Blimey. Whole flurry of errors. All today (well, yesterday - 16 June). Had nothing like this for weeks.

https://boinc.bakerlab.org/rosetta/result.php?resultid=24427279

https://boinc.bakerlab.org/rosetta/result.php?resultid=24460877

https://boinc.bakerlab.org/rosetta/result.php?resultid=24463664

https://boinc.bakerlab.org/rosetta/result.php?resultid=24495408

https://boinc.bakerlab.org/rosetta/result.php?resultid=24513042

---

Ian Cundell, St Albans, UK

https://boinc.bakerlab.org/rosetta/result.php?resultid=24571715

i was viewing the graphics window at the time it failed incase that makes a difference

---

Want to search the BOINC Wiki, BOINCstats, or various BOINC forums from within firefox? Try the BOINC related Firefox Search Plugins

I have just joined the project. On one PC of the 9 WUs it has been sent it has successfully processed 5 but errored out on 4.

from my log:

17/06/2006 04:25:04 Unrecoverable error for result t299__CASP7_JUMPRELAX_SAVE_ALL_OUT_BARCODE_cterm2_nohelix3_hom001__681_83011_0 ( - exit code -1073741819 (0xc0000005))

Clues or advice?


A different unit to that above but some debug info to help devs.
https://boinc.bakerlab.org/rosetta/result.php?resultid=24466158

---

First error ever on this machine (31,000 credit):

https://boinc.bakerlab.org/rosetta/workunit.php?wuid=19785224

stderr out

<core_client_version>5.5.0</core_client_version>
<message>
Incorrect function. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
# random seed: 3706611
# cpu_run_time_pref: 14400
# cpu_run_time_pref: 14400
ERROR:: Exit at: .dock_structure.cc line:401

</stderr_txt>

btw [BOINCUK]Tigher, (0xc0000005) is usually a memory error, in my experience.

btw [BOINCUK]Tigher, (0xc0000005) is usually a memory error, in my experience.

Gulp! Hmmm thanks.

---

Another problem - looks the same from this end as the other ones I had.

https://boinc.bakerlab.org/rosetta/workunit.php?wuid=20802010

When I leave from working on the computer, I'll exit IE to see if that helps.

Bandit's Mom

---

https://boinc.bakerlab.org/rosetta/hosts_user.php?userid=94664

I have been accumulating computation errors lately.

---

https://boinc.bakerlab.org/rosetta/result.php?resultid=24638007

This WU created about three good models with energy minima between -200 and -300. then it failed to do more good models which each succeeding model completing within minutes and always the same energy minimum of about -30. Watching on the graphics showed a stretched protein where no folding was achieved. I "aborted" the model the soft way with 6 restarts of BOINC (to prevent sending out the same WU).

I watched such WU in the past. Perhaps there is a pattern.

---

This WU created about three good models with energy minima between -200 and -300. then it failed to do more good models which each succeeding model completing within minutes and always the same energy minimum of about -30.

I for one have been HOPING to see WUs that would act like that. If you knew that a -300 was possible, and you are sitting at a -30, there are cases where it might be SMART to bail on this one and invest the time in pursuing something with more potential.

I don't know that this is what happened in your case, I'll leave that for the project team to assess. I just wanted to point out that it is the TYPE of thing that I think we'll see more of as the algorythm gets smarter.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

This WU created about three good models with energy minima between -200 and -300. then it failed to do more good models which each succeeding model completing within minutes and always the same energy minimum of about -30.

I for one have been HOPING to see WUs that would act like that. If you knew that a -300 was possible, and you are sitting at a -30, there are cases where it might be SMART to bail on this one and invest the time in pursuing something with more potential.

I don't know that this is what happened in your case, I'll leave that for the project team to assess. I just wanted to point out that it is the TYPE of thing that I think we'll see more of as the algorythm gets smarter.

I agree! Using previous result for "pruning" decision is an idea that for a long time crossed my mind. I'm a bit in chess engine programming and in these engines a lot of "pruning" is done in positions where one side is just too worse to have any chance of reaching the current score with any move. However in the case reported it was most certainly something different, since the models finished successively in a few minutes without really folding the protein (it was stretched in the graphics) and with always the same score. In the end I had over 150 models of which only three had not been "aborted".

