[
https://issues.apache.org/jira/browse/HADOOP-17003?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17159293#comment-17159293
]
Ahmed Hussein commented on HADOOP-17003:
----------------------------------------
The implementation of the retention depends on the log4j aPI used in Hadoop.
> No Log compression and retention in Hadoop
> ------------------------------------------
>
> Key: HADOOP-17003
> URL: https://issues.apache.org/jira/browse/HADOOP-17003
> Project: Hadoop Common
> Issue Type: Bug
> Components: kms
> Reporter: Ahmed Hussein
> Assignee: Ahmed Hussein
> Priority: Major
> Attachments: HADOOP-17003.001.patch, HADOOP-17003.002.patch,
> HADOOP-17003.003.patch, HADOOP-17003.004.patch, HADOOP-17003.005.patch,
> HADOOP-17003.006.patch
>
>
> Hadoop logging lacks several important features. Logs generated end up eating
> disk space
> We need an implementation that satisfies the following three features: 1)
> time-based rolling, 2) retention and 3) compression.
> For example KMS logs have no retention or compression.
> {code:bash}
> -rw-r--r-- 1 hkms users 704M Mar 20 23:59 kms.log.2020-03-20
> -rw-r--r-- 1 hkms users 731M Mar 21 23:59 kms.log.2020-03-21
> -rw-r--r-- 1 hkms users 750M Mar 22 23:59 kms.log.2020-03-22
> -rw-r--r-- 1 hkms users 757M Mar 23 23:59 kms.log.2020-03-23
> -rw-r--r-- 1 hkms users 805M Mar 24 23:59 kms.log.2020-03-24
> -rw-r--r-- 1 hkms users 858M Mar 25 23:59 kms.log.2020-03-25
> -rw-r--r-- 1 hkms users 875M Mar 26 23:59 kms.log.2020-03-26
> -rw-r--r-- 1 hkms users 754M Mar 27 23:59 kms.log.2020-03-27
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
