Message boards : Number crunching : Credit
Author | Message |
---|---|
Mario Kaiser Send message Joined: 14 Dec 08 Posts: 2 Credit: 165,705 RAC: 0 |
Hi, why i have get for WU ID 197543745 CPU Time 34,145.80, claimed credit 119.35 but granted credit 0????? I´m not amused.... Greets Mario |
mikey Send message Joined: 5 Jan 06 Posts: 1895 Credit: 9,202,276 RAC: 3,232 |
Hi, If you look at the details you will see why: <core_client_version>6.4.5</core_client_version> <![CDATA[ <stderr_txt> # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 # cpu_run_time_pref: 28800 failed to create shared mem segment CreateSemaphore failure! Cannot create semaphore! # cpu_run_time_pref: 28800 ====================================================== DONE :: 1 starting structures 34145 cpu seconds This process generated 10 decoys from 10 attempts ====================================================== BOINC :: Watchdog shutting down... BOINC :: BOINC support services shutting down... called boinc_finish </stderr_txt> ]]> Validate state Workunit error - check skipped The unit just stopped crunching so there is a problem with either your pc or the unit, or the program. As you can see by the other person that crunched the same unit they had the exact same problems that you did. Hopefully this is a one off, but crunching is like that sometimes, you put in a ton of work, figuratively speaking, and get nothing for it. |
Evan Send message Joined: 23 Dec 05 Posts: 268 Credit: 402,585 RAC: 0 |
Workunit error - check skipped I don't think anything was wrong. Correct me if I am wrong but I believe that the error message refers to too many results. Mario Kaiser was the third person to send in his result and only two are allowed. |
mikey Send message Joined: 5 Jan 06 Posts: 1895 Credit: 9,202,276 RAC: 3,232 |
Workunit error - check skipped You are correct but it does seem to be a scheduler problem, that he was issued a unit because the other person didn't return it, yet the other person was still allowed to return it and therefore Mario did not get credit. Mario was legitimately issued the unit, he should be the one getting the credits. Hopefully this does not happen that often. |
Mod.Sense Volunteer moderator Send message Joined: 22 Aug 06 Posts: 4018 Credit: 0 RAC: 0 |
This resending tasks when too many results can result is a known BOINC server issue. Rosetta Moderator: Mod.Sense |
mikey Send message Joined: 5 Jan 06 Posts: 1895 Credit: 9,202,276 RAC: 3,232 |
This resending tasks when too many results can result is a known BOINC server issue. I really think Dr. A and group need to work on this one, it says it was opened "Opened 2 years ago, Last modified 3 months ago"! This seems like an easy thing to check, send a unit to your own pc, abort it, send it to a second pc and don't return it on time. Then when the unit gets sent to a 3rd pc, return the unit from the 2nd pc, voila problems. I understand this is NOT your baby and all that, and I do NOT want you to think I am thinking this is a problem you can fix. I was a forum moderator over at Seti for awhile and have dealt with Dr. A and crew in the past. Thank you for the link and we can always hope that a fix will be coming before hell freezes over! |
ramostol Send message Joined: 6 Feb 07 Posts: 64 Credit: 584,052 RAC: 0 |
At the same time one must consider the error message given: failed to create shared mem segment CreateSemaphore failure! Cannot create semaphore! # cpu_run_time_pref: 28800 ====================================================== DONE :: 1 starting structures 34145 cpu seconds This process generated 10 decoys from 10 attempts ====================================================== It clearly states that an error is registered. Furthermore it says that with a preferred runtime of 28800 cpu seconds 10 decoys was computed using 34145 cpu seconds - the ultimate model must have been quite longwinded indeed compared to the others. I should say that something happened when computing the last model, and the result was invalid. BOINC behaved as it should, and all received due credits. As for the “too many results” message: I have twice observed this message being issued not upon the server receiving a result, but when the server issued a task for the third time to a pc. Since the third participants later delivered valid results and got their credits I consider this message as a mere warning, not an explanation of subsequent developments. |
Message boards :
Number crunching :
Credit
©2024 University of Washington
https://www.bakerlab.org