[
https://issues.apache.org/jira/browse/DRILL-5821?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16183309#comment-16183309
]
Paul Rogers commented on DRILL-5821:
------------------------------------
The alternative is to use the metrics. It seem that `direct.capacity` may be
total direct memory. There is no metric for the number of CPUs.
> In the Drill web UI, display configured direct, heap memory
> -----------------------------------------------------------
>
> Key: DRILL-5821
> URL: https://issues.apache.org/jira/browse/DRILL-5821
> Project: Apache Drill
> Issue Type: Improvement
> Affects Versions: 1.11.0
> Reporter: Paul Rogers
> Assignee: Paul Rogers
> Fix For: 1.12.0
>
>
> When diagnosing memory-related issues, the first question we always ask is,
> "how much memory has Drill been given." This turns out to be rather tedious:
> we have to log onto the node running Drill, track down the config value, and
> check the configured amount. Since there are multiple possible config files,
> we have to figure out which was actually used.
> An alternative is to do `ps aux | grep Drill` to look at the command line,
> but this also requires access to the node itself.
> As an easier route, just as we display encryption information on the web UI
> main page, display the following:
> {noformat}
> Resources
> Direct Memory: ## GB
> Heap Memory: # GB
> Cores: ##
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
