[
https://issues.apache.org/jira/browse/QPID-6944?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15059970#comment-15059970
]
Keith Wall edited comment on QPID-6944 at 12/16/15 1:25 PM:
------------------------------------------------------------
Hi Mark,
I have tried running {{Qpid6944}} (on separate host) with a 6.0.0 Broker that
uses AMQPS (TLS) and BDB based virtual host on a platform that uses RHEL 6.6
Santiago. I am not seeing the same CPU behaviour as you (observing CPU using
jvisualvm attached to the Qpid Broker's JVM)
I wasn't too clear on how you were using your reproduction, so I've tried two
things:
# I put a loop in {{Qpid6944#run()}} (so it creates a connection, sends
message, closes connection) and left the program running for some time. CPU
utilisation is flat at 6.1%.
# I put a loop in {{Qpid6944#send()}} (so it creates a connection, sent
messages (1000) , sleep(50ms), repeat last two steps 50 times, close
connection) CPU utilising was even lower (hardly perceivable on the jvisualvm
chart)
I did not run {{AmqpMessageReader}}.
I have tested this with Oracle JDK 1.8 0_60 and Oracle JDK 1.7.0_67.
One difference I cannot eliminate is the Oracle BDB JE. I cannot test with JE
versions higher than 5.0.104 owing to the licence change.
Can you confirm that I am conducting the test as you do? Also, can you retest
with Oracle BDB JE 5.0.104. I'd be surprised if this were the issue, but it'll
eliminate from the enquiry all the same.
I need to review the qpid.log you have sent more carefully. I hope to have
some more time later today. I hope to have some more ideas.
was (Author: k-wall):
Hi Mark,
I have tried running {{Qpid6944}} (on separate host) with a 6.0.0 Broker that
uses AMQPS (TLS) and BDB based virtual host on a platform that uses RHEL 6.6
Santiago. I am not seeing the same CPU behaviour as you (observing CPU using
jvisualvm attached to the Qpid Broker's JVM)
I wasn't too clear on how you were using your reproduction, so I've tried two
things:
# I put a loop in {{Qpid6944#run()}} (so it creates a connection, sends
message, closes connection) and left the program running for some time. CPU
utilisation is flat at 6.1%.
# I put a loop in Qpid6944#send() (so it creates a connection, sent messages
(1000) , sleep(50ms), repeat last two steps 50 times, close connection) CPU
utilising was even lower (hardly perceivable on the jvisualvm chart)
I did not run {{AmqpMessageReader}}.
I have tested this with Oracle JDK 1.8 0_60 and Oracle JDK 1.7.0_67.
One difference I cannot eliminate is the Oracle BDB JE. I cannot test with JE
versions higher than 5.0.104 owing to the licence change.
Can you confirm that I am conducting the test as you do? Also, can you retest
with Oracle BDB JE 5.0.104. I'd be surprised if this were the issue, but it'll
eliminate from the enquiry all the same.
I need to review the qpid.log you have sent more carefully. I hope to have
some more time later today. I hope to have some more ideas.
> High CPU use sending message
> ----------------------------
>
> Key: QPID-6944
> URL: https://issues.apache.org/jira/browse/QPID-6944
> Project: Qpid
> Issue Type: Bug
> Components: Java Broker
> Affects Versions: qpid-java-6.0
> Environment: Red Hat Enterprise Linux Server release 6.7 (Santiago)
> OpenJDK Runtime Environment (build 1.8.0_65-b17)
> OpenJDK 64-Bit Server VM (build 25.65-b01, mixed mode)
> Reporter: Mark Soderquist
> Attachments: enabledebug.png, qpid.0.32.png, qpid.6.0.png,
> qpid.cpu.png, qpid.cpu.png, qpid.log.zip, qpid6944.zip,
> threaddump-1449856663270.tdump
>
>
> The org.apache.qpid.server.transport.SelectorThread is causing live lock.
> While the application still functions it causes the CPU to run at nearly
> 100%. See attached thread dump.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
