[Ipopt] 3.7.0 Restoration phase in the restoration phase failed.

[Andreas Waechter]

Hi Ali,

> Thank you for the response. I can keep them, no problem.

Actually, we could also attach the files to a ticket...

> Just one more question. Is it normal from IPOPT to sometimes report
> "Feasible point for square problem found" for the feasibility problem
> with a HUGE complementary value?
>
> It almost always reports "Optimal Solution Found" since a feasible
> solution MUST be an optimal solution as well.

Exactly, if Ipopt sees that there are as many free variables as equality 
constraints, it takes the luxury of quitting as soon as a (sufficiently) 
fesaible point is found, since it then declares the feasibility problem to 
be solved.  It is possible that at this point the complementarity is huge, 
but that has no real meaning.  In principle, Ipopt should not even deal 
with complementarity at this point, but the "square problem" mode is more 
like an addition to Ipopt to make it handle feasibility problems that it 
hasn't really been designed for.

Regards,

Andreas