Error when doing modal analysis using self-complied OpenSees

[danielree]

Hi there,

I downloaded the latest version from svn (svn://peera.berkeley.edu/usr/local/svn/OpenSees/trunk) and complied it using VS2017 for the Win64 version. I used the generated OpenSees.exe to do a modal analysis -- Eigen analysis of a two-story shear frame (http://opensees.berkeley.edu/wiki/index ... hear_frame). It shows an error as shown below:

OpenSees > source EigenAnal_twoStoreyShearFrame7.tcl
forrtl: severe (408): fort: (19): Dummy character variable 'WHICH' has length 2 which is greater than actual variable length -3689348814741910324

I tried other example like static analysis -- Truss (http://opensees.berkeley.edu/wiki/index ... ss_Example). It works fine.

Could you give me some suggestion on how to solve the problem?

Thanks,
Dan

[fmk]

does the binary version you can download give the same error as the compiled version?

[danielree]

fmk wrote:
> does the binary version you can download give the same error as the
> compiled version?

Hi fmk,

The binary version works fine for the modal analysis.

Although both versions (binary and complied) are Version 2.5.0, the binary is Version 2.5.0 (rev 6536) but the complied version is Version 2.5.0 (rev 6248). Would this cause the problem?

Thanks,
Dan

[fmk]

the important number is that revision number .. we slip bug fixes into the executable w/o doing official releases (due to amount of effort to go through steps of creating an official release). can you download the same revision number as the binary version.

a warning. they are going to shut down th peers server imminently and it will be offline till the machine is patched with latest os (and given my schedule that could be awhile)

[danielree]

fmk wrote:
> the important number is that revision number .. we slip bug fixes into the
> executable w/o doing official releases (due to amount of effort to go
> through steps of creating an official release). can you download the same
> revision number as the binary version.
>
> a warning. they are going to shut down th peers server imminently and it
> will be offline till the machine is patched with latest os (and given my
> schedule that could be awhile)

Hi fmk,

I downloaded the revision 6536. I tried to compile it. But VS2017 shows a lot of errors like:
"C:\Program Files (x86)\Microsoft Visual Studio\2017\Enterprise\Common7\IDE\VC\VCTargets\Microsoft.Cpp.Platform.targets(55,5): error MSB8020: The build tools for Visual Studio 2010 (Platform Toolset = 'v100') cannot be found. To build using the v100 build tools, please install Visual Studio 2010 build tools. Alternatively, you may upgrade to the current Visual Studio tools by selecting the Project menu or right-click the solution, and then selecting "Retarget solution".
6>Done building project "damage.vcxproj" -- FAILED."

I changed to VS2010, but it shows the error when building the project OpenSees:
"1>------ Build started: Project: OpenSees, Configuration: Debug x64 ------
1>Build started 12/4/2017 2:51:55 PM.
1>InitializeBuildStatus:
1> Touching ".\..\..\obj\opensees\debug\OpenSees.unsuccessfulbuild".
1>ClCompile:
1> All outputs are up-to-date.
1>ManifestResourceCompile:
1> All outputs are up-to-date.
1>LINK : warning LNK4075: ignoring '/INCREMENTAL' due to '/FORCE' specification
1> Creating library .\..\..\bin\OpenSees.lib and object .\..\..\bin\OpenSees.exp
1>material.lib(TclModelBuilderUniaxialMaterialCommand.obj) : error LNK2019: unresolved external symbol "void * __cdecl OPS_PySimple3(void)" (?OPS_PySimple3@@YAPEAXXZ) referenced in function "int __cdecl TclModelBuilderUniaxialMaterialCommand(void *,struct Tcl_Interp *,int,char const * *,class Domain *)" (?TclModelBuilderUniaxialMaterialCommand@@YAHPEAXPEAUTcl_Interp@@HPEAPEBDPEAVDomain@@@Z)
1>.\..\..\bin\OpenSees.exe : fatal error LNK1120: 1 unresolved externals
1>
1>Build FAILED.
1>
1>Time Elapsed 00:00:02.48
========== Build: 0 succeeded, 1 failed, 24 up-to-date, 0 skipped =========="

Could you give me some suggestions on how to solve these errors?

Thanks,
Dan

[danielree]

fmk wrote:
> the important number is that revision number .. we slip bug fixes into the
> executable w/o doing official releases (due to amount of effort to go
> through steps of creating an official release). can you download the same
> revision number as the binary version.
>
> a warning. they are going to shut down th peers server imminently and it
> will be offline till the machine is patched with latest os (and given my
> schedule that could be awhile)
Hi fmk,

I also noticed that the revision number in SVN seems conflicting to the revision number of the software. I downloaded the revision 6636 (SVN) but it shows revision 6248 (in software).

So, I am not sure that revision 6536 is the one that I desired or not, as I have not succeeded in building it.

Dan

[danielree]

fmk wrote:
> the important number is that revision number .. we slip bug fixes into the
> executable w/o doing official releases (due to amount of effort to go
> through steps of creating an official release). can you download the same
> revision number as the binary version.
>
> a warning. they are going to shut down th peers server imminently and it
> will be offline till the machine is patched with latest os (and given my
> schedule that could be awhile)

Hi fmk,

I think I may find the reason that causes this problem. The eigenvalue solver in OpenSees will call ARPACK to do the calculation. While the included ARPACK is written in Fortran, passing the array 'which' in function 'DSAUPD' doesn't indicate the length of the array. And this makes Fortran read a random number. Could you give some suggestion on how to solve this C-Fortran interface probelm?

Thanks,
Dan

[YiXiao]

Hi Dan,

I have the same error when using "eigenvalue" command of the self-complied OpenSees. Could you please share the solution to fix this problem? Thank you!

Best regards,
Yi
University of Washington/Tongji University