I've had a few of these today, they seem to use next to no GPU resources, and be almost completely CPU work, even though they are scheduled for the GPU.

Is this a symptom of a dying GPU, or a faulty WU?



254 SETI@home 12/12/2014 09:05:13 Task ap_08jn14ab_B2_P1_00001_20141210_07383.wu_0 exited with zero status but no 'finished' file

---

Happy Crunching,

Graham

The "Result exited with zero status but no 'finished' file" error can mean that the science application is unable to find the last checkpoint that it wrote. For some reason, the latest checkpoint wasn't written to disk, perhaps due to a corruption of the task, the disk or because the directory was locked (possibly due to an anti virus scan or anti spyware scan).

---

With each crime and every kindness we birth our future.

...Is this a symptom of a dying GPU, or a faulty WU?...exited with zero status but no 'finished' file

In short neither. It's a symptom of (the BOINC client) using fixed timeouts in a multithreaded (Modern OS) environment. If it's a substantial change from your host's familiar behaviour, I would consider examining what device drivers changed recently, and system DPC latency behaviour.

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

A little further info...

It seems to be every AP WU that runs on my GTX580. The work on the GTX570 all goes through just fine, AP and MB.

I've just suspended APs until the current one finishes and I'll see what happens with the MBs on that GPU.

Also, all the AP WUs that run on the GTX580 seems to be running almost completely in CPU, with very little GPU usage, about 85%+ CPU time, about 12% - 17% CPU on the GTX570!

Hummm

---

Happy Crunching,

Graham

More info, it's not every WU on that GPU, it seems to be the SAME WU running on that GPU over and over again - it errors. so it restarts and has the same symptoms!

Go I abort that WU or what?

---

Happy Crunching,

Graham

Having watched go round again, I've now aborted it

All GPUs working OK on the last 2 AP WUs and the MB WUs.

Thanks for the comments.

---

Happy Crunching,

Graham

From stderr of that aborted task:

ERROR: some exception inside long FFA, probably video-driver restart, restarting app...

I would recommend to decrease FFA block to avoid such situations in the future.

From stderr of that aborted task:

ERROR: some exception inside long FFA, probably video-driver restart, restarting app...

I would recommend to decrease FFA block to avoid such situations in the future.

I would recommend using the Lunatics 0.43a installer to get the version of AP7 which automatically decreases FFA block when the WU data needs it.

                                                                   Joe