D2OL Problems

[The Reaper]

Since the new client came out I've finished less then 80 in a weeks time. This sould be close to my daily output not my weekly weekly output. Sometimes I get a massage "Incomplete Atom" using 100% CPU power. Other times everytime looks fine, but it will use no CPU power. Rebooting only helps for a couple of hours. I'm ready to throw in the hat on this project and find something else. System specs....

WinXP home (no sp1)
AXP 1600+ @ 1785MHz
256M RAM (256 more on the way)

 

[networkman]

Have you tried back down on the OC? Could be there's something unstable and until you run DC project like D2OL that is VERY cpu/ram intensive that the problem hasn't occured. I've had similar problem with others systems at home that ran fine with games but not with DC projects.

 

[RemyCanad]

Yeah I second that. It may be cranking out tons of them but alot might be getting rejected becuase they are in some way mangled up.

Also I had this problem too at one point. I have a 1.3Ghz duron that was cranking out less than my 200mhz pentium 1 MMX! I just reistalled the client and it cleared it right on up.

 

[The Reaper]

I thought about turning down the OC, but it ran fine on the old client. I'll try reinstalling the client to see if it helps. Thanks for the replys!

 

[The Reaper]

Reinstalling the client didn't help, so I'm backing my OC down to see if it helps. Time will tell.

 

[dmcowen674]

Only 2 nodes running D2OL successfully, 2 machines have stalled and one has never completed a Candidate, it counts up the que and then goes back down to 0 and just keeps doing that.

Stuck - 5103 12238 42 840 6 nci_1150_1472 1 days, 19 hr, 18 min, 08 sec
Stuck - 5704 12176 24 480 2.67 nci_1135_1562 3 days, 15 hr, 42 min, 06 sec
Never completed a unit - 8016 12177 0 0 0 - 0 ms

 

[mastertech01]

Originally posted by: dmcowen674
Only 2 nodes running D2OL successfully, 2 machines have stalled and one has never completed a Candidate, it counts up the que and then goes back down to 0 and just keeps doing that.

Stuck - 5103 12238 42 840 6 nci_1150_1472 1 days, 19 hr, 18 min, 08 sec
Stuck - 5704 12176 24 480 2.67 nci_1135_1562 3 days, 15 hr, 42 min, 06 sec
Never completed a unit - 8016 12177 0 0 0 - 0 ms

Dave delete all your files in the Structs folder and redownload cands. They are coming in slow again so it may be awhile. If you leave a bad cand in que it will keep coming back.

 

[dmcowen674]

OK. I'll give that a try.

 

[The Reaper]

I think I'm gping to quit this project till they get their act together. This didn't happen with the old client. I'm sick of wasting my time. :|

 

[dmcowen674]

Mastertech, deleting the strucks folser worked for 2 of the 3 machines that were stuck but and a big but. Each machine spit out 1 candidate and then frozen on whatever candidate they are on now. The other machine that the candidates go up and down and it goes from starting task to pending back and forth I suspect is some kind of JAVA problem.

They really need to do something with the code, it needs to have some kind of timeout and delete bad candidate function.