[ 
https://issues.apache.org/jira/browse/ZOOKEEPER-4701?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

zhanglu153 updated ZOOKEEPER-4701:
----------------------------------
    Description: 
从版本 3.4.14 升级到版本 3.6.4 后,发现其他 JVM 内存选项(如 - XX: MaxRAMPercentage)无法用于控制 
zookeeper 的内存使用量。经调查发现,原因是在 3.5 版的 zkEnv.sh 脚本中添加了堆大小的默认设置,默认值为 1000M。
{code:java}
ZK_SERVER_HEAP="${ZK_SERVER_HEAP:-1000}"
export SERVER_JVMFLAGS="-Xmx${ZK_SERVER_HEAP}m $SERVER_JVMFLAGS"{code}
当用户使用 - XX: MaxRAMPercentage 控制 zookeeper 的内存使用量时,发现内存配置会被默认的 - Xmx 
配置值覆盖。这不符合用户所需的内存使用配置。

在 ZOOKEEPER-4391 中,无法使用较新的 JVM 内存设置作为 Xmx 覆盖 - ASF JIRA 
(apache.org)],还建议添加一个标志以默认禁用内存配置行为。仅当用户需要使用默认堆大小配置时,才能选择性地释放此默认值。

  was:
After upgrading from version 3.4.14 to version 3.6.4, it was found that other 
JVM memory options such as - XX: MaxRAMPercentage cannot be used to control the 
memory usage of zookeeper. Upon investigation, it was found that the reason was 
due to the addition of a default setting for heap size in the zkEnv.sh script 
in version 3.5, with a default value of 1000M.
{code:java}
ZK_SERVER_HEAP="${ZK_SERVER_HEAP:-1000}"
export SERVER_JVMFLAGS="-Xmx${ZK_SERVER_HEAP}m $SERVER_JVMFLAGS"{code}
When the user uses - XX: MaxRAMPercentage to control the memory usage of 
zookeeper, it is found that the memory configuration will be overwritten by the 
default - Xmx configuration value. This does not meet the memory usage 
configuration required by the user.

In ZOOKEEPER-4391, Unable to use newer JVM memory settings as Xmx overrides - 
ASF JIRA (apache.org)], it is also proposed to add a flag to disable the memory 
configuration behavior by default. This default value can only be optionally 
released when users need to use the default heap size configuration.


> Unable to use other JVM memory options to control zookeeper memory usage.
> -------------------------------------------------------------------------
>
>                 Key: ZOOKEEPER-4701
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-4701
>             Project: ZooKeeper
>          Issue Type: Improvement
>          Components: scripts
>    Affects Versions: 3.5.9, 3.5.10, 3.5.0, 3.5.1, 3.5.2, 3.5.3, 3.5.4, 3.6.0, 
> 3.6.3, 3.7.0, 3.5.5, 3.5.6, 3.5.7, 3.6.1, 3.5.8, 3.6.2, 3.8.0, 3.7.1, 3.6.4, 
> 3.8.1
>            Reporter: zhanglu153
>            Priority: Minor
>              Labels: Zookeeper
>
> 从版本 3.4.14 升级到版本 3.6.4 后,发现其他 JVM 内存选项(如 - XX: MaxRAMPercentage)无法用于控制 
> zookeeper 的内存使用量。经调查发现,原因是在 3.5 版的 zkEnv.sh 脚本中添加了堆大小的默认设置,默认值为 1000M。
> {code:java}
> ZK_SERVER_HEAP="${ZK_SERVER_HEAP:-1000}"
> export SERVER_JVMFLAGS="-Xmx${ZK_SERVER_HEAP}m $SERVER_JVMFLAGS"{code}
> 当用户使用 - XX: MaxRAMPercentage 控制 zookeeper 的内存使用量时,发现内存配置会被默认的 - Xmx 
> 配置值覆盖。这不符合用户所需的内存使用配置。
> 在 ZOOKEEPER-4391 中,无法使用较新的 JVM 内存设置作为 Xmx 覆盖 - ASF JIRA 
> (apache.org)],还建议添加一个标志以默认禁用内存配置行为。仅当用户需要使用默认堆大小配置时,才能选择性地释放此默认值。



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to