Lost Credit
2 Jul 2009 16:59:00 UTC
Topic 194430
(moderation:
I had a work unit that finished and was a success with a lot of credit claimed. Credit = 0 There was not an error of any kind. Was this because of the resent shutdown? This is the 2nd. time in the last year that many hours were put into the work units and I received Zip for credit. Why was no credit given for all that work? Thanks for your response. :-)
Lost Credit
)
Hi,
please unhide your computers (or post it's id), this allows us to check for error messages.
Michael
Team Linux Users Everywhere
RE: RE: I had a work unit
)
Hi Michael,
I already checked, there were no errors listed so I figured it might be due to the shutdown. The work unit was deemed a success with 172.71 credits claimed and 0 given. They have me listed with several computers and I have always had just one.
RE: Hi Michael, I already
)
Outcome "success" only means, that computation finished w/o errors on your computer. Then validation follows. So it would be helpful to get maybe the WU id?
Michael
Team Linux Users Everywhere
RE: I already checked,
)
Hi Ronald,
Thanks for "unhiding" your computer as we can now see the details of what happened.
If you check the task details for the result in question and look right down at the bottom of the page, you will see that the validate state was marked "invalid". The task was sent out to a third computer and when that result was returned, all three were compared. Unfortunately your result was deemed to be the "odd one out" and was consequently given zero credit.
It is possible that differences in hardware and software between the three computers that calculated the result could give enough difference in the respective answers to account for the "invalid" decision. This is rather unlikely and in most cases it is much more likely to be a hardware glitch or fault in the machine denied credit. As your machine appears to be quite old, have you checked it recently for things like overheating due to old and tired fans (both CPU and PSU) or excessive fluff buildup that is impeding the flow of air?
With respect to your "multiple hosts" problem, this can easily happen at your end when you play around with various things to try to get a machine to connect to the project. There is actually a nice simple solution. Just click on the "view computers" link on your account page and then click on the "all computers" link to see all the hosts that you have ever registered with the project. You will see multiple instances of your current host, a 733MHz PIII, as well as an AMD machine you must have used before that. If you click on the "most recent" computer ID (at the top of the page) you will pull up the full details of your current registered host. Down at the bottom of the page you will find a "merge" link that should allow you to merge all the previous instances of this host. Only you can see these details but I believe you will find eight previous host IDs that you can merge (tick the boxes) into the current host. If you wanted to, you could probably merge all the AMD hosts into one as well, giving you a final listing of just two machines.
See if that will work for you.
Cheers,
Gary.
RE: RE: I already
)
Thank you Gary,
All the other Projects worked fine along with Einstein and Einstein has for many years. I learned something new from you. I appreciate it. Have a nice 4th.
Ron
RE: All the other Projects
)
My experience is that the E@H science app is quite good at finding any small "weakness" in the hardware and can throw up errors where other projects' apps don't show any problems. Transient failures like this are the most frustrating things :-).
You're most welcome. Did you manage to get your hosts merged OK?
Cheers,
Gary.
No credit given!! I think
)
No credit given!! I think this may be 3 or 4 units that did not get credit. Please fix - until you do, I will only do SETI work units, and will not even try to do EINSTEIN units. All pending EINSTEIN units have been aborted.
Alan
RE: No credit given!! I
)
Those units did not get credit yet, this is completely normal and is called "pending credit". To ensure the scientific validity of the results returned by the volunteers, every piece of work is distributed to two different PCs. Credit is granted only after
a) both PCs have returned the results
b) both results match within some reasonable margin of error
This process is called validation. In your case, your PC was faster than "the other one". Your will receive credits eventually.
CU
Bikeman
RE: Those units did not get
)
And note that this process of validation and pending credit is exactly the same as is used at SETI.
RE: RE: No credit given!!
)
And some of his units now have credits granted, some are still pending and the rest were aborted by the user. Th user is also using an older 5.2.13 version of Boinc, but that shouldn't make any difference in how soon the credits get granted.