrics of DataNode Volume IO operations
Date Tue, 03 Jan 2017 21:56:58 GMT

    [ https://issues.apache.org/jira/browse/HDFS-11282?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15796299#comment-15796299
] 

Arpit Agarwal edited comment on HDFS-11282 at 1/3/17 9:56 PM:
--------------------------------------------------------------

Thanks [~linyiqun]. Comments below.
# The rendered table looks wrong. The {{|}} need to be escaped. See
!metrics-rendered.png|width=900! 
# Can you add that enabling per-volume metrics may have a performance impact?
# "Average Time" should be replaced with "Mean Time" in descriptions to be accurate.
# The description for the count metrics and the rate metric is the same e.g. TotalMetadataOperations
and MetadataOperationRateNumOps. I am not sure that is correct.
# We can say that metadata operations include stat, list, mkdir, delete, move, open and posix_fadvise.

# The description _Average time of file io error operations in milliseconds_ could be improved.
It measures the mean time in milliseconds from the start of an operation to hitting a failure
(assuming failures were seen on that volume).


was (Author: arpitagarwal):
Thanks [~linyiqun]. Comments below.
# The rendered table looks wrong. The {{|}} need to be escaped. See !metrics-rendered.png!

# Can you add that enabling per-volume metrics may have a performance impact?
# "Average Time" should be replaced with "Mean Time" in descriptions to be accurate.
# The description for the count metrics and the rate metric is the same e.g. TotalMetadataOperations
and MetadataOperationRateNumOps. I am not sure that is correct.
# We can say that metadata operations include stat, list, mkdir, delete, move, open and posix_fadvise.

# The description _Average time of file io error operations in milliseconds_ could be improved.
It measures the mean time in milliseconds from the start of an operation to hitting a failure
(assuming failures were seen on that volume).

> Document the missing metrics of DataNode Volume IO operations
> -------------------------------------------------------------
>
>                 Key: HDFS-11282
>                 URL: https://issues.apache.org/jira/browse/HDFS-11282
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: documentation
>    Affects Versions: 3.0.0-alpha2
>            Reporter: Yiqun Lin
>            Assignee: Yiqun Lin
>            Priority: Minor
>         Attachments: HDFS-11282.001.patch, HDFS-11282.002.patch, metrics-rendered.png
>
>
> In HDFS-10959, it added many metrics of datanode volume io opearions. But it hasn't been
documented. This JIRA addressed on this.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org


Mime
View raw message