1311 Crash

Case number:845829-2003127
Topic:Crash/Hang
Opened by:TomTaylor
Status:Open
Type:Bug
Opened on:Wednesday, November 30, 2016 - 01:47
Last modified:Thursday, December 8, 2016 - 13:07

Crashed while using Wiggle.

AttachmentSize
log.txt224.92 KB
(Wed, 11/30/2016 - 01:47  |  40 comments)


Joined: 10/23/2014
Groups: Contenders

Crash #3 using Wiggle

Joined: 04/24/2014
Groups: None

Replication steps from Bendbob:
Start puzzle.
Freeze all Sheets.
Right click Helix and select Idealize Sidechains
Press keyboard "q" to center protein.
Zoom out using the mouse wheel which will eventually reveal the 2nd piece of the puzzle.
Unfreeze the 2nd piece.
Begin to pull the protein using the left mouse button.
=> As soon as the 2nd piece begins to move the game crashes.

Joined: 05/17/2008
Groups: None

From Log.Txt

***** STARTING THREAD ActionRepack
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0

UNHANDLED EXCEPTION
1: RaiseException +88 bytes (no line)
2: alIsAuxiliaryEffectSlot +295880 bytes (no line)
3: alIsAuxiliaryEffectSlot +172379 bytes (no line)
4: alIsAuxiliaryEffectSlot +165471 bytes (no line)
5: library_main +12199 bytes (no line)
6: library_main +9845 bytes (no line)
7: library_main +13470 bytes (no line)
8: library_main +22713 bytes (no line)
9: library_main +36111428 bytes (no line)
10: library_main +15296 bytes (no line)
11: library_main +3039 bytes (no line)
12: library_main +35165470 bytes (no line)
13: library_main +4883072 bytes (no line)
14: library_main +4881442 bytes (no line)
15: library_main +3195901 bytes (no line)
16: library_main +1863967 bytes (no line)
17: library_main +213199 bytes (no line)
18: library_main +3274838 bytes (no line)
19: library_main +4875769 bytes (no line)
20: library_main +3033413 bytes (no line)
21: library_main +16749 bytes (no line)
22: library_main +21 bytes (no line)
23: no symbol (no line)
24: no symbol (no line)
25: no symbol (no line)
26: BaseThreadInitThunk +18 bytes (no line)
27: RtlInitializeExceptionChain +99 bytes (no line)
28: RtlInitializeExceptionChain +54 bytes (no line)

Joined: 05/19/2009
Groups: Contenders

The REPOSTED version crashes repeatedly when just shaking from start.
Please also call the current repost 1311b, and subsequnt ones c and d to avoid confusion in the bug reports.

***** STARTING THREAD ActionRepack
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0
IRC::run error, code = 5, desc = Remote connection closed
Exiting IRC::run

Joined: 01/12/2015
Groups: None

Hi Bletchly,

Are you running OSX or Windows?

Joined: 05/19/2009
Groups: Contenders

I am running the last Windows client

Joined: 09/24/2012
Groups: Go Science

Shake (in recipe) crashes on Mac OS as well.

Joined: 01/12/2015
Groups: None

Thanks for the reports. I will try and replicate the errors.

-free_radical

alcor29's picture
User offline. Last seen 11 hours 8 min ago. Offline
Joined: 11/16/2012

1311 crash. Crashed when using zoom. Crashed again after posting a msg in chat box and lost all of my laborious work trying to line up sheets in a hexamer environment. 1st log was weird; saw nothing. Attaching whole second log. It too is unusual. Well could not upload word doc to foldit. Should have uploaded file from OS. Oh well. Don't if can start over again today. For what it's worth here is a copy of the last lines of the log.

core.pose.util: WARNING: conf2pdb_chain(): PDBInfo does not exist, returning default map 1=A, 2=B, ...
Autosave puzzles/0002003148/0000462374/default/autosave-recentbest.ir_solution loaded.
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0
core.pose.util: WARNING: conf2pdb_chain(): PDBInfo does not exist, returning default map 1=A, 2=B, ...
Autosave puzzles/0002003148/0000462374/default/autosave-creditbest.ir_solution loaded.
building library...done.
clear density...
ScorePanel new session
GROUP CHAT CHANNEL: AnthropicDreams
BOINC says connect to: irc.fold.it:6667
Using nick: alcor29
Ended Initialization of GameApplication
SRVR_THRD getting notifications...
Entering IRC::run
Sending SOPs:

