11/30/2008 8:08:16 PM|Milkyway@home|Reason: Unrecoverable error for result nm_stripe82_1_71_1228076157_0 ( - exit code -1 (0xffffffff))
11/30/2008 8:08:16 PM|Milkyway@home|Computation for task nm_stripe82_1_71_1228076157_0 finished
11/30/2008 8:08:16 PM|Milkyway@home|Output file nm_stripe82_1_71_1228076157_0_0 for task nm_stripe82_1_71_1228076157_0 absent
11/30/2008 8:08:16 PM|Milkyway@home|Computation for task nm_stripe82_1_22_1228076155_0 finished
11/30/2008 8:08:16 PM|Milkyway@home|Output file nm_stripe82_1_22_1228076155_0_0 for task nm_stripe82_1_22_1228076155_0 absent
11/30/2008 8:08:16 PM|Milkyway@home|Starting nm_stripe82_1_37_1228076156_0
11/30/2008 8:08:16 PM|Milkyway@home|Starting task nm_stripe82_1_37_1228076156_0 using milkyway version 4
11/30/2008 8:08:16 PM|Milkyway@home|Starting nm_stripe82_1_38_1228076156_0
11/30/2008 8:08:16 PM|Milkyway@home|Starting task nm_stripe82_1_38_1228076156_0 using milkyway version 4

Not sure what is going on. I think the first two workunits I did finished normally, at least it looks like it from the log, but i'm not sure. "Insta Purge" SUCKS!It would really be nice to be able to look back on at least the last 8 workunits completed.
Also, Checkpoints are still not working.

---

4870 GPU
4870 GPU

Try setting your venue prefs to leave MW apps in memory.

This will help prevent the app (known to be a little fragile in this regard) from crapping out if the CC switches away from it during execution for any reason.

That will leave just a BOINC BSOD/shutdown or system power down as the other reasons for lost Milkyway tasks.

Travis said earlier there is another version coming which will address the problem of proper checkpointing and then these issues should go away.

HTH,

Alinator

Looking up the exit code at this Boinc FAQ it tells me:

Exit code -1 (0xffffffff) seems to be an incompatibility between the science application and your video card. You'll run into it when you use the graphics or screen saver, tasks running at the time when either is activated will error out with the above error message.

It's always good to report these errors on the project forums, especially when the project doesn't have graphics.

Mostly found on ATI video cards.
Try updating your video card drivers and DirectX version.

But the only sure workaround in case the project does have graphics is not to look at the graphics or to use the screen saver.

-------------------------------------------------------------------------------

Did you try to view graphics for the workunit or have Boinc set as your screensaver?

---

Never surrender and never give up. In the darkest hour there is always hope.

Normally that's good advice, but in this case I was experimenting around and tried deliberately switching away from Milkyway with the pref set to remove from memory.

When I switched it back in the app errored out with the same error code and a 'Can't read into stream integrals' message in stderr.

Might just be co-incidence, but I haven't had any problems since I set it back to leave in memory. I have mine set to do that normally since I also run Leiden, and not all of the trajou applications have checkpointing.

Alinator

Mostly found on ATI video cards.
Try updating your video card drivers and DirectX version.

But the only sure workaround in case the project does have graphics is not to look at the graphics or to use the screen saver.

-------------------------------------------------------------------------------

Did you try to view graphics for the workunit or have Boinc set as your screensaver?

No, I wasn't even on the computer and I don't use any screen saver either. I have my monitor set to turn off after 10 minutes of inactivity
My graphics card is RADEON X300 SE which is an ATI card I believe. I checked for driver updates but mine is current. Something else must be able to cause that particular error.

---

4870 GPU
4870 GPU

Normally that's good advice, but in this case I was experimenting around and tried deliberately switching away from Milkyway with the pref set to remove from memory.

When I switched it back in the app errored out with the same error code and a 'Can't read into stream integrals' message in stderr.

