Re: [Wrapper-user] Wrapper start problems

[Manjiri <man...@gm...>]

Hello,

Thank you for your help.

Yes, you are right, our box did shutdown and caused the apacheds to stop. On
startup, the jvm ping timed out and caused the wrapper to not start. Then
after 2 days, I realized that the ApacheDS service is not running and
attempted to start it.

I will edit the values for the wrapper.max_failed_invocations and the
wrapper.startup.timeout parameters according to your suggestions and let you
know.

Thanks again for your valuable help.

Sincerely,
Manjiri


Leif Mortenson-3 wrote:
> 
> Manjiri,
> Thank you for the logs.  There were a few things that I noticed.
> 
> 1) The first JVM infocation's log output simply stops at 2011/11/21
> 15:30:04 without the Wrapper shutting down.  This could happen if the
> OS was shutdown with a hard power off, or if Wrapper process crashed.
> 
> 2) The Wrapper is then started up again almost an hour later at
> 2011/11/21 16:20:42.  This time the Wrapper launches the JVM, but
> times out after 30 seconds waiting for it to start.   This could be a
> problem with the JVM freezing startup, or that your system was so
> heavily loaded at the time that the JVM was not able to complete its
> startup within that time.  It looks like you have told the Wrapper
> never to try again with the "wrapper.max_failed_invocations=1"
> property.  So it gives up after the first attempt and exits.   If the
> issue was caused by heavy load, then extending your
> wrapper.startup.timeout to a longer value might clear this up for you.
> We have made many improvements to the Wrapper over the years so there
> is a fairly good chance that this has been improved in the latest
> versions.  3.2.3 is over 5 years old.
> I am not able to give you much more information with that version
> unless you can reproduce this problem with the wrapper.debug=true
> property set.   I would prefer that you try to do so with a newer
> version however.
> 
> 3) There was then a 2 day down time.  The next launch attempt was at
> 2011/11/23 13:42:01.  What happened between these two events?  Was the
> system rebooted?   I have seen rare cases where Windows has resource
> problems which can cause it to start failing to launch a given
> process.   When that happens, it is never cleared up until the system
> has been rebooted.
> 
> Cheers,
> Leif
> 
> On Thu, Nov 24, 2011 at 5:52 AM, Manjiri <man...@gm...>
> wrote:
>>
>> Hello,
>>
>> I am sorry for not being able to provide the requested conf and the log
>> file. That issue seemed to disappear for a while. But, observed it again.
>>
>> ERROR  | wrapper  | 2011/11/21 16:21:23 | Startup failed: Timed out
>> waiting
>> for signal from JVM.
>> ERROR  | wrapper  | 2011/11/21 16:21:24 | JVM did not exit on request,
>> terminated
>> INFO   | wrapper  | 2011/11/21 16:21:24 | JVM exited on its own while
>> waiting to kill the application.
>> STATUS | wrapper  | 2011/11/21 16:21:24 | JVM exited in response to
>> signal
>> SIGKILL (9).
>>
>> Any help is highly appreciated.
>>
>> http://old.nabble.com/file/p32874521/apacheds.conf apacheds.conf
>> http://old.nabble.com/file/p32874521/wrapper.log.1 wrapper.log.1
>>
>> Sincerely,
>> Manjiri
> 
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure 
> contains a definitive record of customers, application performance, 
> security threats, fraudulent activity, and more. Splunk takes this 
> data and makes sense of it. IT sense. And common sense.
> http://p.sf.net/sfu/splunk-novd2d
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
> 
> 

-- 
View this message in context: http://old.nabble.com/Wrapper-start-problems-tp31295824p32884528.html
Sent from the Java Service Wrapper mailing list archive at Nabble.com.