[
https://issues.apache.org/jira/browse/IGNITE-27520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ivan Zlenko updated IGNITE-27520:
---------------------------------
Description:
It will be a good idea to have a metric which will showcase current off-heap
memory requirements imposed by all storage engine profiles on the node.
Potentially it will be useful to see breakdown either by each profile or by
engine.
Even if this information could be parsed from the configuration having it in
logs solves couples of issues:
* We can have a summary without need to calculate it each time
* Indication when changes were made to configuration, but we forgot to restart
a node to apply them
* It is better to have all info from one source, like metrics, rather than
compiling it manually from different sources
* Helps identifying incorrect configuration when default values is set for the
profile, even if some specific values is set in configuration
was:
It will be a good idea to have a metric which will showcase current off-heap
memory requirements imposed by all storage engine profiles on the node.
Potentially it will be useful to see breakdown either by each profile or by
engine.
Even if this information could be parsed from the configuration having it in
logs solves couples of issues:
* We can have a summary without need to calculate it each time
* Indication when changes were made to configuration, but we forgot to restart
a node to apply them
* It is better to have all info from one source, like metrics, rather than
compiling it manually from different sources
> Introduce off-heap memory requirements metric
> ---------------------------------------------
>
> Key: IGNITE-27520
> URL: https://issues.apache.org/jira/browse/IGNITE-27520
> Project: Ignite
> Issue Type: Improvement
> Reporter: Ivan Zlenko
> Priority: Major
> Labels: ignite-3
>
> It will be a good idea to have a metric which will showcase current off-heap
> memory requirements imposed by all storage engine profiles on the node.
> Potentially it will be useful to see breakdown either by each profile or by
> engine.
> Even if this information could be parsed from the configuration having it in
> logs solves couples of issues:
> * We can have a summary without need to calculate it each time
> * Indication when changes were made to configuration, but we forgot to
> restart a node to apply them
> * It is better to have all info from one source, like metrics, rather than
> compiling it manually from different sources
> * Helps identifying incorrect configuration when default values is set for
> the profile, even if some specific values is set in configuration
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
