[
https://issues.apache.org/jira/browse/SOLR-10306?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17036148#comment-17036148
]
Jan Høydahl commented on SOLR-10306:
------------------------------------
Please see [GitHub Pull Request
#1256|https://github.com/apache/lucene-solr/pull/1256]
([https://github.com/apache/lucene-solr/blob/solr10306-swappiness/solr/solr-ref-guide/src/taking-solr-to-production.adoc#avoid-swapping)]
for a proposed new section in the Ref Guide. The text is inspired by this
article
[https://unix.stackexchange.com/questions/265713/how-to-configure-swappiness-in-linux-memory-management]
> Document vm.swappiness and mlockall in RefGuide
> -----------------------------------------------
>
> Key: SOLR-10306
> URL: https://issues.apache.org/jira/browse/SOLR-10306
> Project: Solr
> Issue Type: Task
> Components: documentation
> Reporter: Jan Høydahl
> Priority: Minor
> Time Spent: 10m
> Remaining Estimate: 0h
>
> I think we should document sane best practice OS level settings in the ref
> guide, e.g. in
> https://cwiki.apache.org/confluence/display/solr/Taking+Solr+to+Production
> Such as lower system swappiness or ability to use mlockall (like this ES page
> https://www.elastic.co/guide/en/elasticsearch/guide/current/heap-sizing.html)
> I also found this github repo https://github.com/LucidWorks/mlockall-agent -
> it is old, did anyone have good experience with the agent for locking Solr's
> memory?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
