|Re: SMILA stop because of missing ODEServerImpl threads [message #664051 is a reply to message #663796]
||Thu, 07 April 2011 10:08
Originally posted by: juergen.schumacher.attensity.com|
Am 06.04.2011, 16:20 Uhr, schrieb semweb <email@example.com>:
> my SMILA did run fine for a while but then it got slower and slower and
> finally stoped. I noticed that the ListenerWorker thread where still
> there but all ODEServerImpl where gone.
> After I set the number of ListenerWorker thread to 0 and to 1, two new
> ODEServerImpl threads apeared.
> Is it normal that there are two ODEServerImpl thread for each
> ListenerWorker and does anyone know why the ODEServerImpl threads die?
The ODEServerImpl threads are managed by ODE dynamically and created if
any are needed,
but also destroyed if they are idle for some time (at least it seems like
this in a
simple experiment I just did). It also looks like that it creates always
at least 2 threads
and more depending on the number of invocations.
So I assume that the problem is within the ListenerWorker-Thread: It's
still there but not
working correctly anymore. Do you have any error messages in the log that
could help with
analysis? Maybe a Java error like OutOfMemory? I will take a closer look
at this in the next
few days ... not sure yet, when I'll get some time.
Powered by FUDForum
. Page generated in 0.11323 seconds