Yes, that was what I was suggesting.

If your performance tests need enough consistency that the Jenkins slave
agent growth is a real problem, then it seems like you should consider not
executing the slave on the target node at all.  It seems like you may also
want to reboot the computer before starting the performance tests, and take
other measures to assure the computer is not affected by other programs
running on the machine unnecessarily (like virus scanners and corporate
administrative jobs and ...).

Mark Waite

On Fri, Apr 24, 2015 at 6:40 AM <pra...@gmail.com> wrote:

> Hi Mark,
>
> Thanks for your suggestion. We thought it was convenient to run Jenkins
> slave on the machine running the tests because Jenkins would show us the
> progress of the job while it is running and other nice to have information.
>
> Are you suggesting that we launch the target program using psexec from
> within Jenkins master? This definitely looks like a promising solution. We
> will give it a try if other options fail.
>
> For now we are using task scheduler to run Jenkins slave.jar. it seems to
> work.
>
> Thanks
>
>
>
>
> On Thursday, April 23, 2015 at 12:47:49 AM UTC+5:30, Mark Waite wrote:
>
>> If you're running performance tests, isn't the presence of any other
>> processes a risk to the repeatability of your performance test?
>>
>> Wouldn't it be better to use a Jenkins slave (or master) to launch the
>> target program from a remote machine, without a Jenkins slave agent on the
>> machine running  the tests?
>>
>> Mark Waite
>>
>> On Wed, Apr 22, 2015 at 8:17 AM Dunnigan, Terrence J <
>> terrence....@boeing.com> wrote:
>>
> Two thoughts.
>>>
>>>
>>>
>>> 1)      When running Jenkins as a service, can you run as a user,
>>> instead of local system? And can you give that user permissions to your
>>> share?
>>>
>>> 2)      From a command window, “shutdown /r /t 0” will restart the
>>> machine. If you’re not running as a service you’ll need to log in before
>>> the slave reconnects.
>>>
>>>
>>>
>>> Terry
>>>
>>>
>>>
>> *From:* jenkins...@googlegroups.com [mailto:jenkins...@googlegroups.com] *On
>>> Behalf Of *pra...@gmail.com
>>>
>>
>>> *Sent:* Wednesday, April 22, 2015 5:09 AM
>>>
>> *To:* jenkins...@googlegroups.com
>>> *Subject:* Jenkins slave restart
>>>
>>
>>>
>>> Hi,
>>>
>>>
>>>
>>> We have recently adopted Jenkins to manage performance testing lab which
>>> comprises of many performance testing servers.
>>>
>>>
>>>
>>> *A little background of the problem we have:*
>>>
>>>
>>>
>>> Lately we have been observing some discrepancy in the readings due to
>>> memory consumption of Jenkins slave - memory usage goes up and down.
>>>
>>>
>>>
>>> For reliability we need same readings when a single test is executed
>>> multiple times. But that doesn't seem to happen now after adopting Jenkins.
>>>
>>>
>>>
>>> We have observed that Jenkins slave memory usage increases gradually
>>> over a period of time and suspecting that this could be affecting the
>>> stability of the performance tests.
>>>
>>>
>>>
>>> *Question:*
>>>
>>>
>>>
>>> So now we want to figure out a way to restart Jenkins slave after every
>>> performance test we perform on the server. Using Jenkins slave as a service
>>> is preventing us to access some network shares so we were forced to start
>>> Jenkins slave using the java -jar option.
>>>
>>> Is there a way we can restart Jenkins slaves after every test run on the
>>> slave?
>>>
>>>
>>>
>>> I have also attached herewith the list of plugins I have installed.
>>> Please advise if any of these plugins could be causing the increasing
>>> memory usage of Jenkins slave.
>>>
>>> I can share more details about my setup if it helps.
>>>
>>>
>>>
>>> Thank you for your time and advice.
>>>
>>>
>>>
>>> Regards,
>>>
>>> Praneeth
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Jenkins Users" group.
>>>
>> To unsubscribe from this group and stop receiving emails from it, send an
>>> email to jenkinsci-use...@googlegroups.com.
>>>
>>
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/jenkinsci-users/799ea334-37b4-4475-b317-37e8f4d6952c%40googlegroups.com
>>> <https://groups.google.com/d/msgid/jenkinsci-users/799ea334-37b4-4475-b317-37e8f4d6952c%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>  --
>>> You received this message because you are subscribed to the Google
>>> Groups "Jenkins Users" group.
>>>
>> To unsubscribe from this group and stop receiving emails from it, send an
>>> email to jenkinsci-use...@googlegroups.com.
>>
>>
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/jenkinsci-users/9AD8B655B5126D4FAE1397203CDDBF2D16A820%40XCH-PHX-205.sw.nos.boeing.com
>>> <https://groups.google.com/d/msgid/jenkinsci-users/9AD8B655B5126D4FAE1397203CDDBF2D16A820%40XCH-PHX-205.sw.nos.boeing.com?utm_medium=email&utm_source=footer>
>>> .
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>  --
> You received this message because you are subscribed to the Google Groups
> "Jenkins Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to jenkinsci-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/jenkinsci-users/78643d68-4945-40d1-8211-e02062f6112b%40googlegroups.com
> <https://groups.google.com/d/msgid/jenkinsci-users/78643d68-4945-40d1-8211-e02062f6112b%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/CAO49JtG%2BV0SZszc1T-WgZB0HjBNF%2BACARGGzj0fvVkJe6U3KgA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to