Client error Compute error - sorry, but who's fault is this?

Message boards : Number crunching : Client error Compute error - sorry, but who's fault is this?

To post messages, you must log in.

AuthorMessage
Gerhard Klünger

Send message
Joined: 24 Dec 05
Posts: 21
Credit: 936,103
RAC: 0
Message 33821 - Posted: 31 Dec 2006, 11:49:20 UTC
Last modified: 31 Dec 2006, 11:49:41 UTC

Please see results for workunit 47809942
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=47809942

This WU was crunched at least on two different machines from 2 different users. In both cases the computation terminated with errors.

53829172 237278 25 Dec 2006 9:20:09 UTC 25 Dec 2006 10:13:09 UTC Over Client error Compute error 1,368.66 3.23 ---
53839869 176175 25 Dec 2006 11:09:46 UTC 26 Dec 2006 1:17:41 UTC Over Client error Compute error 7,518.06 20.62 ---

If I have to change there something on my system - please tell me.

BTW: If it is not my fault, I don't find it fair not to grant any credit for the work done.
Gruß, Gerhard
ID: 33821 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Tom Philippart
Avatar

Send message
Joined: 29 May 06
Posts: 183
Credit: 834,667
RAC: 0
Message 33825 - Posted: 31 Dec 2006, 12:44:45 UTC - in response to Message 33821.  
Last modified: 31 Dec 2006, 12:48:02 UTC

It isn't very likeley that a WU hits 2 computers which aren't 100% rosetta stable. In my opinion this WU has some bugs.

2nd question:
It is due the credit system:
invalid result=no credit
The credit system can't make a difference between a WU that fails because of a computer and a buggy WU.
ID: 33825 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
MikeMarsUK

Send message
Joined: 15 Jan 06
Posts: 121
Credit: 2,637,872
RAC: 0
Message 33828 - Posted: 31 Dec 2006, 13:17:29 UTC
Last modified: 31 Dec 2006, 13:19:09 UTC

Actually it would well have been due to coincidence. If you look at the other computers results, it's failing about 30% of the time.

I'd say the other computer is very unreliable, and it's just bad luck that the one WU you had which failed happened to also have been processed by an unreliable box.

If you want to check your own PC for reliability, I'd recommend running Prime95's Torture Test for 24 hours, but one failure out of 30 isn't too bad.

Regarding credit for failed work units, unfortunately it makes it possible for people to give themselves unlimited credit if that is allowed.

ID: 33828 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Buffalo Bill
Avatar

Send message
Joined: 25 Mar 06
Posts: 71
Credit: 1,630,458
RAC: 0
Message 33847 - Posted: 31 Dec 2006, 21:16:50 UTC

Credit is granted for failed WU's (Claimed = Granted) up to a maximum of 20 credits. Click on the "Result ID" link for any failed WU and it will show at the bottom of the Result page. It may take a day to show up and I think that is the only place it will show.
ID: 33847 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Gerhard Klünger

Send message
Joined: 24 Dec 05
Posts: 21
Credit: 936,103
RAC: 0
Message 33850 - Posted: 31 Dec 2006, 22:20:42 UTC - in response to Message 33847.  

Credit is granted for failed WU's (Claimed = Granted) up to a maximum of 20 credits. Click on the "Result ID" link for any failed WU and it will show at the bottom of the Result page. It may take a day to show up and I think that is the only place it will show.


Thank all of you for your replys. For years I didn't care for credits, but since the thing happend as described in
BOINC stops working on a running PC XP
and
Can a 'malformed' Workunit crash BOINC?
I had a closer look to the results for what is going on there´and found there some strange things.

I consider this case as settled.

BTW: Happy New Year - at my location this will be in 40 minutes.
Gruß, Gerhard
ID: 33850 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Number crunching : Client error Compute error - sorry, but who's fault is this?



©2024 University of Washington
https://www.bakerlab.org