Bad WUs?

TemplarTemplar You first.
edited October 2003 in Folding@Home
I've been getting this error a lot. I think about 3 times each for each instance (I run two on the same machine. HT and all that fun). It's happened three times in a row, and I'm wondering if it's maybe something I've done... ;(

Comments

  • muddocktormuddocktor
    edited October 2003
    I've seen that bad checksum error on tinker WU's a couple of times, but not very often. It could very well be a WU error but first, what P4 are you running and how far have you overclocked it, if at all. Secondly, have you had problems with processing any Gromacs WU's lately too.

    If it did this 3 times in a row and processed the same WU, then it was most probably a defective Tinker WU, which occasionally happens even with the old Tinker core work.

    EDIT: BTW, if you aren't using it, then add the -advmethods flag in the target on your startup shortcut. This will make sure that the great majority of the work that you get is Gromacs WU's which fold much more efficiently on a P4 than Tinker WU's.
  • TemplarTemplar You first.
    edited October 2003
    CPU-Z shot below.

    I don't really know how to tell what kind I'm folding with. I've gotten nearly that exact error everytime (save the checksum). These two instances haven't turned in a single WU in 3 days though, and ever since I put -forceasm and -advmethods on them, they fold a given WU in about 8 to 10 hours or so.

    However, I don't always supervise this box (family PC, basically), so occasionally the clients may get closed and I won't notice for a day. That may have happened here. I know I've seen this error before though.
  • mmonninmmonnin Centreville, VA
    edited October 2003
    Go thru the log file the the FAH directory and find a line that says

    Project: 'project #' <run, clone, gen>

    If its the same then stanford servers might have sent you the same WU to try it again.

    Looks like the next WU is a gromac tho.
  • TemplarTemplar You first.
    edited October 2003
    Appearently Protein p690_L939_K12M_532K was causing the problem. Both clients were folding it, so that may have something to do with the error. This only happened once, by the way, but for both clients (both were folding it). That WU folded extremely slow (1 Day and a 7 hours). The clients HAD been shutdown without my knowledge, and for a good 8 hours. :doh:

    After it had finished that bad WU, it actually started on it again. I deleted the work directories for the one that wasn't folding it (CPU1 is folding Project 373 and CPU2 is folding 340). Looks like it's fine now. Thanks for the help :D
Sign In or Register to comment.