Failed
run_tests.TestSimulation_mosek_conif.test_simulate (from run_tests.TestSimulation_mosek_conif-20200303000113)
Failing for the past 1 build
(Since Failed
)
Error Message
The solver failed for an unknown reason. Running `.debug()` may pinpoint the trouble. You can also try another solver, or increase the verbosity.
Stacktrace
Traceback (most recent call last): File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/gpkit/solvers/mosek_cli.py", line 87, in optimize solution_filename]).split(b"\n"): File "/usr/local/Cellar/python/3.7.6_1/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py", line 411, in check_output **kwargs).stdout File "/usr/local/Cellar/python/3.7.6_1/Frameworks/Python.framework/Versions/3.7/lib/python3.7/subprocess.py", line 512, in run output=stdout, stderr=stderr) subprocess.CalledProcessError: Command '['mskexpopt', '/var/folders/42/s1whb7rd4mddfcnzk96g_d9h0000gp/T/tmpypkc9_9_/gpkit_mosek', '-sol', '/var/folders/42/s1whb7rd4mddfcnzk96g_d9h0000gp/T/tmpypkc9_9_/gpkit_mosek.sol']' returned non-zero exit status 16. The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/gpkit/constraints/gp.py", line 196, in solve **solver_kwargs) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/gpkit/solvers/mosek_cli.py", line 90, in optimize raise UnknownInfeasible() from e gpkit.exceptions.UnknownInfeasible The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/testing/t_simulation.py", line 40, in test_simulate uncertainty_sets, nominal_solution, directly_uncertain_vars_subs, parallel=False) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/simulations/simulate.py", line 199, in variable_gamma_results number_of_time_average_solves, parallel) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/simulations/simulate.py", line 89, in simulate_robust_model linearizationTolerance=linearization_tolerance) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/robust.py", line 237, in robustsolve self.setup(verbosity, **options) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/robust.py", line 194, in setup new_solution = RobustModel.internalsolve(self._robust_model, verbosity=0) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/robust/robust/robust.py", line 459, in internalsolve return model.solve(verbosity=verbosity) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/gpkit/constraints/prog_factories.py", line 126, in solvefn result = progsolve(solver, verbosity=verbosity, **kwargs) File "/Users/jenkins/workspace/CE_gpkit_PR_research_models/buildnode/macys_VM/optimizer/mosek/gpkit/constraints/gp.py", line 225, in solve raise infeasibility.__class__(msg) from infeasibility gpkit.exceptions.UnknownInfeasible: The solver failed for an unknown reason. Running `.debug()` may pinpoint the trouble. You can also try another solver, or increase the verbosity.
Standard Output
SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.0076% on GP solve 4. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. Final solution let signomial constraints slacken by 0.029%. Calling .localsolve with a higher `pccp_penalty` (it was 200 this time) will reduce final slack if the model is solvable with less. If you think it might not be, check by solving with `use_pccp=False, x0=(this model's final solution)`. SGP not convergent: Cost rose by 0.066% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.024% on GP solve 6. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.12% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.13% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.18% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.12% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 12% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 2.4% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. Solution check warning: Sum of nu^T * A did not vanish. Final solution let signomial constraints slacken by 16%. Calling .localsolve with a higher `pccp_penalty` (it was 200 this time) will reduce final slack if the model is solvable with less. If you think it might not be, check by solving with `use_pccp=False, x0=(this model's final solution)`. SGP not convergent: Cost rose by 0.044% on GP solve 6. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.13% on GP solve 4. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. Final solution let signomial constraints slacken by 0.0031%. Calling .localsolve with a higher `pccp_penalty` (it was 200 this time) will reduce final slack if the model is solvable with less. If you think it might not be, check by solving with `use_pccp=False, x0=(this model's final solution)`. SGP not convergent: Cost rose by 0.089% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.027% on GP solve 4. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.008% on GP solve 4. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.22% on GP solve 3. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 0.0097% on GP solve 4. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 12% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities. SGP not convergent: Cost rose by 9% on GP solve 2. Details can be found in `m.program.results` or by solving at a higher verbosity. Note that convergence is not guaranteed for models with SignomialEqualities.