---

stuck at 74.101% Rosetta 5.22 Windows 0.0000% of CPU usage
Thus, aborted by hand after 3 hours of IDLE time!
https://boinc.bakerlab.org/rosetta/result.php?resultid=24659040

Thanks

---

Click signature for global team stats

For the past week or so I've been getting 2-3 crashes per day. The failed work units show up as "Compute Error" with no credit. Do I need to report this? Or will the appropriate party see these errors and be able to deal with them on their own?

---

Do I need to report this? Or will the appropriate party see these errors and be able to deal with them on their own?

It is "HELPFUL" if you report them. It gives the opportunity to ask you questions about your computing environment so they might learn more about the system that's seeing the failure. It is not "required".

Credit for failed WUs is issued once the daily credit run is made. You will see this when you display the WU details... not on the WU listing. Like this one for example.

It looks like most of them were ended by the "watchdog". One was a -107 error (which is something that's been under review for a while already).

The watchdog is trying to assure your computer doesn't get stuck in an unexpected loop on a work unit. If it notices no progress on a work unit in 5 restarts, then it ends it. Do you restart this computer frequently? Or have a number of other projects running in BOINC?

If you would, go to your General Preferences, and let us know what you have set for "Switch between applications every...minutes", and for "Leave applications in memory while preempted?". And is Rosetta your only BOINC project?

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

A crash.

https://boinc.bakerlab.org/rosetta/result.php?resultid=24876847

It happend when i was shutting down grafics window.

Anders n

---

It looks like most of them were ended by the "watchdog". One was a -107 error (which is something that's been under review for a while already).

Correction, I misread that "watchdog is shutting down" message (again!). I keep thinking this message indicates that the watchdog is shutting down the WU, not just ending itself as a normal end of processing a WU.

Most of their errors were -107s.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

The watchdog is trying to assure your computer doesn't get stuck in an unexpected loop on a work unit. If it notices no progress on a work unit in 5 restarts, then it ends it. Do you restart this computer frequently? Or have a number of other projects running in BOINC?

If you would, go to your General Preferences, and let us know what you have set for "Switch between applications every...minutes", and for "Leave applications in memory while preempted?". And is Rosetta your only BOINC project?

I'll try to answer your questions here:

Machine is rarely restarted, once every 2-3 days.

This is the only project I have under BOINC. No other background/SETI type applications are installed.

I'm not sure where this "General Preferences" dialog is you're referring to. I don't see anything like this in BOINC.

I am an accomplished C++/Java/.NET developer w/ Visual Studio installed on this box if you need me to grab a stack trace, I'd be happy to next time!

I'm not sure where this "General Preferences" dialog is you're referring to. I don't see anything like this in BOINC.

Now that you are viewing this message board, click the "Participants" link in the heading of the screen. In the "Preferences" section, click the link for "view or edit" of General preferences. Any changes made there require BOINC to update to the project to take effect. This is done from the projects tab of BOINC, select Rosetta, then click the update button.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

In followup to Message ID 18855, as long as I don't have IE running, I don't seem to have any BOINC problems. If I leave IE on, I have intermittant BOINC crashes. For me, it does not seem to be the screensaver at this time.

Bandit's Mom

---

Ran into a compute error with 522.

6/20/2006 12:12:35 PM|rosetta@home|Unrecoverable error for result t304__CASP7_JUMPRELAX_SAVE_ALL_OUT_BARCODE_hom001__691_17229_0 ( - exit code -1 (0xffffffff)).

Looks like it was: https://boinc.bakerlab.org/rosetta/workunit.php?wuid=21222160

What other information should I provide?

---

Now that you are viewing this message board, click the "Participants" link in the heading of the screen. In the "Preferences" section, click the link for "view or edit" of General preferences. Any changes made there require BOINC to update to the project to take effect. This is done from the projects tab of BOINC, select Rosetta, then click the update button.

You didn't say what these 'should be' so I'm just reporting what they currently are and not changing anything:

work on batteries: no
work while in use: no
idle: 3 mins
hours: (no restrictions)
leave in memory: no
switch between: 60 mins
multiprocessors: 0 processors (although I have two of them!?)
use at most: 100 percent of CPU

---