[
https://issues.apache.org/jira/browse/AMBARI-5973?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14218042#comment-14218042
]
Hank Jakiela commented on AMBARI-5973:
--------------------------------------
I'm curious to know what "pretty horrible performance" means and how much
memory there was in the machine. If it was a small memory, or a small VM on a
development machine, then it would be very dangerous to generalize to
large-memory machines. Production hadoop usually runs on machines much larger
than those used for development. Large machines (eg, 64 GB) typically have huge
pages enabled for good reason. It would be a mistake to turn that off.
> Usability: Add checks for THP to the Ambari pre-req check for Linux
> -------------------------------------------------------------------
>
> Key: AMBARI-5973
> URL: https://issues.apache.org/jira/browse/AMBARI-5973
> Project: Ambari
> Issue Type: Bug
> Reporter: Andrew Onischuk
> Assignee: Andrew Onischuk
> Fix For: 1.7.0
>
>
> We need to add checks for Transparent Huge Page. With it enabled, we have
> pretty horrible performance, and it's enabled by default on RHEL.
> Reference: <https://access.redhat.com/site/documentation/en-
> US/Red_Hat_Enterprise_Linux/6/html/Performance_Tuning_Guide/s-memory-
> transhuge.html>
> What we need to do pseudo code:
> if (`cat /sys/kernel/mm/redhat_transparent_hugepage/enabled` != 'never');
> then
> echo never > /sys/kernel/mm/redhat_transparent_hugepage/enabled
> end
> We need to add a check, and then add the 'echo never' to the python script.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
