work units in error state

Questions and Answers : Getting started : work units in error state

Author	Message
genfoch01 Send message Joined: 23 Jun 99 Posts: 4 Credit: 873,571 RAC: 0	Message 1116267 - Posted: 12 Jun 2011, 14:43:37 UTC
	So I am running boinc on a windows machine using the gpu. I have been getting wu in an error state with the stderr <core_client_version>6.12.26</core_client_version> <![CDATA[ <message> Maximum elapsed time exceeded </message> ]]> anyone know what this means? It can not be speaking of the wu deadline since thats over 2 weeks away. I now have 7 of these so I am curious as to what this is about and if there is any way to eliminate it. ____________ Angels can fly because they take themselves lightly - G.K.Chesterton
	ID: 1116267 ·

Gundolf Jahn Send message Joined: 19 Sep 00 Posts: 3075 Credit: 341,498 RAC: 85	Message 1116322 - Posted: 12 Jun 2011, 16:15:25 UTC - in response to Message 1116267.
	When have you rebooted that machine the last time? It looks like the CUDA tasks get stuck on your GPU. Are they showing any progress in the Tasks tab of BOINC manager when running? Gruß, Gundolf ____________ Computer sind nicht alles im Leben. (Kleiner Scherz) SETI@home classic workunits 3,758 SETI@home classic CPU time 66,520 hours
	ID: 1116322 ·

Ageless Send message Joined: 9 Jun 99 Posts: 11635 Credit: 1,999,039 RAC: 4,234	Message 1116409 - Posted: 12 Jun 2011, 21:20:22 UTC - in response to Message 1116267.
	<core_client_version>6.12.26</core_client_version> <![CDATA[ <message> Maximum elapsed time exceeded </message> ]]> anyone know what this means? What it means is this: Each task comes equipped with a 'runtime estimate' which is measured in FLOPs (FLoating point OPerations). It's the <rsc_fpops_est>64788201036261.297000</rsc_fpops_est> amount in the <workunit/> tags in client_state.xml file (my example given is from an Einstein task). BOINC will check whether or not the application goes over this estimated amount, and if it does, the application is stopped and that error is given. Another error similar to that is when <rsc_disk_bound>100000000.000000</rsc_disk_bound> gets exceeded, then the message is "Maximum disk space exceeded". Then the amount of disk space that the task should use up is too big. ____________ Jord - BOINC FAQ Service - BOINC User Wiki Real is just a matter of perception.
	ID: 1116409 ·

BilBg Send message Joined: 27 May 07 Posts: 2021 Credit: 3,027,755 RAC: 6,570	Message 1116464 - Posted: 13 Jun 2011, 0:13:48 UTC - in response to Message 1116409. Last modified: 13 Jun 2011, 0:43:03 UTC
	Simply put: If the task for whatever reason runs 10 times longer than the initial estimate (column "To completion") BOINC aborts the task automatically as it decides that the task is hang/stuck. (But why it happened to tasks with "Run time: 0.00" I don't know http://setiathome.berkeley.edu/results.php?hostid=6014929&offset=0&show_names=0&state=5&appid= Maybe because you just reached: Consecutive valid tasks 11 http://setiathome.berkeley.edu/host_app_versions.php?hostid=6014929 ) Info on old threads: http://setiathome.berkeley.edu/forum_thread.php?id=60413&nowrap=true http://setiathome.berkeley.edu/forum_thread.php?id=62226 ____________ - ALF - "Find out what you don't do well ..... then don't do it!" :)
	ID: 1116464 ·

Questions and Answers : Getting started : work units in error state