A user can fake the [edit: claimed credits, not the internal benchmark] by just editing a file just as a user can fake the numbers of decoys generated. The difference is that we can validate the decoys and catch cheaters. Unless we start encrypting data or use redundancy, the system is not going to be fake proof.

We fixed something in the interaction of Rosetta with BOINC to trigger more informative debugging messages upon crashes. Please continue to post what goes wrong!

I have four completed 5.22 wu's that seem to have uploaded OK, but whenever they try to 'report' I get a project down message. I don't see anything on the project status page that would cause this. Anyone else having problems or is it just me?

Hi everybody: Rom and I fixed this silly watchdog thing. I'm sending out work now with ralph 5.21! Thanks for helping us out with this.

Rom tells me it is waiting for the watchdog to finish for debugging.

Here is his response:

"When I added code .... to wait until
the thread is finished, it stalls for up to 30 minutes waiting until
watchdog makes its next check."

I think the watchdog can take up to 2x the cpu run time pref, which may explain the longer stalls.

I ran 3 work units.

Two actually completed but I suspect the "dormant" bug is still present as this first work (149120) unit completed in EXACTLY 1 hour with 36 min of CPU time, and this other one (149885) completed in EXACTLY 2 hours with 81 min of CPU time.

...

Mike

I ran into one of those 'stuck' abrelax wu's overnite:
resid 144202
running ralph 5.19 on my P-M laptop. It showed 100% done at 3:14:58 CPU time, but no activity, screensaver wouldn't come up (blank area only), task manager showed all cycles going to System Idle Process. Suspended acty and re-started Boinc, no-joy. Closed BM and re-started machine, brought it back up, no-joy. Aborted the WU and BM went right back to work. Didn't want to wait ?16hrs? (4x4) to see if it would somehow wake-up and finish on its own!

My experiance with this is that it will always finish within one hour of getting into this mode. If you look at the messages, and see when the WU started, take the minutes and seconds portion (like 5:10:20). When your PC time (wall time) reaches the same minutes and seconds (like 7:10:20), the work unit will complete. I know its weird, but thats what I am seeing.

Mike

Yes, this is really weird.

I have one stuck right now: http://ralph.bakerlab.org/workunit.php?wuid=129926

I took this screendump of the BOINC Manager: http://img.photobucket.com/albums/v680/lsh55/Setiweb/Ralph/BOINC_Mgr_Ralph.jpg


And this is the graphic: http://img.photobucket.com/albums/v680/lsh55/Setiweb/Ralph/RalphWU.jpg

This is really weird, the screen is rolling very fast to the left and this "band" is showing at the bottom of the screen.


I'll leave it as it is and see if it will finish as you say at the minute. Else I'll abort it.

I ran into one of those 'stuck' abrelax wu's overnite:
resid 144202
running ralph 5.19 on my P-M laptop. It showed 100% done at 3:14:58 CPU time, but no activity, screensaver wouldn't come up (blank area only), task manager showed all cycles going to System Idle Process. Suspended acty and re-started Boinc, no-joy. Closed BM and re-started machine, brought it back up, no-joy. Aborted the WU and BM went right back to work. Didn't want to wait ?16hrs? (4x4) to see if it would somehow wake-up and finish on its own!

My experiance with this is that it will always finish within one hour of getting into this mode. If you look at the messages, and see when the WU started, take the minutes and seconds portion (like 5:10:20). When your PC time (wall time) reaches the same minutes and seconds (like 7:10:20), the work unit will complete. I know its weird, but thats what I am seeing.

Mike

Woooot a new user of the day :)

Anders n

I miss him already

it doesn't appear, that many read these boards, or maybe they're just too busy/ignoring the cafe to respond.

and 5.4.6 isn't far off either.