Hi Boris
Ran a single element model with the soil that i have in my full model. Is a von mises yield and potential surface and a Armstrong Frederick tensorial evolution law. Just doing a triaxial test on the single soil element.
Get the following errors:
For OpenSees version 1.6.1.c
*** glibc detected *** double free or corruption: 0x0000000000e2eb90 ***
Segmentation fault
For OpenSees version 1.6.2
Root must be bracketed in ZBRENTstress
Dont know if this gives you any idea of what is going wrong, but can see now that the problem in my model has something to do with the soil material models, and in particular the tensorial evolution law
Liam
OpenSees Crashes on Linux
Moderators: silvia, selimgunay, Moderators
zbrent
Hello Liam,
Ship me over the source script for your small model, I suspect there is a mix up with convention for positive stress (in tension). When you define initial state of stress as in:
# initial stress
set stressp "-0.00001 0 0 0 -0.00001 0 0 0 -0.00001"
note that they are with negative sign, hence in compression.
THe error in ZBRENT means that both the start stress and the elastic plastic trial stress are outside of yield surface...
So email me the source script at jeremic@ucdavis.edu and I'll take a look...
Boris
Ship me over the source script for your small model, I suspect there is a mix up with convention for positive stress (in tension). When you define initial state of stress as in:
# initial stress
set stressp "-0.00001 0 0 0 -0.00001 0 0 0 -0.00001"
note that they are with negative sign, hence in compression.
THe error in ZBRENT means that both the start stress and the elastic plastic trial stress are outside of yield surface...
So email me the source script at jeremic@ucdavis.edu and I'll take a look...
Boris
Hi Boris
I just changed my model in accordance with your advice and i do not get the error anymore, it runs fine through the small element model, and is now running fine on the full pile soil interaction model.
Of course it had to be something simple like that, so thanks for your advice and help.
Do you still want me to send the file?
Thanks again
Liam
I just changed my model in accordance with your advice and i do not get the error anymore, it runs fine through the small element model, and is now running fine on the full pile soil interaction model.
Of course it had to be something simple like that, so thanks for your advice and help.
Do you still want me to send the file?
Thanks again
Liam
Sorry Boris, i spoke too soon
Ran the full pile soil interaction model, and get the problem again. THe model runs fine during the application of gravity to the soil, but once the horizontal load is applied to the pile the model crashes after one time step.
Will play around with a few things to see if i can remedy it
Liam
Ran the full pile soil interaction model, and get the problem again. THe model runs fine during the application of gravity to the soil, but once the horizontal load is applied to the pile the model crashes after one time step.
Will play around with a few things to see if i can remedy it
Liam
senbd me files for inspection
Hello Liam,
Send me files for inspection, it works just fine for us (I'll show some of that stuff at the workshop...).
Boris
Send me files for inspection, it works just fine for us (I'll show some of that stuff at the workshop...).
Boris
Hi Boris
Sorry for the delay, i have been away travelling. I have left the US now so i wont be attending the workshop either.
I am currently getting the files together to send to you. The same problem still occurs with a much increased analysis time using the newer windows versions of OpenSees. So that occurs in windows, while linux has the error messages popping up.
Im not sure if it could be something to do with the solution methods im using, but i will send you through the files. Also, have you done a recompile of the windows version of OpenSees?
Thanks
Liam
Sorry for the delay, i have been away travelling. I have left the US now so i wont be attending the workshop either.
I am currently getting the files together to send to you. The same problem still occurs with a much increased analysis time using the newer windows versions of OpenSees. So that occurs in windows, while linux has the error messages popping up.
Im not sure if it could be something to do with the solution methods im using, but i will send you through the files. Also, have you done a recompile of the windows version of OpenSees?
Thanks
Liam
Hi Boris
Sorry again for the delay.. been trying to set up opensees on an aix machine.
I have sent you the files you require from the email address
lwot001@ec.auckland.ac.nz
Hopefully that can be of some use
Thanks
Liam
Sorry again for the delay.. been trying to set up opensees on an aix machine.
I have sent you the files you require from the email address
lwot001@ec.auckland.ac.nz
Hopefully that can be of some use
Thanks
Liam