Hi Phil,

In the first opstation the cpu for Thread-15 is 8367686,61ms so this is the thread that is looping and the position is in GdhWebSocketServer.java line 642.

In the second opstation it's Thread-6 in line 636.

There is a while loop that starts at line 633 but to figure out exactly between which lines it's looping I would need some more line numbers. So it would be great if you could run jattach a number of times to get the spread of line numbers for the looping thread.

There is for example a continue statement at line 645 so if there are only line numbers between 633 and 645 this would indicate that it's the condition for this continue that causes the loop.

/Claes

Good morning,

ok I do this during the day to take the logs several times.

My Java is still at 50% CPU since yesterday. I think another loop needs to fire to bring it up to 100%.

Thank you so much

/Phil

You don't have to wait between the logs, you can take them immediately after each other. 100% is probably just another thread looping in the same place so I don't think that will give any further information.

/Claes

Hi Claes,

you will find attached the 2 logs for the compressor operator station and the suction operator station.
There are several logs following as requested.
Good luck with the analysis










/Phil

Hi Phil,

Looking closely att the line numbers I'm a bit puzzled because they actually don't match the modified version but the old one. Maybe I made some mistake during the build. Anyway I did the same thing again with great care this time and put the result in http://proview.se/v1/files/tmp/pwrrt_6.0.0_patch02.tar.gz.

Unpack with 'tar -xzf' and copy pwr_rt.jar to $pwr_lib. If jar is installed you can check the content of pwr_rt.jar with 'jar -tvf $pwr_lib/pwr_rt.jar'. jpwr/rt/GdhWebSocketServer.class should have date May 01.

/Claes

Note that the jar-file should be copied to $pwr_lib (/usr/pwrrt/lib) not $pwrp_lib.

/Claes