I moved daniels' post here, as he's got Linux problems with 5.43 as well as having seen them on 5.41.

process exited with code 131 (0x83)
process got signal 11

---

Rosetta Moderator: Mod.Sense

12/16/2006 2:39:16 PM|rosetta@home|Unrecoverable error for result 1opd__BOINC_POSE_ABRELAX_NEWRELAXFLAGS_frags83__1449_313_0 ( - exit code -1073741783 (0xc0000029))

---

Is there something about FRA_ W.U.'s or is it they don't like me, the last

three i've had have all errored.

https://boinc.bakerlab.org/rosetta/result.php?resultid=52186178

12/16/2006 23:26:21|rosetta@home|Unrecoverable error for result FRA_t103_test_LARS_constraints_hom002_9_IGNORE_THE_RESTS_00001_0015478_0.pdb_1427_729_1 ( - exit code -1073741819 (0xc0000005))

---

Philip,

The integrated Intel (945) has been listed before when graphics problems happen.

Is that using the trackpad, nipple or a mouse ?

USB mouse of Logitech pedigree.

First try the newer graphics drivers, I think there are a lot of improvments in them Intel 945GM series drivers for WinXP

Thanks for the link - I installed the newer drivers. But I'm far too scared to try to test this again. At least on a job that's been running for several hours already. Maybe if I see a new job firing up I'll do some target practice on it.

P.S. Have you check to see if your battery is one of the recall (exploding Sony) ones. That laptop is on the risk list.

Oh dear. I might have a potential IED sitting on my lap?! *shudder*

Nah, it's got a Toshiba label. Hopefully that's authentic. But it does have a warning label that includes the word "explosion"... maybe time to add a few items to the christmas shopping list: Plasma screen, a very long VGA cable, remote keyboard and one of them nice blast proof bins.

---

Googloo and Peter, the error code indicates most likely a graphic-related problem. 5.43 has some improvement on this issue, but does not solve the problem totally. We are still working on searching for the real problem right now, meanwhile, please try to leave the screensaver or graphic off. Sorry for the inconvenience.

Is there something about FRA_ W.U.'s or is it they don't like me, the last

three i've had have all errored.

https://boinc.bakerlab.org/rosetta/result.php?resultid=52186178

12/16/2006 23:26:21|rosetta@home|Unrecoverable error for result FRA_t103_test_LARS_constraints_hom002_9_IGNORE_THE_RESTS_00001_0015478_0.pdb_1427_729_1 ( - exit code -1073741819 (0xc0000005))

---

I have this problem also with a intel mac running 10.4.8. The problem is surely not limited to the 945 intel video chipset since i have ati video (mbp). I just don't open the graphics and the simulation goes fine

---

Help! I seem to have been dropped from rosetta@home! I have no tasks, and I can't connect to the project. Is it down? Did I do something? I can ping Rosetta@home but I can't connect.

---

--hedera

Never be afraid to try something new. Remember that amateurs built the ark. Professionals built the Titanic.

Now this is odd. I couldn't connect to Rosetta, as I just said. So I brought up the Windows task manager, and saw that boincmanager.exe was still running (boinc.exe was not). I killed the process, then went to the start menu and started boincmanager.exe again. Lo and behold, there was rosetta@home again, running just fine. I'm running 5.43 now. Since there was zero activity in BOINC manager, and also zero logs, before I restarted, I have no error messages, and the current logs look perfectly normal.

---

--hedera

Never be afraid to try something new. Remember that amateurs built the ark. Professionals built the Titanic.

Thanks, Chu. I figured that out eventually (smacking my palm against my forehead). I have quit using BOINC as my screensaver until you get this fixed. That is, I have changed to the blank screensaver in Windows XP. The unrecoverable errors seem to have ceased for now.

---

Thanks Chu.

But i don't use the screensaver never have and didn't touch graphics button, I was not

there when it happened. As it only ran for around 6 min don't no but as i said

that makes last 3 of the FRA W.U.s that have crashed.

---

P.S. Have you check to see if your battery is one of the recall (exploding Sony) ones. That laptop is on the risk list.

Oh dear. I might have a potential IED sitting on my lap?! *shudder*

Nah, it's got a Toshiba label. Hopefully that's authentic. But it does have a warning label that includes the word "explosion"... maybe time to add a few items to the christmas shopping list: Plasma screen, a very long VGA cable, remote keyboard and one of them nice blast proof bins.[/quote]

It say Toshiba on it since it is in a Toshiba laptop, the actual manufacturer of the cells could still be Sony.

If you havn't checked (with the checker at Toshiba's site) do so, you may get a brand new fresh battery :-)




To others, this does not just happen with Intel Integrated graphics
ATI was the first manufacturer brought up, Intel was also another common one. Though Nvidia has seen this happen.

It may not be totally dependent on the graphics type.

It could also be BOINC (there are fixes to screensaver/graphics start/stopping/stalling in the 5.8.0 client) OR just something simple in the code that more often than not appears when graphics are initiated.

