[ https://issues.apache.org/jira/browse/ARTEMIS-1096?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15959572#comment-15959572 ]
ASF GitHub Bot commented on ARTEMIS-1096: ----------------------------------------- Github user jbertram commented on the issue: https://github.com/apache/activemq-artemis/pull/1180 It might be good to explicitly log what is being used for the global-max-size (e.g. an INFO level) when using the default so users understand it's using Xmx/2. > Load tests exhaust memory if global-max-size is not set > ------------------------------------------------------- > > Key: ARTEMIS-1096 > URL: https://issues.apache.org/jira/browse/ARTEMIS-1096 > Project: ActiveMQ Artemis > Issue Type: Improvement > Components: Broker > Environment: Fedora 25 x86-64, 16G memory > Reporter: Justin Ross > Priority: Minor > > With the following config (note commented out global-max-size), testq doesn't > page or block. Under load tests, the broker eventually dies with an OOME. I > propose that there be some fallback code-level default for global-max-size. > {noformat} > <?xml version='1.0'?> > <configuration xmlns="urn:activemq" > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" > xsi:schemaLocation="urn:activemq /schema/artemis-configuration.xsd"> > <core xmlns="urn:activemq:core" > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" > xsi:schemaLocation="urn:activemq:core "> > <!-- <global-max-size>100Mb</global-max-size> --> > <acceptors> > <acceptor > name="artemis">tcp://localhost:5672?protocols=AMQP,CORE,OPENWIRE</acceptor> > </acceptors> > <addresses> > <address name="testq"> > <anycast> > <queue name="testq"> > <durable>false</durable> > </queue> > </anycast> > </address> > </addresses> > <security-settings> > <security-setting match="#"> > <permission type="consume" roles="admin"/> > <permission type="browse" roles="admin"/> > <permission type="send" roles="admin"/> > </security-setting> > </security-settings> > </core> > </configuration> > {noformat} -- This message was sent by Atlassian JIRA (v6.3.15#6346)