Notice : This nickname is registered and protected. If it is your
Notice : nick, type /msg NickServ IDENTIFY ¬password¬. Otherwise,
Notice : please choose a different nick.
Notice : Password accepted - you are now recognized.
IDENTIFIED
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
***** STARTING THREAD ActionRepack
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0
Sending SOPs:

alcor29's picture
User offline. Last seen 11 hours 8 min ago. Offline
Joined: 11/16/2012

PS. Looking at the log (though am not a coder) it seems that maybe my in client chat had something to do with the wipeout crash.

Joined: 01/12/2015
Groups: None

That really sucks alcor29. I am sorry that it crashed. I am playing around with the puzzle as we speak, but I havent had it crash on me. I am doing tests on both OSX and Windows. If I find an answer I will be sure to post it as soon as possible!

mimi's picture
User offline. Last seen 28 weeks 18 hours ago. Offline
Joined: 11/17/2008
Groups: Contenders

If you just open it and shake it will crash - just tried

Joined: 10/23/2014
Groups: Contenders

1311 fixed version - Crashed while using Wiggle.

Joined: 01/12/2015
Groups: None

Hi Everyone,

Thank you for the crash reports. I was able to reproduce the wiggle crash and I have a working theory as how to fix things.

1) The crash happens only for me if I use an old binary for Foldit. After updating Foldit (letting it update in the client), hitting wiggle at the start of the puzzle does not result in a crash.

2) If it does crash, I deleted all the old puzzle files in my directory and that seemed to stop it from crashing.

With this being said, I am not sure what the actual root of the problem is. I havent had it crash on me yet running recipes or doing hand movements. I will continue trying to get it to crash to find the bug. Thank you for the patience.

free_radical

Joined: 10/23/2014
Groups: Contenders

I was able to crash resetting the puzzle and only using Pull Mode->Wiggle. Unfortunately two times it crashed and two times it didn't crash.

1. Reset puzzle
2. Pull Mode->Wiggle segments 12-18
3. Wait a few minutes

Joined: 10/23/2014
Groups: Contenders

Did some more testing. Crashed 8 out of 10 times. Latest crashes are more consistent taking less than 10 seconds to crash.

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

People who are experiencing crashes, try setting your undo graph properties to something less than 100. I have graph length 80, memory 50%, and I rarely crash on very large puzzles that seem to cause many crashes for others. Just something to try.

Joined: 10/23/2014
Groups: Contenders

Tried Susume's settings and the first time when I started Wiggle foldit immediately shut down. No error message just gone. The second time crashed after starting Wiggle. The good news is there is now something in the error log.

alcor29's picture
User offline. Last seen 11 hours 8 min ago. Offline
Joined: 11/16/2012

Crashed using pull tool this time. Log below

No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
No object matching key 0.
SRVR_THRD getting notifications...
Sending SOPs:

delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
***** STARTING THREAD ActionLocalMinimizePull
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGS: [master 12] 13:0 14:0 15:0 16:0 17:0
core.conformation.symmetry.SymmData: Setting weight of master jump ( jump-id=1 ) to 1.0 (was undefined)
core.conformation.symmetry.SymmData: Setting jump_group JGSS: [master 1] 2:0 3:0 4:0 5:0 6:0
delta_score: 0
Playing sound: 4
***** ENDING THREAD ActionLocalMinimizePull
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
Sending SOPs:

delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
SRVR_THRD getting notifications...
Sending SOPs:

delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
***** STARTING THREAD ActionLocalMinimizePull
Tool on_action_complete called
Tool on_action_complete called
***** STARTING THREAD ActionLocalMinimizePull
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
rotate mouse down
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
Sending SOPs:

delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
delta_score: 0
Playing sound: 4
Tool on_action_complete called
Tool on_action_complete called
SRVR_THRD getting notifications...
Sending SOPs:

***** STARTING THREAD ActionLocalMinimizePull

Joined: 08/06/2010
Groups: Contenders

I don't know the code but I have a theory (actually three)...

