activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gary Tully" <gary.tu...@gmail.com>
Subject Re: repeated closing of socket in activeMQ, connection reset
Date Fri, 19 Dec 2008 12:18:56 GMT
ok, that looks like the wrapper is restarting activemq and the stomp
client is seeing a connection drop which would be expected.

The question is what is causing the wrapper to determine that the jvm is hung.

Is it possible to enable logging for activemq somewhere in the wrapper
to see if we can understand what is happening when the 'jvm is hung'
determination occurs and the jvm is terminated.

see: http://activemq.apache.org/how-do-i-change-the-logging.html

2008/12/19 saurabh85 <saurabh_sharma19@infosys.com>:
>
> Hi,
>    I am giving the logs that I got.
>     First of all, there is a wrapper for active MQ, and the log file for
> activeMq shows like this:
>         ERROR  | wrapper  | 2008/12/18 01:14:34 | JVM appears hung: Timed
> out waiting for signal from JVM.
>
> ERROR  | wrapper  | 2008/12/18 01:14:34 | JVM did not exit on request,
> terminated
>
> DEBUG  | wrapperp | 2008/12/18 01:14:36 | server listening on port 32001.
>
> DEBUG  | wrapper  | 2008/12/18 01:14:36 | Waiting 5 seconds before launching
> another JVM.
>
> DEBUG  | wrapper  | 2008/12/18 01:14:40 | Signal trapped.  Details:
>
> DEBUG  | wrapper  | 2008/12/18 01:14:40 |   signal number=17 (SIGCHLD),
> source="unknown"
>
> DEBUG  | wrapper  | 2008/12/18 01:14:40 | Received SIGCHLD, checking JVM
> process status.
>
> STATUS | wrapper  | 2008/12/18 01:14:40 | JVM exited in response to signal
> SIGKILL (9).
>
> DEBUG  | wrapper  | 2008/12/18 01:14:40 | JVM process exited with a code of
> 1, however the wrapper exit code was already 1.
>
> ERROR  | wrapper  | 2008/12/18 01:14:40 | Unable to start a JVM
>
> STATUS | wrapper  | 2008/12/18 01:14:41 | <-- Wrapper Stopped
>
> In fstomp log file, we have the following messages:
> 1. Error processing frame
> 2. These functions are called one by one:
>    a. handle_the_frame()
>    b. process_the_frame()
>    c. recieve_raw()
> When it comes to receive_raw(), it raises an exception:
> error: (111, 'Connection refused')
> unable to send due to "<class 'socket.error'>: (32, 'Broken pipe')"
>
> Theres another log file which manages everything:
> unable to send "SEND
>
> action:spc_query
>
> mtype:request
>
> destination:/topic/REQUEST
>
> from:some-system-name
>
> to:*:*
>
> I am getting this error time and again which leads to shutdown of activeMQ
> each now and then. Please help this out
>
>
> Gary Tully wrote:
>>
>> What information do you get from the logging?
>> Any chance that is in the default inactivity timeout of 30 seconds
>> which is kicking in?
>>
>> 2008/12/18 saurabh85 <saurabh_sharma19@infosys.com>:
>>>
>>> There is some problem due to which the connection is getting closed
>>> everytime.
>>> Almost everyday, the connection gets reset. I am using active MQ versuion
>>> 5.x
>>>
>>> Can any one tell me the possible reason ?? Is this some problem of
>>> activeMQ
>>> ?
>>> --
>>> View this message in context:
>>> http://www.nabble.com/repeated-closing-of-socket-in-activeMQ%2C-connection-reset-tp21072892p21072892.html
>>> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
>>>
>>>
>>
>>
>>
>> --
>> http://blog.garytully.com
>>
>> Open Source SOA
>> http://FUSESource.com
>>
>>
>
> --
> View this message in context: http://www.nabble.com/repeated-closing-of-socket-in-activeMQ%2C-connection-reset-tp21072892p21085849.html
> Sent from the ActiveMQ - Dev mailing list archive at Nabble.com.
>
>



-- 
http://blog.garytully.com

Open Source SOA
http://FUSESource.com

Mime
View raw message