I wanted to use OptimizationProblems.jl to benchmark some solvers in parallel, but I get a strange error.
using Distributed
addprocs(2; exeflags="--project=$(@__DIR__)")
@everywhere begin
using ADNLPModels, OptimizationProblems, OptimizationProblems.ADNLPProblems
end
This gives the following error:
From worker 2: ┌ Warning: Error requiring `ADNLPModels` from `OptimizationProblems.ADNLPProblems`
From worker 2: │ exception =
From worker 2: │ LoadError: StackOverflowError:
From worker 2: │ Stacktrace:
From worker 2: │ [1] promote_typeof(::Float64, ::Float64, ::Vararg{Any}) (repeats 13902 times)
From worker 2: │ @ Base ./promotion.jl:330
From worker 2: │ [2] hvcat(::NTuple{4199, Int64}, ::Int64, ::Vararg{Number})
From worker 2: │ @ Base ./abstractarray.jl:2019
From worker 2: │ [3] top-level scope
From worker 2: │ @ ~/.julia/packages/OptimizationProblems/nfPUU/data/tetra_duct12.jl:1
From worker 2: │ [4] include(mod::Module, _path::String)
From worker 2: │ @ Base ./Base.jl:418
From worker 2: │ [5] include(x::String)
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/ADNLPProblems.jl:1
From worker 2: │ [6] top-level scope
From worker 2: │ @ ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/tetra.jl:71
From worker 2: │ [7] include(mod::Module, _path::String)
From worker 2: │ @ Base ./Base.jl:418
From worker 2: │ [8] include(x::String)
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/ADNLPProblems.jl:1
From worker 2: │ [9] top-level scope
From worker 2: │ @ ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/ADNLPProblems.jl:15
From worker 2: │ [10] eval
From worker 2: │ @ ./boot.jl:373 [inlined]
From worker 2: │ [11] eval
From worker 2: │ @ ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/ADNLPProblems.jl:1 [inlined]
From worker 2: │ [12] (::OptimizationProblems.ADNLPProblems.var"#4#8")()
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/Requires/Z8rfN/src/require.jl:101
From worker 2: │ [13] macro expansion
From worker 2: │ @ timing.jl:299 [inlined]
From worker 2: │ [14] err(f::Any, listener::Module, modname::String, file::String, line::Any)
From worker 2: │ @ Requires ~/.julia/packages/Requires/Z8rfN/src/require.jl:47
From worker 2: │ [15] (::OptimizationProblems.ADNLPProblems.var"#3#7")()
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/Requires/Z8rfN/src/require.jl:100
From worker 2: │ [16] withpath(f::Any, path::String)
From worker 2: │ @ Requires ~/.julia/packages/Requires/Z8rfN/src/require.jl:37
From worker 2: │ [17] (::OptimizationProblems.ADNLPProblems.var"#2#6")()
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/Requires/Z8rfN/src/require.jl:99
From worker 2: │ [18] listenpkg(f::Any, pkg::Base.PkgId)
From worker 2: │ @ Requires ~/.julia/packages/Requires/Z8rfN/src/require.jl:20
From worker 2: │ [19] macro expansion
From worker 2: │ @ ~/.julia/packages/Requires/Z8rfN/src/require.jl:98 [inlined]
From worker 2: │ [20] macro expansion
From worker 2: │ @ ~/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/ADNLPProblems.jl:8 [inlined]
From worker 2: │ [21] (::OptimizationProblems.ADNLPProblems.var"#1#5")()
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/Requires/Z8rfN/src/init.jl:11
From worker 2: │ [22] __init__()
From worker 2: │ @ OptimizationProblems.ADNLPProblems ~/.julia/packages/Requires/Z8rfN/src/init.jl:18
From worker 2: │ [23] _include_from_serialized(path::String, depmods::Vector{Any})
From worker 2: │ @ Base ./loading.jl:768
From worker 2: │ [24] _require_search_from_serialized(pkg::Base.PkgId, sourcepath::String)
From worker 2: │ @ Base ./loading.jl:854
From worker 2: │ [25] _require(pkg::Base.PkgId)
From worker 2: │ @ Base ./loading.jl:1097
From worker 2: │ [26] require(uuidkey::Base.PkgId)
From worker 2: │ @ Base ./loading.jl:1013
From worker 2: │ [27] (::Distributed.var"#1#2"{Base.PkgId})()
From worker 2: │ @ Distributed /apps/local-fci/tools/julia-1.7.1/share/julia/stdlib/v1.7/Distributed/src/Distributed.jl:79
From worker 2: │ [28] (::Distributed.var"#103#104"{Distributed.CallMsg{:call}})()
From worker 2: │ @ Distributed /apps/local-fci/tools/julia-1.7.1/share/julia/stdlib/v1.7/Distributed/src/process_messages.jl:274
From worker 2: │ [29] run_work_thunk(thunk::Distributed.var"#103#104"{Distributed.CallMsg{:call}}, print_error::Bool)
From worker 2: │ @ Distributed /apps/local-fci/tools/julia-1.7.1/share/julia/stdlib/v1.7/Distributed/src/process_messages.jl:63
From worker 2: │ [30] run_work_thunk(rv::Distributed.RemoteValue, thunk::Function)
From worker 2: │ @ Distributed /apps/local-fci/tools/julia-1.7.1/share/julia/stdlib/v1.7/Distributed/src/process_messages.jl:72
From worker 2: │ [31] (::Distributed.var"#96#98"{Distributed.RemoteValue, Distributed.var"#103#104"{Distributed.CallMsg{:call}}})()
From worker 2: │ @ Distributed ./task.jl:423
From worker 2: │ in expression starting at /home/motoug/.julia/packages/OptimizationProblems/nfPUU/data/tetra_duct12.jl:1
From worker 2: │ in expression starting at /home/motoug/.julia/packages/OptimizationProblems/nfPUU/src/ADNLPProblems/tetra.jl:71
From worker 2: └ @ Requires ~/.julia/packages/Requires/Z8rfN/src/require.jl:51
You can reproduce the error without having multiple processes by simply doing this:
using Distributed
@everywhere begin
include("tetra_duct12.jl")
end
If you simply do include("tetra_duct12.jl") without the @everywhere macro, there is no error. There is an open issue on the same error message: https://github.com/JuliaLang/julia/issues/45454
I wanted to use
OptimizationProblems.jlto benchmark some solvers in parallel, but I get a strange error.This gives the following error:
You can reproduce the error without having multiple processes by simply doing this:
If you simply do
include("tetra_duct12.jl")without the@everywheremacro, there is no error. There is an open issue on the same error message: https://github.com/JuliaLang/julia/issues/45454