Bug Report for Ralph 5.28

Message boards : RALPH@home bug list : Bug Report for Ralph 5.28

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Chu
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 26 Sep 06
Posts: 61
Credit: 12,545
RAC: 0
Message 2305 - Posted: 3 Oct 2006, 0:26:02 UTC
Last modified: 3 Oct 2006, 0:21:05 UTC

Ralph has been updated to 5.28. In this version, we added all-atom sidechain display to the screensaver during Rosetta sidechain refinement stage. Also users now can rotate any of the four displayed model on the screen using a mouse. Several new flags have been added to Rosetta too.
ID: 2305 · Report as offensive    Reply Quote
Profile anders n

Send message
Joined: 16 Feb 06
Posts: 166
Credit: 131,419
RAC: 0
Message 2306 - Posted: 3 Oct 2006, 5:18:26 UTC
Last modified: 3 Oct 2006, 5:05:50 UTC

ID: 2306 · Report as offensive    Reply Quote
idlorj

Send message
Joined: 16 Feb 06
Posts: 2
Credit: 20,030
RAC: 0
Message 2308 - Posted: 3 Oct 2006, 18:04:53 UTC

I recieved the same error

https://ralph.bakerlab.org/result.php?resultid=275739

Of the ones completed only this workunit failed for me.
ID: 2308 · Report as offensive    Reply Quote
Buffalo Bill

Send message
Joined: 13 May 06
Posts: 5
Credit: 12,223
RAC: 0
Message 2311 - Posted: 5 Oct 2006, 3:49:32 UTC

Errors for these:

277562
277561
ID: 2311 · Report as offensive    Reply Quote
Uakey

Send message
Joined: 8 Sep 06
Posts: 1
Credit: 141
RAC: 0
Message 2312 - Posted: 5 Oct 2006, 7:47:18 UTC
Last modified: 5 Oct 2006, 7:36:38 UTC

ID: 2312 · Report as offensive    Reply Quote
Profile anders n

Send message
Joined: 16 Feb 06
Posts: 166
Credit: 131,419
RAC: 0
Message 2313 - Posted: 5 Oct 2006, 8:11:30 UTC - in response to Message 2312.  

ID: 2313 · Report as offensive    Reply Quote
Profile Krzychu P.

Send message
Joined: 16 Feb 06
Posts: 19
Credit: 25,687
RAC: 0
Message 2314 - Posted: 5 Oct 2006, 8:45:35 UTC

I get in message tab:
2006-10-05 10:44:52|ralph@home|Unrecoverable error for result DOCK_1BQL_unbound_perturb_benchmark_1330_19_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

and with next WU:

2006-10-05 10:50:53|ralph@home|Unrecoverable error for result DOCK_1JHL_unbound_perturb_benchmark_1330_19_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

Niepoprawna funkcja = incorrect function
ID: 2314 · Report as offensive    Reply Quote
Profile Krzychu P.

Send message
Joined: 16 Feb 06
Posts: 19
Credit: 25,687
RAC: 0
Message 2315 - Posted: 5 Oct 2006, 9:57:41 UTC

Next two WU's with the same:

2006-10-05 10:58:45|ralph@home|Unrecoverable error for result DOCK_1FQ1_unbound_perturb_benchmark_1330_19_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

2006-10-05 11:04:21|ralph@home|Unrecoverable error for result DOCK_1MEL_unbound_perturb_benchmark_1330_20_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

ID: 2315 · Report as offensive    Reply Quote
Profile feet1st

Send message
Joined: 7 Mar 06
Posts: 313
Credit: 116,623
RAC: 0
Message 2316 - Posted: 5 Oct 2006, 16:36:23 UTC

While displaying graphics for this WU, the first box seems stuck with status message "Try Rotamers", and the graphic in that first box is not refreshing. The Accepted box is jumping around, but no movement in first box.

...while I was typing the above, it went to "Packing" status (step 340,000) and now it seems to be progressing normally and refreshing the graphic.
ID: 2316 · Report as offensive    Reply Quote
Chu
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 26 Sep 06
Posts: 61
Credit: 12,545
RAC: 0
Message 2317 - Posted: 5 Oct 2006, 16:54:58 UTC - in response to Message 2315.  

Thanks. This will be very useful for us to track down those hidden bugs in the code.
Next two WU's with the same:

2006-10-05 10:58:45|ralph@home|Unrecoverable error for result DOCK_1FQ1_unbound_perturb_benchmark_1330_19_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

2006-10-05 11:04:21|ralph@home|Unrecoverable error for result DOCK_1MEL_unbound_perturb_benchmark_1330_20_0 (Niepoprawna funkcja. (0x1) - exit code 1 (0x1))

ID: 2317 · Report as offensive    Reply Quote
Chu
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 26 Sep 06
Posts: 61
Credit: 12,545
RAC: 0
Message 2318 - Posted: 5 Oct 2006, 17:20:45 UTC - in response to Message 2316.  

