[
https://issues.apache.org/jira/browse/IGNITE-16827?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vyacheslav Koptilin updated IGNITE-16827:
-----------------------------------------
Reviewer: Vladislav Pyatkov
> Added new Ignite system property that allows to reduce a number diagnostic
> messages.
> ------------------------------------------------------------------------------------
>
> Key: IGNITE-16827
> URL: https://issues.apache.org/jira/browse/IGNITE-16827
> Project: Ignite
> Issue Type: Improvement
> Reporter: Vyacheslav Koptilin
> Assignee: Vyacheslav Koptilin
> Priority: Minor
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Additional diagnostic messages that have been added under IGNITE-5155 and
> IGNITE-5774 (at least) can produce a huge string objects.
> For example, CacheObjectsReleaseFuture can include thousands futures, and
> therefore this lead to extreme stress on GC and takes a significant time.
> Moreover, it may lead to thread pool starvation on sys pool:
> {noformat}
> 00:15:48.807 ^-- System thread pool [active=8, idle=0, qSize=52]
> ...
> 00:22:49.429 ^-- System thread pool [active=8, idle=0, qSize=346]
> ...
> 07:09:30.306 ^-- System thread pool [active=8, idle=0, qSize=18515]
> ...
> 09:52:42.118 ^-- System thread pool [active=8, idle=0, qSize=26365]
> {noformat}
> {noformat}
> 00:15:18.821 WARN Possible thread pool starvation detected (no task
> completed in last 30000ms, is system thread pool size large enough?)
> ...
> 06:31:14.114 WARN Possible thread pool starvation detected (no task
> completed in last 30000ms, is system thread pool size large enough?)
> ...
> 09:51:49.455 WARN Possible thread pool starvation detected (no task
> completed in last 30000ms, is system thread pool size large enough?)
> {noformat}
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