The full hexamer has a ton of graphical objects on it that a normal puzzle won't have (5 extra copies of every motion). That's 6x more flow into the rendering engine. If the CPU/GPU combination can't keep pace and render all of that within a certain frame rate, how does Foldit respond? Are you properly catching frame buffer underflow messages?

If you miss a number of render frames in a row, what is the code designed to do? If the routines used in pumping those objects to the graphics library take too long, are there other timer-based threads (like chat) that freak out because they missed their normal read cycle?

And theory #3: if the manipulation routines (like shake and wiggle) are working faster than the graphics engine can render, are you overflowing some buffer somewhere? Memory block exhaustion, stack exhaustion, or simple overflows have been the bane of many programs. Foldit may not be an exception.

I know from my day job that the thread that "crashed" isn't always the thread that caused the crash. Other threads could have created the condition that simply didn't let that other thread operate properly.

Joined: 09/24/2012
Groups: Go Science

I don't know if it's linked, but:

-applying susume option (minimal graph properties) and minimizing the windows and only using recipes without shake, the following might happen:

-recipe can last the all night without crashing on one client
-using more clients with this same puzzle, other clients might crash after some time
-viewing a non crashed client after one night, and doing something on screen might crash this client (like if, with the time, the client looses stability).

This remembers me that a huge puzzle can randomly crash without notice after some time: like this 610 residues one: http://fold.it/portal/node/996825

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

Crashed during script - log.txt says it was during ActionBuild, scriptlog says it was during Remix.
Event viewer says exception code 0xe06d7363 in KERNELBASE.dll.

Joined: 08/06/2010
Groups: Contenders

0xc0000005 is an access violation. It means the program was trying to use an area of memory that the operating system had not let the program own yet. Either you ran past the end of a buffer (touching memory you did not own) or you overwrote the area of memory where another variable you did own lived and made it point to some random location you didn't own.

Either way, it's related to memory work and is an extremely common error in most C-based programs as it is relatively easy to
* mess up pointer arithmetic
* not check the sizes of your buffers against the content you are pumping into them

And there are WAY more than just the 2 ways I listed to hit this error. But based on what I think Foldit is doing internally, those seem like the first causes I would try to rule out.

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

Crashed during different script - log.txt says during ActionRebuild, scriptlog also says during Rebuild.
Event viewer says exception code 0xc0000005 in game_library.dll.

Joined: 08/06/2010
Groups: Contenders

This code is what you get when your C++ program raises an exception that isn't caught by your own program's exception handling routines.

https://blogs.msdn.microsoft.com/oldnewthing/20100730-00/?p=13273

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

And again during ActionRebuild.
Event viewer says exception code e06d7363 in KERNELBASE.dll.

Joined: 05/19/2009
Groups: Contenders

If you check my snippet of log, you can see the phrase 'Setting weight of master jump ( jump-id=12 ) to 1.0 (was undefined)'. That should provide a clue as to where the issue comes from. If that weight is in a structure that needs initialization, you may also have pointers in there causing memory exceptions (which I have since also seen).

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

And again during ActionRebuild (or possibly Sending SOPs)
Event viewer: exception code c0000005 in StackHash_0a9e

actiasluna's picture
User offline. Last seen 11 hours 46 min ago. Offline
Joined: 03/05/2015
Groups: Gargleblasters

Fixed version crash: Started running recipe, then poof. FYI I have memory set to 10% for undo graph,

actiasluna's picture
User offline. Last seen 11 hours 46 min ago. Offline
Joined: 03/05/2015
Groups: Gargleblasters

Crashed at least 4x more since. Shame that this can't be played properly. I vote that this version of the puzzle is removed and reintroduced after the tools that are broken are really fixed.

Joined: 09/24/2012
Groups: Go Science

Crashes mainly when dooing something on screen (even mouse over).

