You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "GuoHao (Jira)" <ji...@apache.org> on 2024/05/17 08:59:00 UTC
[jira] [Updated] (HDDS-10868) Reduce Container Balance memory usage
[ https://issues.apache.org/jira/browse/HDDS-10868?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
GuoHao updated HDDS-10868:
--------------------------
Description:
When we balance in a comparative cluster, there are often long-term GC issues with SCM, resulting in re election of SCM HA, such as:
The number of containers for a single data node is around 300w. When in balance, SCM measures GC logs
{code:java}
2024-05-16 20:58:56,078 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 1.771s with 1.831s GC time.
2024-05-16 21:00:23,701 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 24.111s with 24.095s GC time.
2024-05-16 21:00:24,809 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 0.108s without any GCs.
2024-05-16 21:01:56,334 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 21.511s with 21.500s GC time.
2024-05-16 21:06:30,957 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 142.648s with 142.984s GC time. {code}
> Reduce Container Balance memory usage
> -------------------------------------
>
> Key: HDDS-10868
> URL: https://issues.apache.org/jira/browse/HDDS-10868
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: GuoHao
> Priority: Major
>
> When we balance in a comparative cluster, there are often long-term GC issues with SCM, resulting in re election of SCM HA, such as:
> The number of containers for a single data node is around 300w. When in balance, SCM measures GC logs
> {code:java}
> 2024-05-16 20:58:56,078 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 1.771s with 1.831s GC time.
> 2024-05-16 21:00:23,701 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 24.111s with 24.095s GC time.
> 2024-05-16 21:00:24,809 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 0.108s without any GCs.
> 2024-05-16 21:01:56,334 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 21.511s with 21.500s GC time.
> 2024-05-16 21:06:30,957 [JvmPauseMonitor0] WARN org.apache.ratis.util.JvmPauseMonitor: JvmPauseMonitor-87cacbd5-f305-495e-8c1c-0645ddc29ca7: Detected pause in JVM or host machine approximately 142.648s with 142.984s GC time. {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org