The current setup in the screensaver graphic is to flash protein sidechains only in "Try Rotamer" and "Packing" stage. "Try Rotamers" is supposed to be very fast each time, and we called them a lot in the code. So people are expected to see protein sidechains flashed on and off most of the time. I think the trajectory is still moving forward as reflected by the moving protein in the accepted box. So it looks like a graphic issue. I am wondering if the more demanding task (showing sidechains on and off quickly) for the graphic is the reason for more frequent "Unhandle Exception" errors we have seen in this updated application. I tried to google what the error code(0x0000005 means when it happens, some people suggest a hardware issue, but no firm answers. Anybody has an idea?
While displaying graphics for this WU, the first box seems stuck with status message "Try Rotamers", and the graphic in that first box is not refreshing. The Accepted box is jumping around, but no movement in first box.

...while I was typing the above, it went to "Packing" status (step 340,000) and now it seems to be progressing normally and refreshing the graphic.

ID: 2318 · Report as offensive    Reply Quote
Pieface

Send message
Joined: 16 Feb 06
Posts: 64
Credit: 203,513
RAC: 0
Message 2320 - Posted: 5 Oct 2006, 19:54:33 UTC

We used to see them a lot before ?Rom? came in and worked thru the code.
I think that was about the time they set-up Ralph for testing. Maybe he can help? I know back then we had to keep Rosetta in-memory and not let it get swapped out to avoid problems also.
ID: 2320 · Report as offensive    Reply Quote
Profile anders n

Send message
Joined: 16 Feb 06
Posts: 166
Credit: 131,419
RAC: 0
Message 2321 - Posted: 5 Oct 2006, 20:02:23 UTC - in response to Message 2320.  
Last modified: 5 Oct 2006, 19:47:13 UTC

We used to see them a lot before ?Rom? came in and worked thru the code.
I think that was about the time they set-up Ralph for testing. Maybe he can help? I know back then we had to keep Rosetta in-memory and not let it get swapped out to avoid problems also.



I still have all my computers set to yes "leave in memory".

Anders n
ID: 2321 · Report as offensive    Reply Quote
Profile feet1st

Send message
Joined: 7 Mar 06
Posts: 313
Credit: 116,623
RAC: 0
Message 2323 - Posted: 6 Oct 2006, 1:48:36 UTC

I think a part of what we should be testing here is the description of what to expect from the new release and the new graphics. We've had problems before where intentional changes are made... not really described to people, tested on Ralph, rolled out to Rosetta and then people start posted in the "bugs with this release" thread, simply because they don't understand what is happening and that the change is intentional. They don't realize that the science is changing over time and that this is part of the point of the project. etc.

So... could you take a stab at describing the changes to the graphics? And the phases and behavior we should expect to observe? Dr. Baker mentioned in his journal that sidechains would be shown... but that raises more questions then it answers. Because this is all second nature to you, let me explain some questions I can see people will be having:

What are side chains?
Why do they blink on and off?
Why can't I see side chains in the other boxes? Mine must be broken.
Are we doing calculations differently now on the side chains? Or simply illustraiting the calculations we've been doing all along?

I feel these types of questions should be answered HERE on Ralph, so that we can comment on them and ask further questions... have the description revised and improved to address those questions and then use that to post the beginning of the "report problems with v x.yy in this thread", and to then summarize and post to the news on the homepage.
ID: 2323 · Report as offensive    Reply Quote
genes
Avatar

Send message
Joined: 16 Feb 06
Posts: 45
Credit: 43,300
RAC: 0
Message 2324 - Posted: 6 Oct 2006, 2:59:59 UTC - in response to Message 2318.  

I tried to google what the error code(0x0000005 means when it happens, some people suggest a hardware issue, but no firm answers. Anybody has an idea?


I looked at some of the error results that were posted, and I saw for the error code 0xffffffffc0000005, which, if you look at the lower 32 bits is 0xc0000005, which is an access violation. You know, like using a bad pointer, or trying to read or write to memory you don't have.

Hope that helps.
ID: 2324 · Report as offensive    Reply Quote
SafeAggie

Send message
Joined: 5 Oct 06
Posts: 6
Credit: 4,207
RAC: 0
Message 2325 - Posted: 6 Oct 2006, 12:13:22 UTC

Unrecoverable error for result

[list]
*1tit__BOINC_ABRELAX_SAVE_ALL_OUT_truess__1329_17_0 ( - exit code -1073741819 (0xc0000005))
*2vik__BOINC_ABRELAX_SAVE_ALL_OUT_truess__1329_18_0 ( - exit code -1073741819 (0xc0000005))
*2vik__BOINC_ABRELAX_SAVE_ALL_OUT_truess__1329_19_1 ( - exit code -1073741819 (0xc0000005))
[list]

resultid=279007
resultid=277677
resultid=277610
resultid=277591
resultid=277577
ID: 2325 · Report as offensive    Reply Quote
Profile Conan
Avatar

Send message
Joined: 16 Feb 06
Posts: 364
Credit: 1,368,421
RAC: 0
Message 2326 - Posted: 6 Oct 2006, 14:13:52 UTC

2Seems to be a Graphics problem. The graphics just freeze and will not release. The workunit appears to still be running (checked with Task Manager) but can't get rid of graphic screen, have to end process which causes a computation error. If I let it go I get this error
**********************************************************************
Rosetta score is stuck or going too long. Watchdog is ending the run!
Stuck at score 0.966268 for 3600 seconds
**********************************************************************
GZIP SILENT FILE: .xx1c9o.out

The stuck score changes for different failed workunits. Only happening on Windows machine as the Linux machine has no graphics.

