.wait() method should report if acemd is stuck about htmd

[stefdoerr]

Can you check in the directory which it's simulating what is written in the log.txt files and the FINISH_TIME file?

[stefdoerr]

I've had ACEMD get stuck in an infinite minimization loop before. To debug it actually try running acemd directly from the simulation folder (from command line). Then you will see what's happening.

[giadefa]

maybe the wait() could give more info also

On 9 June 2016 at 13:42, Stefan [email protected] wrote:

I've had ACEMD get stuck in an infinite minimization loop before. To debug
it actually try running acemd directly from the simulation folder (from
command line). Then you will see what's happening.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#56 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/AHaqOuWyDgJDyqhXrcT4EHtwhn_R338Wks5qJ_w2gaJpZM4Ix3Zi
.

http://www.acellera.com

   <https://twitter.com/acellera>

https://www.youtube.com/user/acelleracom
https://www.linkedin.com/company/2133167?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A2133167%2Cidx%3A2-1-2%2CtarId%3A1448018583204%2Ctas%3Aacellera
https://www.acellera.com/md-simulation-blog-news/
http://is.gd/1eXkbS

[stefdoerr]

Like what info? Projected runtime maybe? But it would still not help in debugging this I think. Only looking at the acemd realtime output helps for such problems and printing that would be impossible (and messy).

[nesilin]

Hi!
For your knowledge , in fact the tutorial barnase-barstar-building-simulation of htmd which also uses ACEMD like

mdx = AcemdLocal()
mdx.submit('./equil')
mdx.wait()

also got stuck at this point.
A part from that, we have tried what you suggested but the FINISH_TIME file does not exist and the log.txt does not say anything about trouble in the wait method.
Moreover, the log.txt did not report any problem.
We have tried ACEMD with the input file inside the /docked/equil/1 directory and we have obtained the output.coor and output.xsc and others.
Could it be that we are missing the driver of nvidia? Because we have tried to run ACEMD from command line in another machine and as I said it worked but in alejandrovr 's computer did not. In alejandrovr 's computer the error has been the following:

 /home/alejandro/miniconda3/bin/acemd.bin:
 error while loading shared libraries: libcuda.so.1: 
cannot open shared object file: No such file or directory

[stefdoerr]

Hm ok that error means you don't have cuda installed on the machine. ACEMD needs cuda and an NVIDIA GPU to work.

So the mdx.wait() problem was only on the machine of @alejandrovr ?

[j3mdamas]

yeah, gianni said that some of their machines wouldn't be able to run acemd...

[stefdoerr]

Would be nice though if AcemdLocal did not get stuck and reported something. But I need a test machine to see what goes wrong.

[giadefa]

workspace1

On 9 June 2016 at 16:06, Stefan [email protected] wrote:

Would be nice though if AcemdLocal did not get stuck and reported
something. But I need a test machine to see what goes wrong.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#56 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/AHaqOnwZIJgcyPtfjEgk4ySQ7bxvx42rks5qKB39gaJpZM4Ix3Zi
.

http://www.acellera.com

   <https://twitter.com/acellera>

https://www.youtube.com/user/acelleracom
https://www.linkedin.com/company/2133167?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A2133167%2Cidx%3A2-1-2%2CtarId%3A1448018583204%2Ctas%3Aacellera
https://www.acellera.com/md-simulation-blog-news/
http://is.gd/1eXkbS

[j3mdamas]

So, this is issue is about changing AcemdLocal (anything that runs acemd really), to report stuff if stuck.

For the students, well, there are large groups, I hope that there is at least one person per group which can run acemd

[giadefa]

only 4 people need to run

On 9 June 2016 at 16:16, João M. Damas [email protected] wrote:

So, this is issue is about changing AcemdLocal (anything that runs acemd
really), to report stuff if stuck.

For the students, well, there are large groups, I hope that there is at
least one person per group which can run acemd

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#56 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/AHaqOhc-RWF0j94vPIOpTQrkafoN3DFYks5qKCBMgaJpZM4Ix3Zi
.

http://www.acellera.com

   <https://twitter.com/acellera>

https://www.youtube.com/user/acelleracom
https://www.linkedin.com/company/2133167?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A2133167%2Cidx%3A2-1-2%2CtarId%3A1448018583204%2Ctas%3Aacellera
https://www.acellera.com/md-simulation-blog-news/
http://is.gd/1eXkbS

[alejandrovr]

@stefdoerr it did not work for me, neither for my colleagues who does not have nvidia. Thanks for answering, running from command line has helped us.

[giadefa]

Why are you not running on the provided machines with Nvidia GPUs?

On 10 June 2016 at 11:12, alejandrovr [email protected] wrote:

@stefdoerr https://github.com/stefdoerr it did not work for me, neither
for my colleagues who does not have nvidia. Thanks for answering, running
from command line has helped us.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#56 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/AHaqOgqBbPpbueP686NnmzbGNzjEalPuks5qKSp8gaJpZM4Ix3Zi
.

http://www.acellera.com

   <https://twitter.com/acellera>

https://www.youtube.com/user/acelleracom
https://www.linkedin.com/company/2133167?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A2133167%2Cidx%3A2-1-2%2CtarId%3A1448018583204%2Ctas%3Aacellera
https://www.acellera.com/md-simulation-blog-news/
http://is.gd/1eXkbS

[alejandrovr]

Because vmd was not installed. Now it is, however vmd is not fullly working, because it will only run if one does ssh without -Y option, but then no window will appear, only the command line from vmd. That seems enough to make the scripts that use vmd run.

[giadefa]

hmm, no script use VMD if not for visualization?

On 10 June 2016 at 11:24, alejandrovr [email protected] wrote:

Because vmd was not installed. Now it is, however vmd is not fullly
working, because it will only run if one does ssh without -Y option, but
then no window will appear, only the command line from vmd. That seems
enough to make the scripts that use vmd run.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#56 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/AHaqOhuoFHWBT0a-Qp3TntPhj4IhuRO0ks5qKS1YgaJpZM4Ix3Zi
.

http://www.acellera.com

   <https://twitter.com/acellera>

https://www.youtube.com/user/acelleracom
https://www.linkedin.com/company/2133167?trk=tyah&trkInfo=clickedVertical%3Acompany%2CclickedEntityId%3A2133167%2Cidx%3A2-1-2%2CtarId%3A1448018583204%2Ctas%3Aacellera
https://www.acellera.com/md-simulation-blog-news/
http://is.gd/1eXkbS

[alejandrovr]

We found this piece of code that saves a view into a variable:

from ipywidgets.widgets import Box
w = []
for i, m in enumerate(molbuilt):
m.view(sel='protein', style='NewCartoon', hold=True)
m.view(sel='water', style='Lines', hold=True)
h = m.view(sel='resname MOL', style='Licorice', color=0)
w.append(h)
b = Box(children=(w[0],w[1]))
b
Now our script fails at Box, but we think that's some kind of Jupyter stuff not really needed for the simulation process so we are trying to run the script with it. Is that correct? Thanks.

PD: We have eliminated Box command and it has worked fine.

[stefdoerr]

If there is still an issue please report it. Closing for now

[mj-harvey]

What I take from this is that there's a problem with ACEMD sometimes getting stuck minimizing ill-conditioned input. There's a case about that open elsewhere, so I'm oging to close this one. If you think there's another problem, please reopen and elaborate.