I still have the issue with this last version. Juste > Le 31 déc. 2015 à 14:30, Romain Reuillon <[email protected]> a écrit : > > I just fixed some programing errors in this branch. > > Romain > > Le 31/12/2015 13:01, Romain Reuillon a écrit : >> It would be a little bit more complicated since netlogo is turned into an >> osgi bundle by OpenMOLE. However I did a dirty trick to close those >> classloaders in this branch of OpenMOLE: >> <https://github.com/openmole/openmole/tree/closeNetLogoClassLoader>https://github.com/openmole/openmole/tree/closeNetLogoClassLoader >> <https://github.com/openmole/openmole/tree/closeNetLogoClassLoader> >> >> Could you try it? >> >> Le 31/12/2015 12:52, Juste Raimbault a écrit : >>> ok thank you very much for the quick bug fix, hope netlogo will quickly >>> merge it ; in the meantime in can also compile your fork of netlogo and >>> replace the netlogo jar in openmole plugins, will simply replacing the >>> netlogo jar with the new one do the job ? >>> >>> Juste >>> >>> -- >>> Juste Raimbault >>> Doctorant en géographie >>> UMR 8504 CNRS - Géographie-Cités / UMR-T 9403 IFSTTAR - LVMT >>> >>> >>> >>> >>>> Le 31 déc. 2015 à 12:46, Romain Reuillon <[email protected] >>>> <mailto:[email protected]>> a écrit : >>>> >>>> I made a pull request: >>>> <https://github.com/NetLogo/NetLogo/pull/908>https://github.com/NetLogo/NetLogo/pull/908 >>>> <https://github.com/NetLogo/NetLogo/pull/908> >>>> >>>> In the meantime I'll try to implement a quick and dirty solution in >>>> OpenMOLE to see if it solve the pb. >>>> >>>> Le 31/12/2015 11:58, Romain Reuillon a écrit : >>>>> Maybe, the reset method (in the file above) should be called from the >>>>> dispose methode of the headless workspace: >>>>> https://github.com/NetLogo/NetLogo/blob/5.x/src/main/org/nlogo/workspace/ExtensionManager.java >>>>> >>>>> <https://github.com/NetLogo/NetLogo/blob/5.x/src/main/org/nlogo/workspace/ExtensionManager.java> >>>>> >>>>> Romain >>>>> >>>>> Le 31/12/2015 11:46, Romain Reuillon a écrit : >>>>>> Hi Juste, >>>>>> >>>>>> it looks like a bug of netlogo headless that doesn't close the >>>>>> extensions when it is disposed :(... Could you submit a ticket to the >>>>>> netlogo team? >>>>>> >>>>>> Romain >>>>>> >>>>>> Le 31/12/2015 11:37, Juste Raimbault a écrit : >>>>>>> Hi all, >>>>>>> >>>>>>> I'm currently experiencing an issue with a reasonable size exploration >>>>>>> of Netlogo5Task ; >>>>>>> >>>>>>> - Running on a local environement, the exploration fails after around >>>>>>> 1000 completed tasks, with the exception "FileNotFoundException (too >>>>>>> many files open)" >>>>>>> >>>>>>> - I relaunched the experiment and monitored progressively opened file >>>>>>> by the java task for openmole, using >>>>>>> lsof -p [openmole pid] | grep >>>>>>> ~/.openmole/[host]/.tmp/[experiment-dir] | wc -l >>>>>>> where [openmole pid], [host] and [experiment-dir] are replaced with >>>>>>> actual values. It appears indeed that this count grows linearly with >>>>>>> completed tasks, and that *.jar files corresponding to netlogo >>>>>>> extensions (2 used in the model) are indeed kept open, even after task >>>>>>> completion and deletion : adding a 'grep "(deleted)"' in the pipe also >>>>>>> yields linear number of opened files ; what should cause the exception >>>>>>> because of system or jvm file open limit. >>>>>>> >>>>>>> Is it something I miss with the Netlogo5Task like an extra parameter >>>>>>> that tells it to close open files when finished ? or is it an issue >>>>>>> with netlogo or the embedding of netlogo that does not close open files >>>>>>> when finished ? >>>>>>> >>>>>>> Thank you in advance ; and happy new year's Eve tonight ! >>>>>>> >>>>>>> Best, >>>>>>> >>>>>>> Juste >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> Juste Raimbault >>>>>>> Doctorant en géographie >>>>>>> UMR 8504 CNRS - Géographie-Cités / UMR-T 9403 IFSTTAR - LVMT >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> OpenMOLE-users mailing list >>>>>>> [email protected] <mailto:[email protected]> >>>>>>> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >>>>>>> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> OpenMOLE-users mailing list >>>>>> [email protected] <mailto:[email protected]> >>>>>> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >>>>>> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> OpenMOLE-users mailing list >>>>> [email protected] <mailto:[email protected]> >>>>> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >>>>> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> >>>> >>>> _______________________________________________ >>>> OpenMOLE-users mailing list >>>> [email protected] <mailto:[email protected]> >>>> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >>>> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> >>> >>> >>> >>> _______________________________________________ >>> OpenMOLE-users mailing list >>> [email protected] <mailto:[email protected]> >>> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >>> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> >> >> >> >> _______________________________________________ >> OpenMOLE-users mailing list >> [email protected] <mailto:[email protected]> >> http://fedex.iscpif.fr/mailman/listinfo/openmole-users >> <http://fedex.iscpif.fr/mailman/listinfo/openmole-users> > > _______________________________________________ > OpenMOLE-users mailing list > [email protected] > http://fedex.iscpif.fr/mailman/listinfo/openmole-users
_______________________________________________ OpenMOLE-users mailing list [email protected] http://fedex.iscpif.fr/mailman/listinfo/openmole-users