So far the above error is on
https://ralph.bakerlab.org/result.php?resultid=279500
https://ralph.bakerlab.org/result.php?resultid=279505
https://ralph.bakerlab.org/result.php?resultid=247849

Also have had Unhandled Exception Record errors
Reason:Access Violation (0xc0000005) at address 0x00759D94 read attempt to address 0x00000011. Had this on https://ralph.bakerlab.org/result.php?resultid=247851
Reason:Access Violation (0xc0000005) at address 0x00759D7D read attempt to address 0x00000011. had this on
https://ralph.bakerlab.org/result.php?resultid=247850

Have had no successful WU's on this batch of Windows WU's.
Have extended the time for the screen saver to come on to 300 minutes from 20 to see if this stops the problem.
ID: 2326 · Report as offensive    Reply Quote
Chu
Volunteer moderator
Project developer
Project scientist

Send message
Joined: 26 Sep 06
Posts: 61
Credit: 12,545
RAC: 0
Message 2327 - Posted: 6 Oct 2006, 18:43:59 UTC

Thanks for the suggestion, feet1st. I will give it a try here.

What are side chains?
For general knowledge about protein and amino acid, please visit here. I notice that some key terms like "sidechains" are missing and we will update that soon. Here is something I wrote quickly and it definitely needs to be improved.
20 natural amino acids have part of their structures in common and they are connected via peptide-bond linkages to form the topology of a protein structure. The common part is called "backbone". Amino acids differ from each other by its unique chemical group which connects to the C-alpha atom of the backbone and that group is called "side chain".

Why do they blink on and off?
Sidechains are part of protein structures but we only choose to show them on screen during some certain stages when they are being changed by the program.

Why can't I see side chains in the other boxes? Mine must be broken.
So far we only enable the sidechain drawing for the "searching" box.

Are we doing calculations differently now on the side chains? Or simply illustraiting the calculations we've been doing all along?
Sidechain calculation has been part of Rosetta program for a long time, but we only start to have them illustrated on the screen recently.

I understand the above is still far from an adequate description. But I will try my best to provide more information on future updates and this will give people like you guys who volunteer to help testing a better chance to send us feedbacks. Many thanks again.
ID: 2327 · Report as offensive    Reply Quote
Profile Conan
Avatar

Send message
Joined: 16 Feb 06
Posts: 364
Credit: 1,368,421
RAC: 0
Message 2329 - Posted: 6 Oct 2006, 23:43:40 UTC

Well these new Ralph work units are impressive. After creating lots of problems such as freezing my computer when the graphics lock up, it now appears to have stuffed Boinc altogether. During the night Boinc closed down due to the graphics problem. I restarted Boinc which ran for a while then errored out and now Boinc won't run anymore due to Visual C++ runtime errors.
I am now trying to get Boinc running again by either replacing the MSVCP71.dll and MSVCP80.dll files in Boinc or I will just have to reload Boinc.
This is only affecting my Windows Xp machine with the latest updates.
ID: 2329 · Report as offensive    Reply Quote
Profile Conan
Avatar

Send message
Joined: 16 Feb 06
Posts: 364
Credit: 1,368,421
RAC: 0
Message 2330 - Posted: 7 Oct 2006, 1:56:54 UTC

I now have 2 problems. I can no longer get BOINC to run on my AMD 4800+ Windows XP machine. Problem started with the Ralph graphic freezes, now all I get is Visual C++ runtime errors when Boinc.exe attempts to run. I have re-installed Boinc and downloaded .net software for the VC libraries but still can't get Boinc to run anymore. Also rebooted a few times, no joy. Help please.

Problem 2, just had 20 WU's download to one of my Linux machines and all 20 Errored out in minutes of downloading with this error:-

ERROR:: Unable to determine sequence length from pdb file

https://ralph.bakerlab.org/workunit.php?wuid=247972, 247980, 247982, 247983, 247993, 247994, 247995, 247996, 247997, 247998, 247999, 248001, 248002, 248004, 248005, 248006, 248051, 248052, 248053, 248054
ID: 2330 · Report as offensive    Reply Quote
1 · 2 · Next

Message boards : RALPH@home bug list : Bug Report for Ralph 5.28



©2024 University of Washington
http://www.bakerlab.org