search

jump-dev / Gurobi.jl

A Julia interface to the Gurobi Optimizer
http://www.gurobi.com/
MIT License
222 stars 81 forks source link

Segfault on ctrl-c #52

Closed JackDunnNZ closed 4 years ago

JackDunnNZ commented 8 years ago

Is it possible to give Gurobi the same magic as JuliaOpt/CPLEX.jl#45 so that it doesn't segfault on ctrl-c? I'm happy to have a go if needed, but it might be easier for someone that understands the internals better than I do.

Related, I've been working with a model given to me that uses gurobipy, and on that, ctrl-c interrupts the solve and returns the current incumbent. I'm not sure if that would fall out here as a consequence, but it would be pretty cool if it worked!

mlubin commented 8 years ago

No clue here, @joehuchette?

joehuchette commented 8 years ago

If Gurobi has an analogue to cpxsetterminate, then yeah the approach I took with CPLEX should work here as well. grbterminate looks like it might do the job.

(Related: https://groups.google.com/forum/#!topic/gurobi/n0CNeAe91VU)

JackDunnNZ commented 8 years ago

Cool thanks, I'll have a crack at it.

joehuchette commented 8 years ago

I'm guessing it'll be even easier than what I had to do in CPLEX, since you can just call grbterminate directly in @grb_ccall_intercept instead of carrying around extra state in the model object...

JackDunnNZ commented 8 years ago

I copied in the same macro and changed it to use grbterminate, but it's never catching the exception, it just dies. Any ideas? It catches the exception if I just set up an infinite loop before the solve and ctrl-c out of that, but as soon as it enters grboptimize it doesn't seem to catch it.

Change here, not that it's more than a copy/paste... https://github.com/JuliaOpt/Gurobi.jl/commit/16c3f714a7bf5547ead0415c98aeacaf679b151a

JackDunnNZ commented 8 years ago

@mlubin You were right, what I have works fine as is on Ubuntu so it must be an OS X problem