Message boards :
Number crunching :
13-14 Feb 2012 - WU restart - but some short-lived
Message board moderation
Author | Message |
---|---|
Send message Joined: 10 Aug 07 Posts: 54 Credit: 813,704 RAC: 116 |
Greetings!! I noticed that WU have restarted - Thanks! I started to worry when the first 4 out of 5 had short run-times. (the 5th is still running.) I am not sure how to interpret sterr out. Is this an error? Or just close-of-file? All 4 say the same thing for stderr output (time stamps different..): <core_client_version>6.10.58</core_client_version> <![CDATA[ <stderr_txt> 20:00:52 (4370): called boinc_finish </stderr_txt> ]] Here is other data: userid=81571 hostid=10169683 - AMD 64 wu, task, run time wuid=6566397 resultid=14306839 _20.20 wuid=6566039 resultid=14306118 121.79 wuid=6565907 resultid=14305855 __8.17 wuid=6565400 resultid=14304840 _10.84 each of these 4 were validated with another user - with a short run-time. Is it just chance that all 4 were, naturally, short-lived? Or was there an error - somewhere? Thanks!! Jay |
Send message Joined: 7 Feb 13 Posts: 19 Credit: 1,478 RAC: 0 |
Greetings!! Boinc_finish means the task has ended. Here is other data: If they validate then they either the tasks completed properly and the results are good else another computer made exactly the same errors and produced a "bad" result that matches your "bad" result. That is very, very unlikely. Is it just chance that all 4 were, naturally, short-lived? Yes, just chance. Most Sixtrack tasks run longer, some are very short like these. Yes, it seems unlikely you would receive 4 short ones so close together but it happens. |
Send message Joined: 10 Aug 07 Posts: 54 Credit: 813,704 RAC: 116 |
Thanks for the info.... Now I know what a non-error looks like.. Thanks again, Jay |
Send message Joined: 12 Dec 05 Posts: 31 Credit: 9,709,398 RAC: 0 |
don't get too excited yet. remember, only 5 days ago Eric said that we might not see new work for a whole week or two. now perhaps he got his analysis done early and tasks are back. then again, it could just be a short injection of test tasks again. i wouldn't be surprised if the "ready to send" queue on the server runs dry again before we start seeing new tasks on a regular basis again... |
Send message Joined: 9 Jan 08 Posts: 66 Credit: 727,923 RAC: 0 |
By the way, some work started trickeling in again. So Eric might be done now. |
©2024 CERN