Hi,

one of my hosts is crunching this WU. Yesterday, after about 360 hrs of computing, the completion % was at more than 60%, today the % fell back to 0%, though the CPU time didn\'t fall back to 0.

Is this normal?

Thanks

darkpella

---

If you display the graphics, what date does it say?

If you display the graphics, what date does it say?

It shows the world color map, date is 01/02/1811.

---

So it really has gone back to the beginning. The last trickle received by the CPDN server was the conclusion of the second phase, so it looks as if something has gone wrong in the post-processing for that phase. Another participant had a model that got stuck in the post-processing, but never went any farther (here).

If you have a backup, then restoring that might be a good idea - it might get through the post-processing on a second attempt. Otherwise, I would leave it for one trickle back in the first phase; if by that time it hasn\'t corrected itself then abort it.

Other people might be more generous to the model, but my experience is that once they\'ve gone wrong they stay that way. Sorry not to be more encouraging. Perhaps someone else might have an idea for a workaround.

I think Iain is right. Your best chance is a rerunning from a backup if you have one available. If not, you can rerun the model from where it is at, but you won\'t get any credits until you get into the 3rd phase, and you\'re taking the chance that something bad could happen to it at or before the end of the 2nd phase.

I\'ve had this happen to me a few times over the years. Usually it occurred when there was an unclean shutdown of the PC while running the model.