[
https://issues.apache.org/jira/browse/HDFS-10459?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eric Badger updated HDFS-10459:
-------------------------------
Affects Version/s: (was: 0.23.0)
(was: 0.22.0)
2.9.0
Description: The computation works on threshold = 1, but not on
threshold < 1. I propose making blockThreshold equal to the ceiling of
total*threshold. Since we need to be >= blockThreshold to get out of safe mode,
>14.9 is the same as =15, since blocks work in integer values. (was: The fix
added in HDFS-2002 only works in cases where threshold < 1, but not when
threshold = 1. There is a '+1' added to the computation because of this
assumption. I propose that instead of adding a '+1', we just set blockThreshold
to be the ceiling of blocksTotal*threshold. Since we need to be >=
blockThreshold to get out of safe mode, this will work. >14.9 is the same as
=15, since blocks work in integer values. )
Summary: getTurnOffTip computes needed block incorrectly for
threshold < 1 (was: getTurnOffTip computes needed block incorrectly for
threshold = 1)
> getTurnOffTip computes needed block incorrectly for threshold < 1
> -----------------------------------------------------------------
>
> Key: HDFS-10459
> URL: https://issues.apache.org/jira/browse/HDFS-10459
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 2.9.0
> Reporter: Eric Badger
> Assignee: Eric Badger
> Attachments: HDFS-10459.001.patch
>
>
> The computation works on threshold = 1, but not on threshold < 1. I propose
> making blockThreshold equal to the ceiling of total*threshold. Since we need
> to be >= blockThreshold to get out of safe mode, >14.9 is the same as =15,
> since blocks work in integer values.
--
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