Might just be co-incidence, but I haven't had any problems since I set it back to leave in memory. I have mine set to do that normally since I also run Leiden, and not all of the trajou applications have checkpointing.

Alinator

It just seems strange that this behavior results in an error / exit code that is listed as being related to a graphics issue.

---

Never surrender and never give up. In the darkest hour there is always hope.

There are a couple problems reading some of the input files in versions 0.4 and lower. When Dave compiles the windows code and upgrades to 0.6 this should go away.

---

I set my prefrences to leave apps in memory. Hop that works. I hate having the work crap out on me.

---

4870 GPU
4870 GPU

There are a couple problems reading some of the input files in versions 0.4 and lower. When Dave compiles the windows code and upgrades to 0.6 this should go away.

LOL...

I was just getting ready to ask about where the x86 build of .6 was! ;-)

Guess the answer is, "Surprise, there isn't one yet." :-)

@ Gavin: I don't remember offhand, but IIRC there can be some ambiguity in how different projects interpret the BOINC specific error codes. It doesn't appear to be falling back to a Windows system error code.

This might be something to bring to Jord's attention for further investigation and an up date to his FAQ.

<edit> @ imcrazynow: I just thought of something else which might have caused it. Check your BOINC message log and see if the benchmarks ran right before the error. Depending on how many projects you run it might be a few task switches after the BM run.

Alinator

There are a couple problems reading some of the input files in versions 0.4 and lower. When Dave compiles the windows code and upgrades to 0.6 this should go away.

Unfortunately v 0.6 doesn't work either... in fact the code you've posted,when compiled on windows, can't even read the input files and crashes in calculate_integrals()...

---

Join Support science! Joinc Team BOINC United now!

There are a couple problems reading some of the input files in versions 0.4 and lower. When Dave compiles the windows code and upgrades to 0.6 this should go away.

Unfortunately v 0.6 doesn't work either... in fact the code you've posted,when compiled on windows, can't even read the input files and crashes in calculate_integrals()...

Weird, i might not have updated the code right... I'll re-release it and hopefully that should work.

---

As a matter of fact Yes,

11/30/2008 7:54:24 PM|Milkyway@home|Reason: requested by project
11/30/2008 8:07:42 PM||Running CPU benchmarks
11/30/2008 8:07:42 PM||Suspending computation - running CPU benchmarks
11/30/2008 8:08:13 PM||Benchmark results:
11/30/2008 8:08:13 PM|| Number of CPUs: 2
11/30/2008 8:08:13 PM|| 1325 floating point MIPS (Whetstone) per CPU
11/30/2008 8:08:13 PM|| 4235 integer MIPS (Dhrystone) per CPU
11/30/2008 8:08:14 PM||Resuming computation
11/30/2008 8:08:16 PM|Milkyway@home|Deferring communication for 1 min 0 sec
11/30/2008 8:08:16 PM|Milkyway@home|Reason: Unrecoverable error for result nm_stripe82_1_71_1228076157_0 ( - exit code -1 (0xffffffff))
11/30/2008 8:08:16 PM|Milkyway@home|Computation for task nm_stripe82_1_71_1228076157_0 finished
11/30/2008 8:08:16 PM|Milkyway@home|Output file nm_stripe82_1_71_1228076157_0_0 for task nm_stripe82_1_71_1228076157_0 absent
11/30/2008 8:08:16 PM|Milkyway@home|Computation for task nm_stripe82_1_22_1228076155_0 finished
11/30/2008 8:08:16 PM|Milkyway@home|Output file nm_stripe82_1_22_1228076155_0_0 for task nm_stripe82_1_22_1228076155_0 absent
11/30/2008 8:08:16 PM|Milkyway@home|Starting nm_stripe82_1_37_1228076156_0

This computer is attached to 8 projects but only MW and SETI are allowed work right now.

---

4870 GPU
4870 GPU