Welcome to MilkyWay@home

more maximum time limit elapsed bug stuff

Message boards : News : more maximum time limit elapsed bug stuff
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 48853 - Posted: 18 May 2011, 21:07:57 UTC

Tried yet another fix on the server end of things. Let me know if you're still seeing maximum time limit elapsed errors.
ID: 48853 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jesse Viviano

Send message
Joined: 4 Feb 11
Posts: 86
Credit: 60,913,150
RAC: 0
Message 48862 - Posted: 19 May 2011, 13:09:35 UTC
Last modified: 19 May 2011, 13:22:39 UTC

I am still getting maximum time exceeded errors on the n-body work units, though one or a few are now completing successfully. My computer is now going into high priority mode as well. One failed at 82%. Another failed at 85%. It looks like I am getting a mix of successes and failures. These failures were under BOINC x64 version 6.10.60. Since BOINC released a new version, 6.12.26, I have upgraded since then.
ID: 48862 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 48866 - Posted: 19 May 2011, 20:44:22 UTC - in response to Message 48862.  

I am still getting maximum time exceeded errors on the n-body work units, though one or a few are now completing successfully. My computer is now going into high priority mode as well. One failed at 82%. Another failed at 85%. It looks like I am getting a mix of successes and failures. These failures were under BOINC x64 version 6.10.60. Since BOINC released a new version, 6.12.26, I have upgraded since then.


Is it failing for both the n-body simulation application and the regular one? Or just one of the two?
ID: 48866 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile The Gas Giant
Avatar

Send message
Joined: 24 Dec 07
Posts: 1947
Credit: 240,884,648
RAC: 0
Message 48867 - Posted: 19 May 2011, 20:55:20 UTC - in response to Message 48866.  

I am still getting maximum time exceeded errors on the n-body work units, though one or a few are now completing successfully. My computer is now going into high priority mode as well. One failed at 82%. Another failed at 85%. It looks like I am getting a mix of successes and failures. These failures were under BOINC x64 version 6.10.60. Since BOINC released a new version, 6.12.26, I have upgraded since then.


Is it failing for both the n-body simulation application and the regular one? Or just one of the two?

With just 4 clicks you to can see Jesse's returned results...
ID: 48867 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 48870 - Posted: 19 May 2011, 22:15:47 UTC - in response to Message 48867.  

I am still getting maximum time exceeded errors on the n-body work units, though one or a few are now completing successfully. My computer is now going into high priority mode as well. One failed at 82%. Another failed at 85%. It looks like I am getting a mix of successes and failures. These failures were under BOINC x64 version 6.10.60. Since BOINC released a new version, 6.12.26, I have upgraded since then.


Is it failing for both the n-body simulation application and the regular one? Or just one of the two?

With just 4 clicks you to can see Jesse's returned results...


When I clicked they were all correct...
ID: 48870 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
TJ

Send message
Joined: 12 Aug 09
Posts: 262
Credit: 92,631,041
RAC: 0
Message 48871 - Posted: 19 May 2011, 22:33:33 UTC - in response to Message 48870.  

I did a few yesterday without problem I think.
However they are very soon out of the database if wingman has them ready as well.
I have only stil two n-body waiting for validation, but mine ran without error.
But that was all before you (Travis) did an update.
Greetings from,
TJ
ID: 48871 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jesse Viviano

Send message
Joined: 4 Feb 11
Posts: 86
Credit: 60,913,150
RAC: 0
Message 48872 - Posted: 19 May 2011, 23:03:53 UTC - in response to Message 48866.  

I recently upgraded my BOINC client, and I am still getting some n-body work units that fail due to maximum time limits exceeded. Some n-body work units also pass. Few if any of the regular work units are giving me problems, and the last problem regular work unit I had was with the ATI client failing to record the results to stderr, and that was a week or two ago and has nothing to do with this problem.

I wonder if you might get some help if you visit http://boinc.berkeley.edu/dev/ and use the message boards there or the BOINC projects email lists at http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_projects for help. There might be a setting that you might be overlooking that could be obvious to another person. The same problem happens to the best of us (e.g. someone failing to declare a variable as volatile in C when using it to spin-wait on an I/O transaction and therefore the compiler optimizes the variable by moving it to a register).
ID: 48872 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 48876 - Posted: 20 May 2011, 2:58:52 UTC - in response to Message 48872.  


I wonder if you might get some help if you visit http://boinc.berkeley.edu/dev/ and use the message boards there or the BOINC projects email lists at http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_projects for help. There might be a setting that you might be overlooking that could be obvious to another person. The same problem happens to the best of us (e.g. someone failing to declare a variable as volatile in C when using it to spin-wait on an I/O transaction and therefore the compiler optimizes the variable by moving it to a register).


I've actually been working with Dave Anderson directly on this one. Something weird is going on with our scheduler.
ID: 48876 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile The Gas Giant
Avatar

Send message
Joined: 24 Dec 07
Posts: 1947
Credit: 240,884,648
RAC: 0
Message 48877 - Posted: 20 May 2011, 3:32:58 UTC - in response to Message 48870.  

I am still getting maximum time exceeded errors on the n-body work units, though one or a few are now completing successfully. My computer is now going into high priority mode as well. One failed at 82%. Another failed at 85%. It looks like I am getting a mix of successes and failures. These failures were under BOINC x64 version 6.10.60. Since BOINC released a new version, 6.12.26, I have upgraded since then.


Is it failing for both the n-body simulation application and the regular one? Or just one of the two?

With just 4 clicks you to can see Jesse's returned results...


When I clicked they were all correct...

LOL...time to increase the time before deletion a little more?
ID: 48877 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jesse Viviano

Send message
Joined: 4 Feb 11
Posts: 86
Credit: 60,913,150
RAC: 0
Message 48882 - Posted: 20 May 2011, 17:02:32 UTC

Could it be possible that your settings are not accounting for the possibility that when BOINC responds that there are 12 processors to use that it is failing to account for the hyperthreading my 6-core hyperthreaded processor is presenting? BOINC might be expecting twice the processing power it is actually getting. I noticed that some of the work units that fail on a Core i7 with hyperthreading work well on a Core 2 without hyperthreading.
ID: 48882 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : News : more maximum time limit elapsed bug stuff

©2024 Astroinformatics Group