DirectX December refresh has just been released by the way.

P.S. I only asked about the mouse as i remember a boinc news post where some problems with older Intellipoint drivers...

---

Team mauisun.org

My granted credit is really low with this new version and its associated work units. It's killing my RAC.

Hmm. I haven't noticed the average granted credit going down significantly, but it has become MUCH more variable post 5.43.

My claimed credit has always averaged around 73.2 +/- 1. Granted credit for pre-5.43 WU's averaged 87.4, with std dev 4.77. My granted credit for 5.43 WU's is admittedly somewhat lower at 84.3, but the std dev is a whopping 14.4! Whereas >95% of pre-5.43 jobs had GC's between 77 and 95, I'm now seeing figures all over the map from 63 to 118.

If I understand the new credit scheme correctly, you might expect variability to be high when you're doing new types of WU's that don't have a long history of granted credits. So I might just be getting a lot of new WU's. But it does seem a bit coincidental. Can anyone explain?

---

hedera It sounds like you had the same problem that I've been seeing and reported here. BOINC loses contact and doesn't display anything. You must have at least seen a project list though to try and update, so your symptoms are a little different then mine.

---

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/

Phillip, it really depends upon the timeframe you are looking at. There were some really LONG running models out there with some of the new work on docking and fibrils. Those tasks had huge varience in credit per model. And part of the reason was that the models could take soooo looong. And part of the reason was that the new science code they built in to Rosetta had a sort of quick kill switch. It would work a given model about half way and decide if it was reasonable to work it to completion or if we should cut out losses here and start on a new one. The later release, I don't see it mentioned in the release notes, improved the runtime on these tasks. I recall seeing about post saying 3x improvement. This minimized the varience somewhat.

Bottom line, I believe you are correct, we're now in to types of WUs with CAPRI and docking and fibrils which have a higher varience in credit per model. Or, should I say, a higher varience in how long it takes to crunch a given model to earn the credit?

---

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/

Still having graphics problems with 5.43, I'm afraid. I started everything up this morning and put task graphics on, not in the screen saver, just a window. Came back a couple of hours later, the graphics window is gone, and I had this error:

12/17/2006 9:45:17 AM|rosetta@home|Unrecoverable error for result 5croA_BOINC_POSE_ABRELAX_NEWRELAXFLAGS_frags83__1449_434_0 ( - exit code -1073741819 (0xc0000005))

This was about 35 minutes after the task had picked up again on startup. My Windows event log had no errors at all.

Feet1st, my earlier error didn't quite match yours as I showed nothing whatever in the BOINC Manager, not even a project to update. Every tab in the Manager was blank, and when I tried to connect to Rosetta with my password, using the Advanced dropdown, a just got "unable to connect".

---

--hedera

Never be afraid to try something new. Remember that amateurs built the ark. Professionals built the Titanic.

I was watching the graphics screen, but walked away and when I came back I noticed this error in my messsages log. This is the first error I have ever received running Rosette.

12/17/2006 2:48:49 PM|rosetta@home|Unrecoverable error for result 1tig__BOINC_POSE_ABRELAX_VARY_SC_BOND_ANGLES_NEWRELAXFLAGS_frags83__1451_898_0 ( - exit code 1073807364 (0x40010004))

---

Feet1st, my earlier error didn't quite match yours as I showed nothing whatever in the BOINC Manager, not even a project to update. Every tab in the Manager was blank...

Actually that is EXACTLY what happens to me (and others). I see, since everything was blank you tried to attach to the project again. Seems to be a BOINC problem. We're discussing it more in the thread I linked, as it does not seem to relate to a specific Rosetta version.

---

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/

I moved daniels' post here, as he's got Linux problems with 5.43 as well as having seen them on 5.41.

process exited with code 131 (0x83)
process got signal 11

---

I moved daniels' post here, as he's got Linux problems with 5.43 as well as having seen them on 5.41.

process exited with code 131 (0x83)
process got signal 11

I am seeing the same thing on both of my Linux machines.
I can make the WU finish by shutting down the computer and restarting it.
I tried to suspend the project and let WCG run for a while then resume the
Rosetta WU but that didn't work.
The Rosetta projects get to about 58.71% done then just set there for hours doing nothing.
I did see this before "once" that I remember with 5.41 but 5.43 has become a brick
wall. Both machines have 256 meg of RAM. One has Linspire Linux and the other has Mandriva 2007.

https://boinc.bakerlab.org/rosetta/result.php?resultid=51996472
https://boinc.bakerlab.org/rosetta/result.php?resultid=51738512

Jim

Hi,
I can't get new projects since yesterday.

All I get in the reports is :

18.12.2006 19:37:53|rosetta@home|Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
18.12.2006 19:37:53|rosetta@home|Reason: Requested by user
18.12.2006 19:37:53|rosetta@home|(not requesting new work or reporting completed tasks)
18.12.2006 19:37:58|rosetta@home|Scheduler request succeeded

Can anyone tell me what's wrong? I didn't change anything any my other projects run just fine.

Thanx.