Forget about my last post at bottom, after 50h one WU resets from 46,000Ts to 33,000 (about), then after couple of h, error occured. Then in the morning, smietrical WU wich was about 50,000Ts has crashed too.
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/results.php?userid=15528&PHPSESSID=89676cd37c545539a38d6902534d2de7

I'm totally confused, i was trying everything: HT on and off, memtest are ok, machine is not OC, i even tryed to decrease performance. Still exit code -5.
I have win 2k with all pathes and SP. Components are rather good. Gigabyte MB, Intel Prescott proc. Temp are very low 51 C for high stress?
I'm close to madness it's very anoying.
THNX for any help
>>Hello to everyone with p4 HT. I had some problems with error shown in topic. I was looking for any help or explanation and I found thread and info that this error is similar to many causes, specially hardware.
I start with the bios and found function "limit threads to 3", switch to enable. Now it works fine, nice and smooth. I hope this post would help anyone.
All the best
P.S sorry for my english:)

I have the same problem. All my results end up with the same exit code.
Unrecoverable error for result 34or_100167514_0 ( - exit code -5 (0xfffffffb))
after 12 hrs of computing.

But I own an Athlon XP 2600+. I do have Win2K with all service packs as well.

Any suggestions?

Thanks
Diego

>I have the same problem. All my results end up with the same exit code.
>Unrecoverable error for result 34or_100167514_0 ( - exit code -5 (0xfffffffb))
>after 12 hrs of computing.
>
>But I own an Athlon XP 2600+. I do have Win2K with all service packs as well.
>
>Any suggestions?
>
>Thanks
>
Often -5 is a hardware error. CPDN stresses system hardware as much or more than just about any other software around. It is more intense, and intolerant of errors than other DC projects or games. Minor processor errors or memory errors lead to aborted models. Best to check out your system

Hardware specific tests and checks can be found here

http://www.climateprediction.net/board/viewtopic.php?t=2126

Basic hardware maintenance advice can be found here.

http://www.climateprediction.net/board/viewtopic.php?t=2124

Good luck.

> >
> Often -5 is a hardware error. CPDN stresses system hardware as much or more
> than just about any other software around. It is more intense, and intolerant
> of errors than other DC projects or games. Minor processor errors or memory
> errors lead to aborted models. Best to check out your system
> ....

Hello I was checking this a lot of times. Strange that Rendering with couple milion of polys in 3dsMAX works fine, but BOINC don't
The strangest thing is that last results was processed on same configuration, but Overclocked to 3,2, and after OC it was working fine LOL

Same problem.

After hunders of hours of computing, the outcom was -5 :(
Standard P IV 2.0 GHz 512Mb Mem.
No overclocking, no special HW or SW.
XP SP2.

Shouldn't boinc correct this kind of errors?
If it *is* a HW error, or the system *is* unstable, retry the computation at a later time.
Just discarding hundreds of hours of research is sad :_(