Page 1 of 1

`GLIBCXX_3.4.11' not found

PostPosted: Tue Apr 10, 2012 3:33 pm
by Will94
Hello,

I provide technical support for the Economics Department at a large public university. I had a faculty member request Dynare, which I installed for him yesterday. When he tries to use the software, he receives the following feedback:

Code: Select all
>> dynare structuralNews4April2012_Est
 
Configuring Dynare ...
[mex] Generalized QZ.
[mex] Sylvester equation solution.
[mex] Kronecker products.
[mex] Sparse kronecker products.
[mex] Bytecode evaluation.
[mex] k-order perturbation solver.
[mex] k-order solution simulation.

/installers/dynare-4.2.5/matlab/dynare_m: /usr/bin/matlab/sys/os/glnxa64/libstdc++.so.6: version `GLIBCXX_3.4.11' not found (required by /installers/dynare-4.2.5/matlab/dynare_m)

??? Error using ==> dynare at 114
DYNARE: preprocessing failed

I've done a little looking with google, and apparently MATLAB ships with its own libraries which can conflict with the system's. I am hesitant to remove or rename the MATLAB libraries as this is a shared machine which is used by numerous people in our department. The system runs Mandriva Linux 2010.2 x64.

Any advice on how to solve this problem would be greatly appreciated.


Thank you!

Will N.,

Bryan, TX USA

Re: `GLIBCXX_3.4.11' not found

PostPosted: Tue Apr 10, 2012 8:17 pm
by Will94
I upgraded the libraries under the MATLAB folder which fixed the 'GLIBCXX_3.4.11 not found' problem.

However, my end user has a new problem. I will copy and paste his latest email to me below as it contains the best description of the problem that I can provide. If someone can describe the source of his problems and point me to a document/FAQ/etc with a solution(s), I would greatly appreciate it.

FYI, it would be advantageous for him to be able to run his jobs on one of our Linux machines as they have 32 GB of RAM each... far more than his Windows desktop PC.

Code: Select all
I uploaded my working folder again.  Here is the error message.


You did not declare endogenous variables after the estimation command.
??? Error using ==> read_variables at 84
Can't find datafile: structuralNewsDataBookForDynare

Error in ==> dynare_estimation_init at 290
rawdata =
read_variables(options_.datafile,options_.varobs,[],options_.xls_sheet,options_.xls_range);

Error in ==> dynare_estimation_1 at 37
  [data,rawdata,xparam1] = dynare_estimation_init(var_list_);

Error in ==> dynare_estimation at 62
    dynare_estimation_1(var_list,varargin{:});

Error in ==> structuralNews4April2012_Est at 168
dynare_estimation(var_list_);

Error in ==> dynare at 120
evalin('base',fname) ;



When I run the same program with my Windows workstation, the following line comes, and Dynare launch the estimation procedure.

You did not declare endogenous variables after the estimation command.
Loading 194 observations from structuralNewsDataBookForDynare.xls



Thank you!

Will N.,
Bryan, TX USA

Re: `GLIBCXX_3.4.11' not found

PostPosted: Tue Apr 10, 2012 9:48 pm
by bkjecn
It seems that Dynare is having trouble reading the data into Matlab. When working in linux, it is probably not a good idea to try to read in an Excel file... almost always breaks. You can avoid Excel by instead saving the data used for estimation in a .mat file. The variables need to be saved in a single .mat file as column vectors with names that match the names in your .mod file. You may have to play with it a bit to get it working... but it shouldn't be too big of a problem.