Re: [Wrapper-user] Service stop while in delay

[Leif M. <le...@ta...>]

Nic,
Don't bother with the bug. I forgot that the wrapper.restart.delay was added
for 2.2.9 and has not yet been released. I went ahead and got this fixed
today.
It required adding an additional state to the main state engine. This
lets the wrapper
continue to process all control events while it is waiting for the delay
to expire.

I also added a new advanced property which makes it possible to enable
output
of the current wrapper state engine. It is useful for understanding how
the internals
of the Wrapper work. Just add a wrapper.state_output=true line to your
wrapper.conf
file. This will output about 4 lines per second of information like the
following:

wrapper | WrapperState=STARTED, JVMState=STARTED JVMStateTimeout=28
wrapper | WrapperState=STARTED, JVMState=STARTED JVMStateTimeout=28
wrapper | WrapperState=STARTED, JVMState=STARTED JVMStateTimeout=28
wrapper | WrapperState=STARTED, JVMState=STARTED JVMStateTimeout=28
wrapper | WrapperState=STARTED, JVMState=STARTED JVMStateTimeout=27
wrapper | JVM exited unexpectedly.
wrapper | WrapperState=STARTED, JVMState=DOWN JVMStateTimeout=0
wrapper | WrapperState=STARTED, JVMState=LAUNCH(DELAY) JVMStateTimeout=5
wrapper | WrapperState=STARTED, JVMState=LAUNCH(DELAY) JVMStateTimeout=4
wrapper | WrapperState=STARTED, JVMState=LAUNCH(DELAY) JVMStateTimeout=4
wrapper | WrapperState=STARTED, JVMState=LAUNCH(DELAY) JVMStateTimeout=4

I did it this way rather than enabling the output through the log
levels, because it is pretty
low level and is difficult to observe when mixed with lots of other
debug level output.

Could you give this new version a try and confirm that it solves the
problems you were
seeing? (You can checkout from CVS?)

Cheers,
Leif

Leif Mortenson wrote:

>Nic,
>You are right. Hadn't thought of that. In the original design, the delay
>was fixed at
>5 seconds, so it was never a problem. The Wrapper is a single threaded
>app. When it
>is pausing at the restart, that thread simply waits for the specified
>time before
>continuing, so it makes sense that it would do that. I'll have to modify
>it so that it is
>another state in the main loop. That way it will still be able to
>respond to system
>events while it is paused.
>Can you submit a bug on this?
>Cheers,
>Leif
>
>asdfas sdfdsf wrote:
>
>  
>
>>Hi there, Leif
>>That wrapper.restart.delay parameter works great periodically running
>>my app. Back from holiday, but my guys here uncovered another small
>>prob in the meanwhile.
>>When the service is in delay it doesn't stop using the command-line
>>"net stop" (but works with the service manager). It will stop after
>>the delay, but the 'net stop' command returns with an error.
>>Cheers
>>Nic
>>
>>
>>
>>_________________________________________________________________
>>Join the world痴 largest e-mail service with MSN Hotmail.
>>http://www.hotmail.com
>>
>>
>>
>>-------------------------------------------------------
>>This sf.net email is sponsored by:ThinkGeek
>>Welcome to geek heaven.
>>http://thinkgeek.com/sf
>>_______________________________________________
>>Wrapper-user mailing list
>>Wra...@li...
>>https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>>    
>>
>
>
>
>
>-------------------------------------------------------
>This sf.net email is sponsored by:ThinkGeek
>Welcome to geek heaven.
>http://thinkgeek.com/sf
>_______________________________________________
>Wrapper-user mailing list
>Wra...@li...
>https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>  
>