With window minimized (and recipe), I didn't experience a crash yet.

Joined: 09/24/2012
Groups: Go Science

Crash during recipes: only on shake all (not on shake selected)

Joined: 10/23/2014
Groups: Contenders

Crash using Wiggle

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

It looks like the last symmetry science puzzle was 1191, in February. Since then, in March/April, the Symmetry Design beginner puzzle was crashing repeatedly. It is possible a bug was created in symmetry in that time frame and has never been fixed.

http://fold.it/portal/node/2002237 Beginner Puzzle "Symmetry Design" crashing on different machines

Joined: 11/16/2012

I would add that as recently as this October, the Symmetry Design Beginner Puzzle was crashing repeatedly. Any time a shake was invoked manually or from within a script, it would crash. Very similar to the symptoms I am experiencing with 1311.

Joined: 09/24/2012
Groups: Go Science

Indeed. No crash with this recipe that uses only selected shakes (no shake all):
http://fold.it/portal/recipe/102015

Susume's picture
User offline. Last seen 3 days 16 hours ago. Offline
Joined: 10/02/2011

I decided to try an old symmetry puzzle (1150) that did not have any complaints of crashes in puzzle comments or feedback when it was open. It crashes reliably during Rebuild within a minute of starting the enhanced DRW script. So it does appear that symmetry in general is broken, not just puzzle 1311.

I have shared with scientists the solution I tested with, named "1150 Crashing symmetry Dec 2016"
Script is TvdL enhanced DRW 2.8.0, accept all default settings.

Crash 1: during Rebuild
Exception code: 0xc0000374
Faulting module path: C:\Windows\AppPatch\AcXtrnal.DLL
log - Copy (30).txt (note that another client was running; crash was during Rebuild but is not at end of log)

Crash 2: during Rebuild
Exception code: 0xc0000374
Faulting module path: C:\Windows\AppPatch\AcXtrnal.DLL
log - Copy (31).txt

Crash 3: during Rebuild
log - Copy (32).txt

Crash 4: during Rebuild
Exception code: 0xc0000374
Faulting module path: C:\Windows\AppPatch\AcXtrnal.DLL
log - Copy (33).txt

Crash 5: during Rebuild
Exception code: 0xc0000374
Faulting module path: C:\Windows\SysWOW64\ntdll.dll
log - Copy (34).txt

Crash 6: during Rebuild
Exception code: 0xc0000005
Faulting module path: C:\Windows\SysWOW64\ntdll.dll
log - Copy (35).txt (this one has an Unhandled Exception in it)

actiasluna's picture
User offline. Last seen 11 hours 46 min ago. Offline
Joined: 03/05/2015
Groups: Gargleblasters

Latest crash ... immediately upon clicking wiggle after closing cuts. Lots of "can't find xxx" in the log.txt, I think that might be related to the game_dynamic issue which obviously doesn't help when symmetry is also broken. Log txt and mac crash log attached.
FYI I am running a fresh client (with the options.txt and all.macro copied into the correct - Resources - folder) last night in an attempt to clear the slate so troubleshooting 1311 would be "cleaner"... it did not help, first try on 1311 crashed on shake (no log on this unless it shows up in the log.txt attached here). All old foldit files are "quarantined" (I had to do this back in August because it was finding any client it could to run, from my OLD foldit backup folders, apparently. After game_dynamic started showing up it didn't know where to look or so it seems.)

actiasluna's picture
User offline. Last seen 11 hours 46 min ago. Offline
Joined: 03/05/2015
Groups: Gargleblasters

Another crash but accidentally overwrote the log.txt file... on shake after a couple minutes of tenuous working. Caught the mac crashlog though so it's attached. (now I go back to working on the one that I can actually play for more than a couple minutes).

actiasluna's picture
User offline. Last seen 11 hours 46 min ago. Offline
Joined: 03/05/2015
Groups: Gargleblasters

Just one more crashlog for the experts, trying to run DRWremix on the expired puzzle to refine an interesting solution.

log.txt won't help as I'm running other clients so mac crashlog and scriptlog attached.

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons