NLP is not solving in ubuntu OS whereas It is solving in Windows OS with all versions of package and Julia is same (Juniper optimzer with Ipopt and HiGHS)

Specific Domains Optimization (Mathematical)
#1

I am trying to Solve a Non Linear model using Juniper with Ipopt and GiGHS. I am getting solution in my windows operating system, whereas Ubuntu OS is not giving any results immediately as it is saying NLP couldn’t solve optimally and No Integral solution found. I am using julia 1.6.1 in both OS and sampe packages with same version. Please find the log below for same model and help me out.

Windows OS:
Status of relaxation: LOCALLY_SOLVED
Time for relaxation: 81.11199998855591
Relaxation Obj: 7444.76604825103

   MIPobj              NLPobj       Time

=============================================
Presolving model
257 rows, 160 cols, 514 nonzeros
65 rows, 64 cols, 130 nonzeros
0 rows, 0 cols, 0 nonzeros
Presolve: Optimal

Solving report
Status Optimal
Primal bound 11.1231861754
Dual bound 11.1231861754
Gap 0% (tolerance: 0.01%)
Solution status feasible
11.1231861754 (objective)
0 (bound viol.)
0 (int. viol.)
0 (row viol.)
Timing 0.01 (total)
0.01 (presolve)
0.00 (postsolve)
Nodes 0
LP iterations 0 (total)
0 (strong br.)
0 (separation)
0 (heuristics)
11.1232 0.0 25.2

FP: 73.69899988174438 s
FP: 1 round
FP: Obj: 7445.767441685866
Obj: 7445.767441685866

Ubuntu OS:

Start values are not feasible.
Status of relaxation: LOCALLY_SOLVED
Time for relaxation: 135.4964361190796
Relaxation Obj: 7444.653740204127

   MIPobj              NLPobj       Time

=============================================
Presolving model
257 rows, 160 cols, 514 nonzeros
65 rows, 64 cols, 130 nonzeros
0 rows, 0 cols, 0 nonzeros
Presolve: Optimal

Solving report
Status Optimal
Primal bound 11.1229585924
Dual bound 11.1229585924
Gap 0% (tolerance: 0.01%)
Solution status feasible
11.1229585924 (objective)
0 (bound viol.)
0 (int. viol.)
0 (row viol.)
Timing 0.00 (total)
0.00 (presolve)
0.00 (postsolve)
Nodes 0
LP iterations 0 (total)
0 (strong br.)
0 (separation)
0 (heuristics)
┌ Warning: NLP couldn’t be solved to optimality
└ @ Juniper ~/.julia/packages/Juniper/0Z1vO/src/fpump.jl:405
11.123 - 53.5

FP: 53.602248191833496 s
FP: 1 round
FP: No integral solution found

#2

Can you provide a piece of code that allows us to reproduce your problem?

#3

Have a look at this article. It outlines all the things that help people answer help requests.

#4

To “be sure” that the versions are the same, you should create an environment specific to this project in one of the two OS, run the code and then copy the whole folder including Manifest.toml and Project.toml in the other OS, run Pgk.instantiate() and run the model again.