After running Climate on this computer 907566 for more as 3 years without real problems, suddenly all wu of climate started failing in about 20 seconds.I am not sure if this was at the same moment I did the update to 6.12.26. Today I removed boinc totaly from this computer and installed the 6.10.58 release. Then I attached to the project and also now the workunits are failing directly. I am running OSX 10.6.8. Is there anybody that has knows a solution for this.

It's explained in the thread immediately before yours, about crashes in 45 seconds.

---

Backups: Here

This is not the same situation. I my situation also hadcm3n wu crach. During the last month all wu on this computer are failing. On all my other computers I see sometimes a failed wu but most of the wu are running fine.

Unfortunately, BOTH types of models have been faulty recently.

Perhaps a better place to keep an eye on, is the News and Announcements thread at the top of Number crunching. This is where posts of importance are made.

---

Backups: Here

On the Mac computer:

The error you're getting for AM3 regional models is:
Model crashed: INITDUMP: Wrong no of atmos prognostic fields

Not sure what that means. Seems like the researchers made a technical mistake with that batch of tasks.

However, the HadCM3 models crash with this:
process exited with code 22 (0x16, -234)

which AFAIK means permissions problems. Perhaps there's something on your end that needs fixing?

This is the third report I've seen of tasks starting to crash immediately on a Darwin system after upgrading to BOINC 6.12.26.

I've just posted this sticky about this; the solution is a project reset (or detach and reattach if you prefer).

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer