I’m using Mosek in cvx for my gp optimization model. I received this Error:

*** Error(1400): blx[66] is too large Reference to non-existent field 'sol'.

There is not enough info about this error, I do not have sol variable neither blx. I have an iterative loop and in each iteration I generate a matrix of random numbers. This error happens irregularly and when it does not happen I receive a "solved" message from mosek. Worth to mention that, I tried sedumi instead of mosek and the result is "infeasible" for all the iterations.

I cannot understand how to cure this error. Any clue to solve this error is appreciated.

Thanks for your reply,
I only have access to mosek, sedumi and sdpt3 in cvx. and because I’m using gp it is not easy to migrate my complete model to another solver (e.g. Matlab opt. tool or cplex etc…)
I give you more details maybe you can give me a help on it:
I tried sedumi and sdpt3 for my gp model and I receive “failed” status with optimal value equal NaN.
I tried mosek for my gp model and it irregularly gives me the error I have mentioned in my first post. I tried different precision (best, medium, low) and it does not help.
I put the complete Error message here: *** Error(1400): blx[65] is too large Reference to non-existent field 'sol'. Error in cvx_mosek>solve (line 486) Error in cvxprob/solve (line 247) [ x, status, tprec, iters2, y, z ] = shim.solve( [ At, Anew2 ], [ b ; bnew ], c, cones, true, prec, solv.settings, eargs{:} ); Error in cvx_end (line 88) solve( prob );

the Error (1400) is generated on rethrow( estruc );line of the following code which is a part of cvx_end.m :

`if ~isempty( estruc ),
if strncmp( estruc.identifier, 'CVX:', 4 ),
error( estruc.identifier, estruc.message );
else
rethrow( estruc );
end
end`

hi, dear saralk, have you solved your problem? now i have the same issue, if you have the way to handle it, can you tell me please? thank you very much