A log of 170 Go !

Case number:699969-2010088
Topic:General
Opened by:Bruno Kestemont
Status:Open
Type:Bug
Opened on:Thursday, July 30, 2020 - 05:56
Last modified:Saturday, October 24, 2020 - 14:29

My C drive was full because of this !

=====================
Version:
ff622263f4d003fd3cb579c9f521e423
on Windows

Debug.txt:

version: 1
{
"backtrace" : " 1: no symbol (no line)\n 2: no symbol (no line)\n 3: no symbol (no line)\n 4: no symbol (no line)\n 5: no symbol (no line)\n 6: no symbol (no line)\n 7: no symbol (no line)\n 8: no symbol (no line)\n 9: no symbol (no line)\n"
"build_id" : "20200625-6f685ae5d0-win_x86"
"crash_id" : "5a62c4eb-fbd7-4ef6-af16-482e71f2d6db"
"current_actions" : "MacroScriptRunnerAction\,ActionLocalMinimize"
"current_tool" : "Pull Mode"
"file" : "interactive/util/Thread.cc:300"
"machine_id" : "6d5108b9-fc6d-4e3f-85ef-80f84614c9de"
"macro_id" : "0"
"message" : "Unhandled exception: Can't start thread: °NÌÙ"
"player_id" : "447652"
"puzzle_id" : "2010041"
"puzzle_running_time" : "14015"
"running_time" : "127577"
"timestamp" : "1596057074"
}
verify: 115a8672ed1a0ef41a352b804979fd36

(Thu, 07/30/2020 - 05:56  |  7 comments)


bkoep's picture
User offline. Last seen 21 hours 19 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

Wow, that's a big log file! There are several different problems contributing to excessive log messages, and I think the next release will have a fixes for a couple of these, which should help.

Joined: 09/24/2012
Groups: Go Science

Record 276 Go log (3 clients crashed).

Version (updated): 13b4a7177bdfe0751a5380d1a24dc136
on Windows 7.

Version database: 79ebb0c96fcc1973fc3f9f37c0e14705
Version ressources: 76dfb4c1b3dd003d6606314b7ecd98dc

This happens when the clients are on the same puzzle. Here 1870 trimer.

bkoep's picture
User offline. Last seen 21 hours 19 min ago. Offline
Joined: 11/15/2012
Groups: Foldit Staff

Thanks Bruno. Is this a devprev client or the main client? Can you post the build ID?

(Sorry, I can't tell the build ID from these version numbers.)

Joined: 09/24/2012
Groups: Go Science

A log of 332 Gb (disk full error)

Windows 7, MAIN

1cefe0fe933f2a6ccef256615d422c0d

428fc0948d0f404d685c4e477f984796

214153bebec0533121c44416c1fa1d1b

Binary:

cmp-binary-1cefe0fe933f2a6ccef256615d422c0d

It happens when I run several clients on the same puzzle, using Lua command saving and loading from disk. This provoques freezes of almost all clients.

Joined: 09/24/2012
Groups: Go Science

Very long lists of "spaces" seem to be the main cause of inflating logs. See:

https://fold.it/portal/node/2010241#comment-42703

Joined: 09/24/2012
Groups: Go Science

Record equalized: 335 Mb, disk full. (on up to date devpev).
This happens when at least one client is on Medium wiggle power on end game (may be when I use the lua command save to disk in recipes). Here I had 2 clients on devprev. I think the recipe Constructor was used (it does a lot of different thinks => unable to know what action provoqued such a long log).

jeff101's picture
User offline. Last seen 1 day 15 hours ago. Offline
Joined: 04/20/2012
Groups: Go Science
I had similar problems on Puzzle 1905, but the 
log.txt files it made were only about 2-25 Gbytes. 
Nevertheless, such files were large enough to give 
unresponsive Foldit clients that I chose to quit 
and restart.

My hunch is that the large log.txt files occurred 
when Recipes tried to execute LUA commands like:
save.LoadSolution or save.LoadSolutionByName
as described at:
https://foldit.fandom.com/wiki/Foldit_Lua_Functions#save.LoadSolution
Loading the same solution files manually caused
no such problems.

The same Recipes have worked before on other Foldit
Puzzles like 1829 and 1847. Puzzle 1905 contained 
several cuts, like the ones between segments 34-35, 
64-65, 103-104, and 113-114. Each of these cuts gave 
a group of 3 lines in log.txt like below:

core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 34-1, 34-2, 34-3, 35-1!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 34-2, 34-3, 35-1, 35-2!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 34-3, 35-1, 35-2, 35-3!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 64-1, 64-2, 64-3, 65-1!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 64-2, 64-3, 65-1, 65-2!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 64-3, 65-1, 65-2, 65-3!
...
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 103-1, 103-2, 103-3, 104-1!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 103-2, 103-3, 104-1, 104-2!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 103-3, 104-1, 104-2, 104-3!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 113-1, 113-2, 113-3, 114-1!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 113-2, 113-3, 114-1, 114-2!
core.kinematics.AtomTree: [ ERROR ] No proper DoF can be found for these four atoms: 113-3, 114-1, 114-2, 114-3!

Such groups of lines occurred repeatedly in the 
log.txt file, making it grow quickly in size.

I'm guessing that each line above is for a different
dihedral angle among backbone atoms 1-3 (1=nitrogen, 
2=alpha-carbon, and 3=carbon on the C=O group). Maybe 
2 of these dihedrals are on the Ramachandran Map, and
the 3rd distinguishes cis from trans peptide bonds.
When there are cuts, these 3 dihedral angles become
ill-defined, giving log.txt messages like above.

If you can track down this bug and fix it, I would
appreciate it.
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