"ERROR: LoadError: TaskFailedException: IOError: stream is closed or unusable" with @distributed on multinode cluster

I have some code that evaluates trajectories of a stochastic master equation in parallel, using a @distributed for with reduction for evaluating average quantities. The code is available here.

Some jobs fail with the following error, which I can’t understand. Can anyone provide some insight or possible ideas on how to diagnose it?

ERROR: LoadError: TaskFailedException:
IOError: stream is closed or unusable
Stacktrace:
 [1] uv_write_async(::Sockets.TCPSocket, ::Ptr{UInt8}, ::UInt64) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [2] uv_write(::Sockets.TCPSocket, ::Ptr{UInt8}, ::UInt64) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [3] unsafe_write(::Sockets.TCPSocket, ::Ptr{UInt8}, ::UInt64) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [4] unsafe_write at ./io.jl:593 [inlined]
 [5] macro expansion at ./gcutils.jl:91 [inlined]
 [6] write at ./io.jl:616 [inlined]
 [7] serialize_array_data at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:246 [inlined]
 [8] serialize(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Array{Complex{Float64},2}) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:263
 [9] serialize(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Array{Array{Complex{Float64},2},1}) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:268
 [10] serialize_any(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Any) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:629
 [11] serialize(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Any) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:608
 [12] serialize_any(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Any) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:629
 [13] serialize(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Any) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:608
 [14] serialize_any(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Any) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:629
 [15] serialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:608 [inlined]
 [16] serialize_msg(::Distributed.ClusterSerializer{Sockets.TCPSocket}, ::Distributed.CallMsg{:call_fetch}) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:90
 [17] #invokelatest#1 at ./essentials.jl:709 [inlined]
 [18] invokelatest at ./essentials.jl:708 [inlined]
 [19] send_msg_(::Distributed.Worker, ::Distributed.MsgHeader, ::Distributed.CallMsg{:call_fetch}, ::Bool) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:185
 [20] send_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:134 [inlined]
 [21] #remotecall_fetch#145(::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}, ::typeof(remotecall_fetch), ::Function, ::Distributed.Worker, ::Function, ::Vararg{Any,N} where N) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/remotecall.jl:385
 [22] remotecall_fetch(::Function, ::Distributed.Worker, ::Function, ::Vararg{Any,N} where N) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/remotecall.jl:382
 [23] #remotecall_fetch#148 at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/remotecall.jl:417 [inlined]
 [24] remotecall_fetch at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/remotecall.jl:417 [inlined]
 [25] (::Distributed.var"#159#160"{typeof(+),ContinuousMeasurementFI.var"#15#19"{Float64,Float64,TimerOutputs.TimerOutput,Int64,Float64,ContinuousMeasurementFI.var"#dW#16"{Float64},RemoteChannel{Channel{Bool}}},UnitRange{Int64},Array{UnitRange{Int64},1},Int64,Int64})() at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/macros.jl:270
Stacktrace:
 [1] wait at ./task.jl:251 [inlined]
 [2] fetch at ./task.jl:266 [inlined]
 [3] iterate at ./generator.jl:47 [inlined]
 [4] collect_to!(::Array{Array{Float64,2},1}, ::Base.Generator{Array{Task,1},typeof(fetch)}, ::Int64, ::Int64) at ./array.jl:667
 [5] collect_to_with_first!(::Array{Array{Float64,2},1}, ::Array{Float64,2}, ::Base.Generator{Array{Task,1},typeof(fetch)}, ::Int64) at ./array.jl:646
 [6] collect(::Base.Generator{Array{Task,1},typeof(fetch)}) at ./array.jl:627
 [7] preduce(::Function, ::Function, ::UnitRange{Int64}) at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/macros.jl:274
 [8] macro expansion at /home/matros/.julia/packages/ProgressMeter/g1lse/src/ProgressMeter.jl:595 [inlined]
 [9] macro expansion at /home/matros/.julia/packages/ContinuousMeasurementFI/tmNbJ/src/Eff_QFI_HD_Dicke.jl:135 [inlined]
 [10] macro expansion at /home/matros/.julia/packages/TimerOutputs/7Id5J/src/TimerOutput.jl:214 [inlined]
 [11] #Eff_QFI_HD_Dicke#12(::Float64, ::Float64, ::Float64, ::Float64, ::Int64, ::TimerOutputs.TimerOutput, ::typeof(Eff_QFI_HD_Dicke), ::Int64, ::Int64, ::Float64, ::Float64) at /home/matros/.julia/packages/ContinuousMeasurementFI/tmNbJ/src/Eff_QFI_HD_Dicke.jl:134
 [12] (::ContinuousMeasurementFI.var"#kw##Eff_QFI_HD_Dicke")(::NamedTuple{(:κ, :κcoll, :ω, :η, :outpoints),Tuple{Float64,Float64,Float64,Float64,Int64}}, ::typeof(Eff_QFI_HD_Dicke), ::Int64, ::Int64, ::Float64, ::Float64) at ./none:0
 [13] top-level scope at /home/matros/continuous_measurement_dicke/sims_20200217/simulateTama.jl:28
 [14] include_relative(::Module, ::String) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [15] include(::Module, ::String) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [16] exec_options(::Base.JLOptions) at /home/matros/julia-1.3.0/lib/julia/sys.so:?
 [17] _start() at /home/matros/julia-1.3.0/lib/julia/sys.so:?
in expression starting at /home/matros/continuous_measurement_dicke/sims_20200217/simulateTama.jl:28
ARGS = ["50", "5000", "1.0", "0.00005", "1.0", "1.0", "0.01", "1.0", "200"]
ContinuousMeasurementFI.SuperOperator{Complex{Float64},Int64}
ContinuousMeasurementFI.SuperOperator{Complex{Float64},Int64}
      From worker 73:
      From worker 73:   signal (11): Segmentation fault
      From worker 73:   in expression starting at none:0
      From worker 73:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 73:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 73:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:
      From worker 76:   signal (11): Segmentation fault
      From worker 76:   in expression starting at none:0
      From worker 73:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 76:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 76:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 73:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 73:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 73:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 76:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   process_tcp_streams at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 73:   #101 at ./task.jl:333
      From worker 73:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 73:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 73:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 73:   start_task at /buildworker/worker/package_linux64/build/src/task.c:659
      From worker 73:   unknown function (ip: 0xffffffffffffffff)
      From worker 73:   Allocations: 16906524 (Pool: 16903554; Big: 2970); GC: 18
      From worker 76:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 76:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 76:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 76:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 76:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   process_tcp_streams at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 76:   #101 at ./task.jl:333
      From worker 76:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 76:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 76:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 76:   start_task at /buildworker/worker/package_linux64/build/src/task.c:659
