It also looks like that on the client end their flops are being overestimated drastically.
Is there any way we can fix this server side, or do clients need to be updated? For example: wow, on mine, in the client_state.xml, it was set to: <flops>58418537435568693000000000000000.000000</flops> I have an i7 2600k @ 4.0GHz. http://dnahome.cs.rpi.edu/dna/forum_thread.php?id=85#554 --Travis On May 4, 2011, at 1:22 AM, Travis Desell wrote: > Looking in the database, it seems like he computer who had this problem (and > there are others) have really low values for et_avg in the database: > > mysql> select pfc_n, et_n, et_avg, et_var, et_q from host_app_version where > host_id = 1017 or host_id = 2270; > +-------+------+----------------------+-----------------------+----------------------+ > | pfc_n | et_n | et_avg | et_var | et_q > | > +-------+------+----------------------+-----------------------+----------------------+ > | 15 | 15 | 5.42546789377713e-10 | 3.81303780501951e-21 | > 5.71955670752926e-20 | > | 26 | 26 | 4.96156582191146e-17 | -2.79897154084492e-31 | > 1.46049779705622e-22 | > +-------+------+----------------------+-----------------------+----------------------+ > > Could that be causing the problem? > > Why would the estimated time be so far off? > > --Travis > > > On May 4, 2011, at 1:18 AM, Travis Desell wrote: > >> Even more problems. I've been using the new credit policy on DNA@Home as >> well, and users are getting errors along these lines: >> >> >> 5/3/2011 10:54:42 PM DNA@Home Aborting task test_update3_0_3866_0_0: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_4009_0_0: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_4006_0_1: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_4005_0_1: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_4004_0_1: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_3964_0_0: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_3953_0_1: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_3937_0_1: >> exceeded elapsed time limit 0.010911 >> 5/3/2011 11:03:27 PM DNA@Home Aborting task test_update3_0_3916_0_0: >> exceeded elapsed time limit 0.010911 >> >> For some reason, they're using some weird time limits.. any fixes for this? >> >> >> On May 1, 2011, at 3:12 PM, Travis Desell wrote: >> >>> Just recently we've been having an error where for some of our nbody >>> workunits the client errors out with the message: >>> >>> Maximum elapsed time exceeded >>> >>> After just a couple seconds. >>> >>> For example: >>> >>> http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=20284800 >>> >>> However, in the workunit the rsc_fpops_bound is set: >>> >>> mysql> select rsc_fpops_bound, rsc_fpops_est, rsc_memory_bound, >>> rsc_disk_bound from workunit where id = 13655742; >>> +----------------------+-----------------+------------------+----------------+ >>> | rsc_fpops_bound | rsc_fpops_est | rsc_memory_bound | >>> rsc_disk_bound | >>> +----------------------+-----------------+------------------+----------------+ >>> | 2.32172068444737e+16 | 232172068444737 | 500000000 | >>> 52428800 | >>> +----------------------+-----------------+------------------+----------------+ >>> 1 row in set (0.00 sec) >>> >>> >>> Any reason why this could be happening? >>> >>> >>> ---------------------------------------------------------------------------------------------------------- >>> Travis Desell <deselt @ cs.rpi.edu> >>> 1-518-867-1054 >>> Adjunct Professor & Postdoctoral Research Assistant >>> Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA >>> http://www.cs.rpi.edu/~deselt/ >>> MilkyWay@Home ( http://milkyway.cs.rpi.edu/ ) >>> DNA@Home ( http://dnahome.cs.rpi.edu/ ) >>> Worldwide Computing Laboratory ( http://wcl.cs.rpi.edu/ ) >>> ---------------------------------------------------------------------------------------------------------- >>> >>> _______________________________________________ >>> boinc_dev mailing list >>> [email protected] >>> http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev >>> To unsubscribe, visit the above URL and >>> (near bottom of page) enter your email address. >> >> ---------------------------------------------------------------------------------------------------------- >> Travis Desell <deselt @ cs.rpi.edu> >> 1-518-867-1054 >> Adjunct Professor & Postdoctoral Research Assistant >> Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA >> http://www.cs.rpi.edu/~deselt/ >> MilkyWay@Home ( http://milkyway.cs.rpi.edu/ ) >> DNA@Home ( http://dnahome.cs.rpi.edu/ ) >> Worldwide Computing Laboratory ( http://wcl.cs.rpi.edu/ ) >> ---------------------------------------------------------------------------------------------------------- >> > > ---------------------------------------------------------------------------------------------------------- > Travis Desell <deselt @ cs.rpi.edu> > 1-518-867-1054 > Adjunct Professor & Postdoctoral Research Assistant > Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA > http://www.cs.rpi.edu/~deselt/ > MilkyWay@Home ( http://milkyway.cs.rpi.edu/ ) > DNA@Home ( http://dnahome.cs.rpi.edu/ ) > Worldwide Computing Laboratory ( http://wcl.cs.rpi.edu/ ) > ---------------------------------------------------------------------------------------------------------- > ---------------------------------------------------------------------------------------------------------- Travis Desell <deselt @ cs.rpi.edu> 1-518-867-1054 Adjunct Professor & Postdoctoral Research Assistant Rensselaer Polytechnic Institute, 110 8th Street, Troy NY 12180, USA http://www.cs.rpi.edu/~deselt/ MilkyWay@Home ( http://milkyway.cs.rpi.edu/ ) DNA@Home ( http://dnahome.cs.rpi.edu/ ) Worldwide Computing Laboratory ( http://wcl.cs.rpi.edu/ ) ---------------------------------------------------------------------------------------------------------- _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