      From worker 76:   unknown function (ip: 0xffffffffffffffff)
      From worker 76:   Allocations: 16906533 (Pool: 16903560; Big: 2973); GC: 17
      From worker 79:
      From worker 79:   signal (11): Segmentation fault
      From worker 79:   in expression starting at none:0
      From worker 81:
      From worker 81:   signal (11): Segmentation fault
      From worker 81:   in expression starting at none:0
      From worker 57:
      From worker 57:   signal (11): Segmentation fault
      From worker 57:   in expression starting at none:0
      From worker 57:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 57:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 81:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 81:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 81:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 79:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 79:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 79:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 79:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 78:
      From worker 78:   signal (11): Segmentation fault
      From worker 78:   in expression starting at none:0
      From worker 57:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 78:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 78:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 78:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 81:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 67:
      From worker 67:   signal (11): Segmentation fault
      From worker 67:   in expression starting at none:0
      From worker 79:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 78:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 67:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 67:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 67:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 67:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 67:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 67:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 81:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 81:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 67:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 55:
      From worker 55:   signal (11): Segmentation fault
      From worker 55:   in expression starting at none:0
      From worker 81:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 78:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 67:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 55:   sig_match_fast at /buildworker/worker/package_linux64/build/src/gf.c:2160 [inlined]
      From worker 55:   jl_lookup_generic_ at /buildworker/worker/package_linux64/build/src/gf.c:2243 [inlined]
      From worker 55:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2301
      From worker 55:   deserialize_fillarray! at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:1094
      From worker 55:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 55:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 79:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 57:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 67:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 67:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 67:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 55:   deserialize_array at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 57:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 57:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 79:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 78:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 81:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 81:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 55:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 67:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 81:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 81:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 79:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 79:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 79:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 79:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 78:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   deserialize at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Serialization/src/Serialization.jl:722 [inlined]
      From worker 78:   deserialize_msg at /home/matros/julia-1.3.0/share/julia/stdlib/v1.3/Distributed/src/messages.jl:99
      From worker 78:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 78:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 55:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 55:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 55:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   process_tcp_streams at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 67:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 57:   #101 at ./task.jl:333
      From worker 79:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 79:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 67:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 67:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 57:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 57:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 79:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 78:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 78:   jl_f__apply at /buildworker/worker/package_linux64/build/src/builtins.c:627
      From worker 78:   jl_f__apply_latest at /buildworker/worker/package_linux64/build/src/builtins.c:665
      From worker 57:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 57:   start_task at /buildworker/worker/package_linux64/build/src/task.c:659
      From worker 57:   unknown function (ip: 0xffffffffffffffff)
      From worker 57:   Allocations: 16904401 (Pool: 16901426; Big: 2975); GC: 18
      From worker 81:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 55:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 67:   handle_deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 78:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 79:   message_handler_loop at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   process_tcp_streams at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 81:   #101 at ./task.jl:333
      From worker 81:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2141 [inlined]
      From worker 81:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 55:   deserialize at /home/matros/julia-1.3.0/lib/julia/sys.so (unknown line)
      From worker 55:   _jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2135 [inlined]
      From worker 55:   jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2305
      From worker 81:   jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1631 [inlined]
      From worker 81:   start_task at /buildworker/worker/package_linux64/build/src/task.c:659
      From worker 81:   unknown function (ip: 0xffffffffffffffff)
      From worker 81:   Allocations: 16906535 (Pool: 16903564; Big: 2971); GC: 18

I’m running these simulations on a multinode cluster. Julia is started by passing the node list with the --machine-file parameters.

1 Like

Did this ever get resolved? I’ve got a similar issue occurring with a Julia package I wrote.

Nope. I just modified the code to minimize the communication between workers, and I don’t get this error anymore, although I don’t know if the two things are related. I was passing big matrices around, now they are initialized independently by each worker.

Ah, yep I’m passing big matrices around as well. I’ve switched to multi-threading in the latest unreleased version, so maybe that will help? We’ll see. Thanks for the quick reply!